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
|
|---|---|---|---|---|---|
Hypothalamic neurons are main regulators of energy homeostasis . Neuronal function essentially depends on plasma membrane-located gangliosides . The present work demonstrates that hypothalamic integration of metabolic signals requires neuronal expression of glucosylceramide synthase ( GCS; UDP-glucose:ceramide glucosyltransferase ) . As a major mechanism of central nervous system ( CNS ) metabolic control , we demonstrate that GCS-derived gangliosides interacting with leptin receptors ( ObR ) in the neuronal membrane modulate leptin-stimulated formation of signaling metabolites in hypothalamic neurons . Furthermore , ganglioside-depleted hypothalamic neurons fail to adapt their activity ( c-Fos ) in response to alterations in peripheral energy signals . Consequently , mice with inducible forebrain neuron-specific deletion of the UDP-glucose:ceramide glucosyltransferase gene ( Ugcg ) display obesity , hypothermia , and lower sympathetic activity . Recombinant adeno-associated virus ( rAAV ) -mediated Ugcg delivery to the arcuate nucleus ( Arc ) significantly ameliorated obesity , specifying gangliosides as seminal components for hypothalamic regulation of body energy homeostasis .
The investigation of pathogenetic mechanisms underlying obesity has attained significant interest , as obesity has become an endemic metabolic disturbance worldwide . Elevated peripheral energy storage can develop as a consequence of alterations in the neuronal feedback circuits regulating energy homeostasis . The hypothalamus is the main CNS integrator of peripheral energy signals , matching energy intake to energy expenditure for body weight maintenance [1] . Among the most extensively studied peripheral molecules involved in regulating energy homeostasis and feeding behavior in the CNS are the adipocyte-derived hormone leptin as well as insulin [2] , [3] . Among numerous leptin- and insulin-sensitive brain areas , the hypothalamic Arc is one of the main regions integrating peripheral energy signals and initiating adaptive metabolic and behavioral responses [4] . Recently , several CNS regions targeted by leptin have emerged that are involved in the regulation of energy metabolism , such as the brain stem nucleus of the solitary tract ( NTS ) and reward circuits involving the ventral tegmental area [5] , [6] . Still , leptin is suggested to exert anti-obesity effects by signaling through “long form” leptin receptors ( ObR ) abundantly present on both orexigenic neuropeptide Y ( NPY ) /agouti-related peptide ( AgRP ) neurons and anorexigenic pro-opiomelanocortin ( POMC ) neurons in the Arc . Excess NPY signaling abates sympathetically mediated thermogenesis , thereby reducing energy expenditure [7] . NPY and AgRP expression is attenuated upon ObR-induced phosphatidylinositol-3-OH-kinase ( PI3k ) signaling [8] . Conversely , leptin stimulates the expression of the POMC-derived neurotransmitter α-melanocyte-stimulating hormone ( α-MSH ) through the Janus kinase/signal transducer and activator of transcription ( Jak-Stat ) pathway [9] . Alpha-MSH , a potent agonist of melanocortin receptors , inhibits food intake and stimulates the expenditure of excess energy in the body , thus preventing obesity development [10] . Insulin exerts its anorexigenic effects in hypothalamic neurons by directly stimulating insulin receptor autophosphorylation and activation of PI3k . Even though both insulin and leptin receptor stimulation leads to activation of PI3k and subsequent formation of phosphatidylinositol ( 3 , 4 , 5 ) -triphosphate ( PIP3 ) [11] , it has been shown that both hormones exert converging direct actions on POMC neurons , while having opposite effects on AgRP/NPY neurons [12] . GCS is the key enzyme for the biosynthesis of glycosphingolipids ( GSLs ) and gangliosides , a class of acidic GSLs abundantly expressed by neurons and glial cells [13] , [14] . Ganglioside-depleted neurons are viable and show apoptosis rates comparable to wild-type neurons [15] . GSLs including gangliosides contribute to the formation of membrane microdomains , which are important mediators of intracellular signal transduction [16] . GCS expression is crucial for initial postnatal brain maturation and Ugcgf/f//NesCre mice with constitutive Ugcg deletion in brain tissue under the control of the nestin promoter die within 3 wk after birth [15] . In 2003 , it was shown that GM3 synthase-deficient mice are more sensitive to insulin , thereby protecting these mice from high-fat-diet-induced insulin resistance [17] . A different ganglioside species , GD1a , has been shown to exert activating effects on tyrosine kinase receptors [18] . To address the functional role of GCS in neuronal regulation of energy homeostasis , we have generated and characterized mice with inducible neuron-specific Ugcg deletion in adult mouse CNS ( Ugcgf/f//CamKCreERT2 mice ) . Cre activity in this mouse model was restricted to distinct populations of forebrain neurons . Hypothalamic nuclei involved in the regulation of energy homeostasis were targeted by this approach . Explicitly , Cre activity was absent in the brain stem NTS , which also contributes to regulation of energy homeostasis . The present study highlights GCS-derived gangliosides as mediators for ObR-dependent signal transduction at the hypothalamic neuronal membrane . GCS-depleted neurons failed to show ObR activation upon leptin stimulation . Major neuronal gangliosides GM1 and GD1a were recruited to ObR upon ligand stimulation and subsequent signal transduction depended on ganglioside expression in hypothalamic neurons . Ugcgf/f//CamKCreERT2 mice deficient in GSLs in hypothalamus developed progressive obesity and decreased sympathetically mediated thermogenesis . rAAV-mediated Ugcg delivery to the hypothalamic Arc with ensuing nucleus-specific GSL synthesis significantly ameliorated obesity .
Ugcgflox/flox ( Ugcgf/f ) mice were bred with mice expressing the inducible CreERT2 recombinase under the control of the Calcium/Calmodulin-dependent Kinase II-alpha ( CamK ) promoter , resulting in forebrain neuron-specific Ugcg deletion ( Ugcgf/f//CamKCreERT2 ) followed by ganglioside depletion after tamoxifen injection ( Figure 1A ) . Generation of Ugcgf/f mice and CamKCreERT2 mice has been described earlier [15] , [19] . Beta-galactosidase ( X-Gal ) staining of brains from R26R/Ugcgf/+//CamKCreERT2 reporter mice indicated strong Cre activity in distinct hypothalamic nuclei , namely in the Arc ( Figures 1B and S1B ) , in the paraventricular nucleus , and in median preoptic area ( MnPO ) ( Figure S1A , B ) . Additional Cre activity was detected in the lateral hypothalamic area ( LHA ) , in hippocampus , and in the cerebral cortex ( Figure S1A , B ) . Notably , Cre activity was absent in the ventromedial hypothalamus and the NTS in the brain stem ( Figure S1A , B ) . Ganglioside depletion was confirmed in Cre-targeted areas by GD1a immunofluorescence , whereas non-targeted areas retained GD1a expression ( Figure 1B and Figure S1A ) . Consistent with the expected Cre-activity pattern , in situ hybridization showed Ugcg mRNA depletion in hippocampus , cerebral cortex , amygdala , as well as hypothalamic nuclei ( Figure S1C ) . Recombination events were absent in peripheral organs and peripheral nervous tissue ( Figure S1D ) . Neuron-dense total hippocampi showed significant and stable ganglioside reduction 3 wk postinduction ( p . i . ) , as assessed by thin layer chromatography ( TLC ) ( Figure S1E ) . Residual gangliosides in the dissected tissue were assumed to result from glial cells as well as from innervating nerve fibers emerging from nontargeted neurons [14] . Ceramide levels in Cre-targeted neuronal populations were unchanged ( Figure 1C ) , and a slight increase in sphingomyelin could be detected ( Figure S1F ) . In order to investigate if ganglioside depletion abated general neuronal function and integrity in Ugcgf/f//CamKCreERT2 mice , both electron microscopy and electrophysiological slice recordings were done at late time points p . i . Electron microscopy from Arc neurons displayed normal ultrastructure of the neuronal nucleus , organelles , and an intact , regular plasma membrane of Ugcgf/f//CamKCreERT2 mice both 6 and 12 wk p . i . ( Figure 1D ) . Basic biophysical parameters [spontaneous firing rate , action potential ( AP ) width , and AP rate of rise] from slice recordings of Arc neurons 12 wk p . i . were unaltered ( Figure S2A ) . The resting membrane potential and the AP threshold were marginally increased in Ugcgf/f//CamKCreERT2 mice , however not to an extent that impairs neuronal function ( Figure S2B ) . In order to confirm these findings in vitro , immortalized mouse hypothalamic cells ( N-41 cells ) expressing GCS-derived gangliosides ( Figure S3A , B ) were treated with n-butyldeoxynojirimycin ( NB-DNJ ) specifically inhibiting GCS [20] . NB-DNJ treatment resulted in approximately 80%–90% ganglioside depletion ( Figure S3C ) . Consistent with the findings in Ugcgf/f//CamKCreERT2 mice , membrane integrity and normal cellular ultrastructure of ganglioside-depleted N-41 cells was confirmed by electron microscopy ( Figure S3D ) . Additionally , passive and active membrane properties of cultured primary GCS-deficient hypothalamic Ugcgf/f//NesCre neurons [15] were examined by whole-cell recordings . There were no differences toward control cells in membrane resistance , capacitance , and resting potential ( Figure S3E ) . Spikes evoked by somatic current injection had unaltered threshold , amplitude , and duration ( Figure S3F ) . These results indicate that basic neuronal integrity and general function are not affected by Ugcg deletion and subsequent lack of plasma membrane gangliosides . Coinciding with neuronal ganglioside depletion 3 wk p . i . female and male Ugcgf/f//CamKCreERT2 mice displayed progressive body weight increase ( Figure 2A , B ) . This phenotype was not detected in heterozygous mice ( Figure S4A ) , as residual GCS activity accounted for maintenance of neuronal ganglioside biosynthesis [15] . Ugcgf/f//CamKCreERT2 mice were larger than control littermates 16 wk p . i . ( Figure 2C ) . Hematoxylin and eosin ( HE ) staining revealed enlarged adipocytes in epigonadal white adipose tissue ( WAT ) ( Figure 2D ) . In line with this , epigonadal WAT pad weight was significantly elevated ( Figure 2E ) . Whole body nuclear magnetic resonance ( NMR ) analysis revealed that body weight increase was due to progressive accumulation of body fat ( Figure 2F ) ; lean mass was only marginally elevated 4 wk p . i . ( Figure S4B ) . Adjusted for body weight the initial increase of fat and lean mass was proportional , whereas at later stages fat mass overrode lean mass gain ( Figure S4C ) . Liver steatosis and morphological changes in major peripheral organs of obese Ugcgf/f//CamKCreERT2 mice were not detected 9 wk p . i . ( Figure S4D ) . Serum enzyme activities indicative for liver function ( glutamate dehydrogenase , glutamic oxaloacetic transaminase , and glutamic pyruvic transaminase ) were unaltered ( Figure S4E ) . Likewise , serum cholesterol , urea , glucose , and creatinine did not show any biologically relevant abnormalities ( Figure S4F ) . Coincident with obesity , Ugcgf/f//CamKCreERT2 mice were less glucose tolerant than Ugcgf/f mice 12 wk p . i . ( Figure S4G ) and insulin sensitivity was marginally impaired 10 wk p . i . ( Figure S4H ) . These results demonstrate that Ugcgf/f//CamKCreERT2 mice develop progressive obesity that is evident in all adipose compartments with constant lean mass and a shift in body composition toward fat accumulation . As tight regulation of energy homeostasis is crucial for body weight maintenance [1] , a metabolic characterization was carried out in order to study the relation of energy intake to energy expenditure . Food intake and metabolizable energy ( EMET ) adjusted to body weight were slightly elevated in Ugcgf/f//CamKCreERT2 mice before the onset of obesity 3 wk p . i . ( Figure 3A , B ) when gangliosides were already depleted in Cre-targeted brain regions . Hyperphagia was no longer evident 6 and 11 wk p . i . , as food intake and EMET were simply elevated due to higher body weight ( Figure 3A , B ) . Fecal excretion of free fatty acids ( FFAs ) as well as energy content of feces and extraction efficiency from the food ( Figure S5A ) were unaltered . Thus , abnormalities in food intake do initially contribute to obesity development , but not for obesity maintenance . Energy expenditure was monitored by indirect calorimetry for 21 h . Before onset of body weight gain , the metabolic rate was indistinguishable from Ugcgf/f mice 2 wk p . i . ( Figure S5B ) . When adjusted for body weight , the average metabolic rate tended to be lower in Ugcgf/f//CamKCreERT2 mice at 5 and 9 wk p . i . ( Figure 3C ) . Spontaneous locomotor activity is one contributor to daily energy expenditure and has been reported to be decreased in obese rodents [21] . However , both before the onset of weight gain and during progressive adiposity , spontaneous open field activity of Ugcgf/f//CamKCreERT2 mice was indistinguishable from control littermates ( Figure S5C ) . The respiratory exchange ratio ( RER ) provides information on metabolic fuel preferences [22] . Ugcgf/f//CamKCreERT2 mice displayed significantly elevated average daily RER values ( Figure 3D ) . This finding suggests a shift from lipid oxidation toward lipid storage [22] . In line with this , fat mobilization in response to fasting as assessed by measuring plasma nonesterified free fatty acids ( NEFAs ) was impaired . Significantly decreased plasma NEFAs were detected in Ugcgf/f//CamKCreERT2 mice 11 wk p . i . ( Figure S5D ) , suggesting a reduced capability to mobilize lipid stores when challenged by food withdrawal . After the onset of weight gain , Ugcgf/f//CamKCreERT2 mice displayed a prominent drop in core body temperature , as exemplarily depicted 10 wk p . i . ( Figure 3E ) . Adipocytes in intrascapular brown adipose tissue ( iBAT ) were enlarged ( Figure S6A ) , suggesting reduced triglyceride turnover . Ultrastructural analysis of iBAT furthermore revealed mitochondrial disorganization as well as a lower average mitochondrial size ( Figure S6B , C ) . Thermogenesis in iBAT is regulated by synergistic actions of thyroid hormones and sympathoadrenergic signaling [23] . Free triiodothyronine ( fT3 ) and free thyroxine ( fT4 ) levels were normal in Ugcgf/f//CamKCreERT2 mice ( Figure S6D , E ) . Thus , thyroid dysfunction was unlikely to account for inappropriate thermoregulation . Decreased sympathetic outflow to adipose tissue is assumed to be associated with impaired lipid mobilization [24] . In fact , both iBAT sympathetic activity , as assessed by norepinephrine ( NE ) turnover rate ( Figure 3F , Figure S6F ) , and NE content ( Figure S6G ) were decreased in Ugcgf/f//CamKCreERT2 mice . These results demonstrate that Ugcgf/f//CamKCreERT2 mice develop progressive obesity and a shift in body composition toward fat accumulation initially supported by hyperphagia , but maintained due to hypometabolism and hypothermia . Several distinct hypothalamic and nonhypothalamic brain regions were targeted by Cre activity in Ugcgf/f//CamKCreERT2 mice . Arc neurons in Ugcgf/f//CamKCreERT2 mice expressing the long form of the ObR were targeted by Cre activity , as demonstrated by co-immunofluorescence of PStat3 and beta-galactosidase ( b-Gal ) in R26R/Ugcgf/+//CamKCreERT2 reporter mice ( Figure 4A ) . Other leptin-responsive neurons outside the Arc also targeted by Cre activity , such as the MnPO are likely in part contributing to the observed phenotype . However , ObR-expressing neurons in the LHA seem to be recessed by Cre activity ( Figure S7A , B ) . In order to furthermore clarify the role of the Arc in obesity development , we injected recombinant adeno-associated viruses encoding either Ugcg and lacZ ( rAAV-Ugcg/LacZ ) or only lacZ ( rAAV-Empty/LacZ ) bilaterally into the Arc of Ugcgf/f//CamKCreERT2 mice after ganglioside depletion before 4 wk p . i . Injection of rAAV-Ugcg/LacZ significantly ameliorated obesity , underlining the importance of Ugcg expression in the Arc for body weight maintenance ( Figure 4B ) . Consistently , serum leptin levels tended to be lower in rAAV-Ugcg/LacZ-treated mice ( Figure 4C ) . We verified correct targeting of the Arc by X-Gal staining of the brains injected with rAAV-Ugcg/LacZ and displayed targeted regions in a schematic drawing as well as a typical staining ( Figures 4D–F and S7C ) . Animals that were not targeted by rAAV-Ugcg/LacZ in the Arc ( rAAV-Ugcg/LacZ missed ) did not improve their weight gain ( Figure S7D ) . Restored ganglioside biosynthesis in the Arc of rAAV-Ugcg-treated animals compared to mice injected with viruses encoding empty plasmid was demonstrated by GD1a immunofluorescence ( Figure 4G and Figure S7E ) . Taken together , these results indicate that loss of GCS expression in the Arc is significantly involved in part of the metabolic deregulation seen in Ugcgf/f//CamKCreERT2 mice . Since the number of neurons in the Arc did not differ between Ugcgf/f//CamKCreERT2 mice and controls ( Figure S8A ) , a functional analysis of the Arc was performed . Leptin signaling in the hypothalamus is crucial for the maintenance of body weight and energy homeostasis . As adipocyte-secreted leptin is a major regulator of body weight in the CNS , we hypothesized that leptin signaling might be disturbed in GCS-deficient neurons of Ugcgf/f//CamKCreERT2 mice . In order to test this hypothesis , we investigated hypothalamic Stat3 phosphorylation ( PStat3 ) in the Arc after peripheral leptin stimulation . Decreased PStat3 was detected by immunofluorescence in the Arc ( Figure 5A ) and by Western blot in mediobasal hypothalamus ( Figure S8B ) . Interestingly , baseline Stat3 levels were elevated in Ugcgf/f//CamKCreERT2 mice ( Figure S8C ) . The PStat3/Stat3 ratio was decreased both at baseline and upon leptin challenge ( Figure S8D ) . It has been shown that deficient ObR signaling due to leptin resistance of the Arc in mice with diet-induced obesity ( DIO ) is a consequence of long-term elevated leptin levels [25]–[27] . The suppressor of cytokine signaling 3 ( SOCS-3 ) is a major negative regulator of the ObR that is elevated in rodent models of leptin resistance [25] , [28] . In line with progressive obesity , Ugcgf/f//CamKCreERT2 mice show indeed elevated leptin levels 7 wk p . i . ( Figure 5B ) . However , expression of hypothalamic Socs-3 did not rise with increasing obesity and leptin levels , as measured 2 , 6 , and 9 wk p . i . ( Figure 5C ) . Moreover , hypothalamic ObR expression , usually elevated in leptin-resistant rodents [29] , [30] , was normal in Ugcgf/f//CamKCreERT2 mice 6 wk p . i . ( Figure 5D ) . To further investigate if GCS-derived gangliosides regulate proper leptin receptor signaling at the level of the plasma membrane in hypothalamic neurons , we first assured that loss of gangliosides would not interfere with ObR transport to the membrane , which would have impaired ObR signaling per se . ObR was labeled by an in situ proximity ligation assay ( PLA ) on non-detergent-perturbed cells by two ObR antibodies . The number of detected surface ObR PLA spots on cells treated with NB-DNJ was similar to control cells ( Figure 5E ) , indicating that ObR at the plasma membrane of ganglioside-depleted hypothalamic cells is not significantly changed compared to control cells . As GCS-derived gangliosides have previously been shown to modulate the activity of plasma-membrane-located receptors through close interactions in both adipocytes [17] and neurons [31] , we investigated ObR interactions with major neuronal gangliosides . The PLA indicating close proximity events [32] indeed revealed proximity between ObR and gangliosides GM1 and GD1a . In demonstration of activity-dependent interaction between GSL and ObR , the number of GD1a/ObR and GM1/ObR PLA spots increased upon stimulation with leptin ( Figures 5F , G and S8E ) . Complex formation between GD1a/GM1 with ObR was further corroborated by co-immunoprecipitation ( Co-IP ) of ObR and GD1a/GM1 in saline- and leptin-stimulated N-41 cells ( Figures 5H and S8F ) . As N-41 cells do not express the complex neuronal gangliosides GD1b and GT1b , potential interactions with ObR had to be analyzed in hypothalamic tissue of Ugcgf/f mice . GD1b and GT1b could not be co-precipitated with ObR ( Figure 5H ) . Ganglioside-depleted cells were then assessed for leptin-dependent signal transduction . Ganglioside-depleted cells did not show the leptin-stimulated increased complex formation between ObR and Jak ( Figures 5I and S8G ) . Time- and dose-dependent Jak phosphorylation could be induced by leptin treatment in N-41 cells and was decreased in NB-DNJ-treated GSL-depleted cells ( Figures 5J and S8H ) . It has to be noted that NB-DNJ evokes ganglioside depletion by only approximately 80%–90% ( Figure S3C ) . Thus , residual gangliosides in the plasma cell membrane may explain the appearance of a P-Jak signal at a late time point after stimulation of NB-DNJ-treated cells . Ganglioside-depleted N-41 cells showed decreased Jak phosphorylation 30 min after stimulation with 0 . 5 µg/ml leptin ( Figure 5J , K ) . These results have now shown that two major neuronal GCS-derived gangliosides , GD1a and GM1 , form dynamically leptin-stimulated complexes with ObR on the plasma membrane and that loss of gangliosides decreases signal transduction in hypothalamic neurons . It is known that mice with deficient leptin receptor ( db/db mice ) function develop obesity and lack hypothalamic responsiveness to leptin stimulation [33] . Regarding the finding that neuronal gangliosides enhance ObR signaling , we hypothesized that hypothalamic neuronal function may be altered in Ugcgf/f//CamKCreERT2 mice . In order to investigate this question , neuronal activity after intraperitoneal ( i . p . ) leptin injection was evaluated by c-Fos staining [34] . Leptin-induced c-Fos formation was normal in non-obese Ugcgf/f//CamKCreERT2 mice 1–2 wk p . i . ( Figure 6A ) . Since ganglioside depletion coincides with the start of the obesity development , Ugcgf/f//CamKCreERT2 mice that were weight-matched to control littermates were analyzed 3–4 wk p . i . Decreased leptin responsiveness could already be observed in the Arc of these mice ( Figure 6B ) as well as in the Arc of obese mice 6 wk p . i . ( Figure 6C ) . Neurons in the nontargeted and non-ganglioside-depleted VMH retained responsiveness to leptin at all time points ( Figure 6D–F ) . As expected , the nontargeted brain stem NTS of Ugcgf/f//CamKCreERT2 mice showed regular leptin-induced c-Fos staining 6 wk p . i . ( Figure S9 ) . Altogether , these results indicate a primary deficiency of ganglioside-depleted hypothalamic neurons to respond adequately to peripheral leptin signals . Antagonistic orexigenic NPY and anorexigenic POMC neurons in the hypothalamic Arc are first-order responsive neurons initiating metabolic adaptations to altered peripheral leptin levels [4] . In order to determine leptin-dependent NPY and POMC neuronal function , neuronal activity and ObR activation were assessed by semiquantitative analysis of c-Fos , PStat3 , and PIP3 formation in response to peripheral leptin injections . Leptin engaged POMC neurons ( α-MSH positive ) in control mice , as indicated by increased c-Fos ( Figure 7A ) . Significantly elevated PStat3 ( Figure 7B ) and PIP3 formation ( Figure S10A ) confirmed activation of their ObR . Before ganglioside depletion ( 1–2 wk p . i . ) , POMC neurons of Ugcgf/f//CamKCreERT2 mice responded normally to leptin . However , c-Fos , PStat3 , and PIP3 formation were not elevated in response to leptin in obese GSL-deficient mice 6 wk p . i . ( Figures 7A , B and S10A ) . No significant changes were found in mediobasal hypothalamus ( MBH ) baseline mRNA expression of Pomc and cocaine- and amphetamine-regulated transcript ( Cart ) mRNA 6 wk p . i . ( Figure S10B ) . While a slight decrease in c-Fos–positive NPY neurons was found in leptin-injected control mice , leptin did not show any such effect in Ugcgf/f//CamKCreERT2 mice 6 wk p . i . ( Figure 7C ) . Similarly , leptin did not raise PStat3 levels in NPY neurons of Ugcgf/f//CamKCreERT2 mice 6 wk p . i . ( Figure 7D ) and did not have any direct effect on PIP3 formation ( Figure S10C ) . Remarkably , basal mRNA expression of Agrp and Npy was markedly elevated in the MBH of Ugcgf/f//CamKCreERT2 mice 6 and 9 wk p . i . , with Agrp already increasing 2 wk p . i . ( Figure S10D ) . In summary , this study has indicated that GCS expression and sufficient gangliosides in neurons of the adult CNS play a seminal role in the regulation of body weight and energy homeostasis . Analysis of the leptin receptor signaling pathway , being one of the most prominent regulators of CNS metabolic control [35] , [36] , revealed that GCS-derived gangliosides interact with ObR on the plasma cell membrane , thereby facilitating ObR-dependent signal transduction ( Figure 8A ) . In Ugcgf/f//CamKCreERT2 mice , leptin responsiveness and neuronal function are impaired in hypothalamic neurons involved in the regulation of energy metabolism ( Figure 8B ) . Consequently , defective ObR signaling contributes to the observed metabolic imbalance and obesity development of mice with ganglioside deficiency in the CNS .
Although the seminal role of CNS feedback responses to peripheral energy signals for the regulation of energy homeostasis has been extensively studied , the role of the lipid microenvironment for energy signal receptor function has not yet been addressed . The present study demonstrates that GCS-derived GSLs are critically involved in a to-date unknown mechanism of hypothalamic control of body weight . In line with the finding that neurons of the constitutive Ugcgf/f//NesCre mice do not show increased apoptosis [15] , ganglioside-deficient hypothalamic neurons are viable and they show normal membrane and organelle appearance both in vivo and in vitro . Electrophysiological recordings from Arc neurons in slices of Ugcgf/f//CamKCreERT2 mice at 12 wk p . i . did not show a major disruption of membrane functions . However , resting membrane potential and action potential threshold were both shifted to slightly more depolarized values . The molecular mechanism underlying the altered membrane potential remains presently elusive . However , it is well feasible that the shift of threshold is secondary to the slight depolarization , which might inactivate a fraction of Na+ channels . In line with the largely normal properties of neurons from brain slices , biophysical parameters of primary hypothalamic neurons devoid of gangliosides were unaltered . Thus , failure of basic electrophysiological membrane functions is unlikely to cause the observed phenotype of mice with ganglioside deficiency . Therefore , the present work focuses on interactions of leptin receptors with the ganglioside-containing lipid microevironment in which receptors are embedded . We show with independent methods that two major neuronal GCS-derived gangliosides , GD1a and GM1 , closely interact with leptin receptors on the neuronal membrane . This interaction is dynamically enhanced by stimulation with leptin . Both Ugcgf/f//CamKCreERT2 mice and ganglioside-depleted hypothalamic cells display deficient ObR signal transduction upon leptin stimulation , as assessed by decreased leptin-induced Jak phosphorylation , Stat3 phosphorylation , and PIP3 formation . Corroborated in situ by deficient leptin responsiveness in Arc neurons of Ugcgf/f//CamKCreERT2 mice , these results indicate that GCS-derived GSLs , primarily gangliosides , are seminal regulators for neuronal leptin signal transduction . Consequently , Ugcgf/f//CamKCreERT2 mice with deficient leptin-induced hypothalamic neuronal responsiveness develop progressive obesity . Numerous hypothalamic feedback systems involved in body weight maintenance are known [1] , [4] . Admittedly , the robust phenotype of Ugcgf/f//CamKCreERT2 mice may be caused by several peripheral hormones and defective ensuing signaling events occurring in various Cre-targeted CNS regions of this mouse model . The brain stem NTS , though an important mediator of metabolic control [37] , is not targeted by activity under the CamK II alpha-dependent Cre recombinase used in this study . Consequently , the NTS shows normal responsiveness to leptin in obese Ugcgf/f//CamKCreERT2 mice and can be excluded to contribute to the observed phenotype . Recent reviews also highlight the LHA as an important regulator of energy balance [38] , [39] . In fact , compensating neurocircuits involving nontargeted CNS regions may be considered for the return of food intake from initial hyperphagia to normal levels in obese mice despite the striking increase in orexigenic neuropeptides . Even though X-Gal staining could be seen in parts of the LHA , we could not verify Cre targeting of a major part of ObR-expressing LHA neurons in Ugcgf/f//CamKCreERT2 mice ( Figure S7B ) . In strong support to this line of reasoning , we demonstrate that partial Ugcg replenishment in the Arc mediated by stereotactic injection of rAAV significantly ameliorates obesity and hyperleptinemia in Ugcgf/f//CamKCreERT2 mice . Even though limited infection of closely attached tissue by rAAV injection could not be definitely excluded , mainly Arc neurons were targeted by this approach , as assessed by X-Gal stainings of brains co-injected with LacZ-expressing viruses . The present investigation has thus been restricted to GCS effects focused on the MBH harboring Arc neurons . Deficient leptin signaling as a consequence of leptin resistance occurs predominantly in the Arc of DIO mice with severe long-term hyperleptimenia [25]–[27] , [40] . Socs-3 is a major negative feedback pathway of ObR signaling [41] . Thus , elevated Socs-3 expression levels are found in the hypothalamus of leptin-resistant rodent models [25] , [42] . In line with observations in obese db/db mice with nonfunctioning ObR [25] , [42] , Socs-3 expression in the Arc remains indistinguishable from control littermates in nonobese and obese Ugcgf/f//CamKCreERT2 mice 2 , 6 , and 12 wk p . i . Elevated hypothalamic ObR expression , as it occurs in DIO mice [29] , [30] , has also been proposed as a potential mechanism playing a role in the development of leptin resistance [28] . However , normal ObR expression in Ugcgf/f//CamKCreERT2 mice supports the hypothesis that the ObR signaling in their neurons must be deficient due to ganglioside loss and not merely due to secondary leptin resistance . Furthermore , the nontargeted hypothalamic VMH and brain stem NTS retain leptin responsiveness even in obese mice 6 wk p . i . These results in combination with the decreased ObR signal transduction in ganglioside-depleted and non-leptin-resistant N-41 cells strongly suggest that loss of GCS-derived GSLs including gangliosides GD1a and GM1 is the reason for failing ObR activation and subsequently inhibited intracellular signaling . GCS-depleted Arc neurons display normal leptin sensitivity 1–2 wk p . i . , a time point when gangliosides are still present . Furthermore , onset of body weight gain , deficient neuronal activity in the Arc , and abolished ObR signaling coincide with ganglioside depletion 3 wk p . i . This strongly suggests that the mentioned defects are due to ganglioside depletion in these cells rather than due to lack of the enzyme GCS itself . Further evidence for the postulate that ganglioside deficiency-dependent inhibition of ObR signaling in hypothalamic neurons leads to impaired neuronal function is based on our in situ results in ganglioside-depleted Arc of both obese and nonobese Ugcgf/f//CamKCreERT2 mice . Whereas leptin injection increases c-Fos immunoreactivity and thus neuronal activity in the Arc neurons of fasted lean mice , this response did not occur in GCS-deficient neurons . Leptin specifically engages POMC neurons . Even though the effects of PI3k- and Stat3-dependent signaling in POMC neurons do not overlap [12] , [43] , both pathways are activated by leptin [11] , [12] , [44] , [45] and contribute to maintenance of energy homeostasis [46] . In ganglioside-depleted POMC neurons , neither PStat3 nor PIP3 formation is increased by peripheral leptin injections , strongly suggesting that defects in both pathways may contribute to partial failure of obesity prevention . As peripheral leptin stimulates both pathways through ObR activation [11] , [47] , defective ObR function is very likely to be assumed . In NPY neurons , it has been demonstrated that Jak-Stat3 signaling plays an important role in maintaining NPY/AgRP-mediated energy homeostasis [48] . Additional ObR-mediated PI3k activation seems to be required for inhibiting Npy and Agrp gene expression [8] . Npy and Agrp expression is markedly increased in the MBH of Ugcgf/f//CamKCreERT2 mice , which may be a consequence of absent leptin-induced PStat3 formation in NPY neurons . On the other hand , leptin-induced PIP3 formation does not differ in neither of the groups , which goes in line with the hypothesis that leptin-dependent PIP3 formation in AgRP/NPY neurons is stimulated by an indirect mechanism involving synaptic transmission [12] . Overactive NPY neurons in obese ObR-deficient Leprfa/fa rats were shown to inhibit sympathetic nervous outflow to BAT and cause hypothermia [7] as observed in Ugcgf/f//CamKCreERT2 mice . With regard to the fact that Npy and Agrp but not Pomc expression differ in Ugcgf/f//CamKCreERT2 mice , the role of GCS expression in regulating neuropeptide expression and secretion has to be elucidated . Especially the role of hypothalamic insulin receptor signaling , which also regulates the expression of Pomc and Npy/Agrp in part similar to ObR signaling [11] and is antagonized by GM3 in the periphery [17] , constitutes a promising target for further clarifying the differential neuropeptide expression . Moreover , a potential contribution of Cre-targeted ObR-expressing neurons in the median preoptic area of Ugcgf/f//CamKCreERT2 mice to hypothermia may also be considered . Dynamic membrane microdomains are widely accepted as critical components involved in membrane receptor functions [16] , [49] . Since GCS-derived gangliosides are important constituents of these microdomains , they potentially interact with and regulate a variety of membrane components including receptors such as Trk receptors [31] and insulin receptors [17] . In contrast to mice with neuron-specific insulin receptor deletion , which only display a gender- and diet-dependent subtle increase in body weight [11] , [50] , the obesity and glucose intolerance observed in db/db mice can be rescued by neuron-specific re-expression of ObR [51] . Furthermore , deficient ObR signaling in POMC neurons of the Arc itself leads to the development of mild obesity [52] . In consideration of these findings—despite the existence of potential alternative pathways that might be impaired in neurons of Ugcgf/f//CamKCreERT2 mice—we ascribe ObR and its regulation of activity to a major function in our model pointing to a novel mechanism for CNS metabolic regulation . We demonstrate that GCS-derived gangliosides GD1a and GM1 closely interact with ObR . The leptin-induced increase in GD1a/ObR and GM1/ObR interaction assumes recruitment of these gangliosides to the ObR upon leptin stimulation . These results in combination with the demonstrated deficient ObR signaling in ganglioside-depleted hypothalamic neurons both in vivo and in vitro leads us to surmise that the lipid microenvironment surrounding the ObR can significantly modulate leptin-dependent intracellular signal transduction in hypothalamic neurons . Altogether , these results provide evidence that GM1 and GD1a are actively involved in enhancing the effects of leptin in hypothalamic neurons . As insulin receptors contain a lysine residue predicted for interaction with GM3 [53] , loss of GM3 synthase showed already a prominent effect on peripheral insulin receptor signaling [17] . It is a widely accepted concept that in the state of insulin resistance in peripheral adipocytes , the IR segregates from caveolae into GM3-enriched microdomains [53] , an endogenous inhibitory mechanism [17] . Indeed , elevated GM3 synthase expression could be detected in adipose tissue of obese Zucker fa/fa rats and ob/ob mice [54] . Pharmacologic GCS inhibition in the periphery has been shown to exert beneficial effects on peripheral insulin sensitivity and liver steatosis [55] , [56] . With regard to the fact that different ganglioside species can exert either stimulatory [18] , [57] or inhibitory [17] effects on membrane receptors , the mentioned studies including the present work support the concept that any perturbation , either loss or excess , of membrane GSLs can alter receptor function . Contributions of GCS-derived lipid raft components apart from gangliosides , namely neutral GSLs in the CNS , to leptin receptor function have yet to be elucidated and constitute a challenging target for future investigations . Besides gangliosides , lactosylceramide has been shown to contribute to formation of lipid microdomains [58] . We , however , propose in the present study that in line with the findings for the insulin receptor , hypothalamic leptin receptor signaling is to a significant extent regulated through interactions with the dominant gangliosides GD1a and GM1 . Recent studies have highlighted the central role of systemic ceramide biosynthesis and GCS in the regulation of energy homeostasis [59] , [60] . In accordance with earlier findings [15] , [61] , we show that neuronal ceramide levels in Ugcgf/f//CamKCreERT2 mice are indistinguishable from control mice , virtually excluding any effects of ceramides . In conclusion , our study highlights the expression of neuronal GCS-derived GSLs , foremost gangliosides , as a novel class of hypothalamic metabolic regulators . Gangliosides GM1 and GD1a enhance the action of leptin on intracellular signaling and neuronal activity , most likely through dynamic changes of the lipid microenvironment of the ObR . We demonstrate by independent methods that gangliosides GD1a and GM1 strongly interact with the ObR upon leptin stimulation . Loss of these gangliosides leads to impaired responsiveness . By this relevant influence on hormone signaling , Ugcg deletion in adult mouse CNS leads to development of progressive obesity , hyperleptinemia , and glucose intolerance . The obesity can be partially ameliorated by restoration of GCS activity and ganglioside expression in the hypothalamic Arc of Ugcgf/f//CamKCreERT2 mice . Neuronal GCS expression therefore constitutes a novel mechanism for hypothalamic regulation of body weight maintenance .
Animals were kept in specific-pathogen-free barrier facilities . Ugcgf/f mice [15] and inducible CamKCreERT2 mice were bred to generate Ugcgf/f//CamKCreERT2 mice and control littermates . Mice were induced with tamoxifen 6 wk after birth for 1 wk as described [19] . We performed experiments in female mice , unless stated otherwise . Mice were fasted overnight ( o/n ) . Blood glucose levels were analyzed prior to i . p . injection of glucose ( 2 g/kg body weight ) . Glucose levels were determined from tail vein blood ( Glucometer Accu Check , Aviva , Roche ) . Food was withdrawn 4 h prior to the insulin sensitivity assay . Mice were injected i . p . with 0 . 75 U/kg human insulin ( Eli Lilly ) , and glucose levels were determined as described above ( see also Text S1 ) . Staining was carried out as described earlier [62] . Cryosections ( male mice ) were incubated with mouse-α-GD1a ( 1∶100 , Millipore ) followed by secondary donkey-α-mouse-Alexa-Fluor 488 ( 1∶200 , Invitrogen ) . Analysis was performed by confocal microscopy ( TCS-SL , Leica ) . Mice were fasted o/n and injected with leptin ( 5 mg/kg , Peprotech ) or saline between 8 . 00 a . m . and 10 . 00 a . m . Animals were sacrificed at indicated time points and transcardially perfused with 4% paraformaldehyde ( PFA ) . We prepared 40 µm cryosections covering the Arc . Alternating sections were collected in series for subsequent free-floating section immunostainings . First antibodies used for immunostaining were rabbit-α-PStat3 ( 1∶100 , Cell Signaling Technology ) , rabbit-α-c-Fos ( 1∶100 , Santa Cruz ) , and FITC-conjugated α-PIP3 ( 1∶100 , Echelon ) . Secondary antibody was donkey-α-rabbit-Alexa-Fluor 488 ( 1∶200 , Invitrogen ) . Sections were subsequently incubated with either goat-α-NPY ( 1∶50 , Santa Cruz ) or sheep-α-MSH ( 1∶1000 , Millipore ) followed by secondary antibodies ( donkey-α-goat-Alexa-Fluor 546 , donkey-α-sheep-Alexa-Fluor 546 , 1∶200 , Invitrogen ) . Stainings were analyzed by confocal microscopy ( TCS-SL , Leica ) . Neurons with nuclear ( c-Fos , PStat3 ) or cytoplasmic ( PIP3 , NPY , and α-MSH ) staining above background were considered positive . Immunofluorescence for beta galactosidase was performed as described earlier [63] . The AAV Helper-free System ( Agilent Technologies Inc . ) was used for preparation of rAAV . Full-length mouse Ugcg cDNA was cloned into the pAAV-MCS vector from the Helper-free System ( pAAV-Ugcg ) . Viruses were generated according to the manufacturer's guidelines and purified as described earlier [64] . Bilateral stereotaxic injections were performed as described [65] . We injected 400 nl virus solution containing equal volumes of viruses carrying Ugcg and lacZ ( ∼1 . 8*1011 genome copies/ml ) into the Arc of each hemisphere ( caudal to bregma: 1 . 4 mm , 1 . 44 mm; lateral: 0 . 25 mm; ventral: 5 . 7 mm ) . After surgery , mice were maintained with ad libitum access to lab chow and body weight was monitored weekly . Immortalized hypothalamic neurons were purchased from CELLutions Biosystems ( mHypoE N-41 , Cedarlane ) and cultured according to the manufacturer's guidelines . GCS was inhibited with NB-DNJ treatment ( 100 µM , 7 d , Sigma; Tocris ) . Eight thousand N-41 cells were seeded onto coverslips and incubated at 37° o/n . The 3 h serum-starved cells were stimulated with leptin ( 100 ng/ml , Peprotech ) for 10 min , washed with PBS and fixed in 4% PFA for 15 min . Cells were blocked with 5% skim milk/PBS . PLA was performed with primary antibodies against ObR ( 1∶50 , Santa Cruz ) , GD1a ( 1∶100 , Millipore ) , and GM1 ( 1∶10 , Matreya ) . PLA was performed according to the manufacturer's guidelines ( Duolink Orange Detection System , Olink Biosciences ) . Formation of PLA spots was analyzed by fluorescence microscopy ( Zeiss Cell Observer ) . Mice were injected with leptin or saline as described above and sacrificed 30 min later . The MBH was dissected homogenized on ice in lysis buffer ( 20 mM HEPES , 25 mM KCl , 250 mM sucrose , 2 mM MgCl2 , 0 . 5 mM DTT , 1% digitonin ) containing proteinase inhibitor ( Roche ) and phosphatase inhibitor cocktail ( Sigma ) . Immortalized hypothalamic cells were treated with either saline or 100 µM NB-DNJ for 7 d , serum starved for 4 h , and subsequently treated with either saline or leptin ( 1 , 000 ng/ml , Peprotech , 1 h ) . Cells were lysed on ice in lysis buffer . Protein concentrations were determined by Bradford assay ( Sigma ) . Western blots were performed as described earlier [25] . Primary antibodies: rabbit-α-PStat3 , rabbit-α-Stat3 , rabbit-α-PJak , rabbit-α-Jak ( 1∶1 , 000 , Cell Signaling Technology ) , mouse-α-tubulin ( 1∶5 , 000 , Zymed Labs ) , and rabbit-α-actin ( 1∶1 , 000 , Santa Cruz ) . Secondary antibodies: HRP-conjugated α-rabbit-IgG ( 1∶1 , 000 , Dako ) and HRP-conjugated α-mouse-IgG ( 1∶5 , 000 , Santa Cruz ) . Bands were visualized by chemiluminescence ( Amersham ) and quantified ( ImageJ , NIH ) . Four hours serum-starved N-41 cells were treated with leptin ( 1 , 000 ng/ml , 25 min ) . Cells were lysed in IP buffer [50 mM HEPES , pH 7 . 0 , 150 mM NaCl , 10% glycerol , 1% Triton-X , 1 . 5 mM MgCl2 , 1 mM EDTA , proteinase inhibitor cocktail ( Roche ) ] . Co-IP for ObR/Jak was performed as described earlier [36] . ObR/GD1a- and ObR/GM1-Co-IP and subsequent lipid extraction and analysis was performed as described earlier [31] , [53] . Anti-ObR were incubated at 4°C o/n . Immunoprecipitated lipids were desalted on an RP-18 column , spotted on a TLC , and run in solvent ( chloroform/methanol/0 . 2% CaCl2; 60∶35∶8 , by vol . ) . GD1a was visualized with mouse α-GD1a ( 1∶1 , 000 , 4°C overnight , Millipore ) on the TLC by immune overlay staining as described earlier [66] . Serum leptin and NEFAs were determined by commercially available kits according to the manufacturer's guideline [Leptin-ELISA ( Linco ) ; NEFA-HR2 kit ( WAKO Chemicals ) ] . NEFAs were measured in male mice . NETO rate in iBAT was determined as described earlier [67] . Tissue NE was measured by reversed-phase HPLC with electrochemical detection ( Chrome Systems , Germany ) ( see also Text S1 ) . Body weight was measured once a week . Metabolic measurements were carried out in an open circuit respiratory system ( SM-MARS , Sable Systems , USA ) . VO2 and VCO2 per mouse were analyzed for 21 h to determine the RER = VCO2/VO2 and HP ( mW ) . Whole body composition was determined by noninvasive NMR analysis ( Mini-Spec , Bruker Optics ) . Core body temperature was measured with a rectal probe ( ALMENO 2390-1 , Ahlborn ) ( see also Text S1 ) . Total RNA of the MBH was extracted from nonfasted mice as described earlier [68] . RNA was reversely transcribed by Superscript II Reverse Transcriptase ( Invitrogen ) and cDNA was quantified using the LC FastStart DNA Master SYBR Green I kit ( Roche ) according to the manufacturer's guidelines and the Light Cycler ( Roche ) ( see also Text S1 ) . R26R/Ugcgf/+//CamKCreERT2 mice and R26R/Ugcgf/+ mice were induced with tamoxifen i . p . 6 wk after birth as described . At 3 d p . i . , animals were sacrificed , and brains were removed and frozen on dry ice . X-Gal staining was performed as described previously [69] . Similarly , β-galactosidase activity in brains of rAAV-Ugcg/LacZ- , rAAV-Empty/LacZ- , and rAAV-LacZ-injected mice was visualized 7 d after virus injection . GSLs were extracted and separated into neutral and acidic fractions containing gangliosides as described earlier [15] . The amount of GSLs spotted onto a plate by a TLC applicator ( Camag , USA ) was normalized to tissue protein content determined by the Lowry method [70] . TLC running solvent for acidic GSL was chloroform/methanol/0 . 2% CaCl2 ( 45∶45∶10 by vol ) . GSLs were visualized with 0 . 2% orcinol in 10% sulphuric acid at 120°C for 10 min . Ceramide was extracted as described earlier [15] and spotted onto a TLC plate . Running solvent for ceramide was chloroform/methanol/acetic acid ( 190∶9∶1 by vol ) , and ceramide was visualized with 10% CuSO4 in 8% H3PO4 at 180°C for 10 min . Lipid content was quantified by densitometry ( Shimadzu , Japan ) . Unless stated elsewhere , results were analyzed by a two-tailed , unpaired Student's t test ( Graph Pad Prism , Graph Pad Software , Inc . ) . To analyze main effects of genotype on metabolizable energy or energy expenditure , body weight was employed as a co-factor in a linear regression model to account for the confounding effect of body size on energy metabolism parameters [71] . p≤0 . 05 was considered statistically significant and marked * . p≤0 . 01 was marked ** , and p≤0 . 001 was marked *** . | Obesity is a growing health threat that affects nearly half a billion people worldwide , and its incidence rates in lower income countries are rising dramatically . As obesity is a major risk factor for type II diabetes and cardiovascular disease , significant effort has been put into the exploration of causes , prevention , and potential treatment . Recent research has demonstrated that a region of the brain called the hypothalamus is a major integrator of metabolic and nutrient signals , adapting food intake and energy expenditure to current metabolic needs . Leptin or insulin receptors located in the plasma cell membrane of neurons sense energy signals from the body . They transmit this information inside the cell , which then regulates neuronal function . In this study , we show that leptin receptors interact with gangliosides , a class of plasma membrane lipids . This interaction is a prerequisite for proper receptor activation . Consequently , ganglioside loss in hypothalamic neurons inhibits leptin receptor signal transduction in response to energy metabolites . Furthermore , mice lacking gangliosides in distinct forebrain areas , amongst them the hypothalamus , develop progressive obesity and hypothermia . Our results suggest a previously unknown regulatory mechanism of plasma membrane lipids for hypothalamic control of body weight . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"neurochemistry",
"protein",
"interactions",
"neuroscience",
"gene",
"function",
"animal",
"models",
"model",
"organisms",
"membrane",
"receptor",
"signaling",
"membranes",
"and",
"sorting",
"biology",
"proteomics",
"neuroendocrinology",
"mouse",
"biochemistry",
"signal",
"transduction",
"cellular",
"neuroscience",
"cell",
"biology",
"genetics",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics",
"signaling",
"cascades"
]
| 2013 | Neuronal Expression of Glucosylceramide Synthase in Central Nervous System Regulates Body Weight and Energy Homeostasis |
Neurons are equipped with homeostatic mechanisms that counteract long-term perturbations of their average activity and thereby keep neurons in a healthy and information-rich operating regime . While homeostasis is believed to be crucial for neural function , a systematic analysis of homeostatic control has largely been lacking . The analysis presented here analyses the necessary conditions for stable homeostatic control . We consider networks of neurons with homeostasis and show that homeostatic control that is stable for single neurons , can destabilize activity in otherwise stable recurrent networks leading to strong non-abating oscillations in the activity . This instability can be prevented by slowing down the homeostatic control . The stronger the network recurrence , the slower the homeostasis has to be . Next , we consider how non-linearities in the neural activation function affect these constraints . Finally , we consider the case that homeostatic feedback is mediated via a cascade of multiple intermediate stages . Counter-intuitively , the addition of extra stages in the homeostatic control loop further destabilizes activity in single neurons and networks . Our theoretical framework for homeostasis thus reveals previously unconsidered constraints on homeostasis in biological networks , and identifies conditions that require the slow time-constants of homeostatic regulation observed experimentally .
Neurons in the brain are subject to varying conditions . Developmental processes , synaptic plasticity , changes in the sensory signals , and tissue damage can lead to under- or overstimulation of neurons . Both under- and overstimulation are undesirable: prolonged periods of excessive activity are potentially damaging and energy inefficient , while prolonged low activity is information poor . Neural homeostasis is believed to prevent these situations by adjusting the neural parameters and keeping neurons in an optimal operating regime . Such a regime can be defined from information processing requirements [1 , 2] , possibly supplemented with constraints on energy consumption [3] . As homeostasis can greatly enhance computational power [4–6] , and a number of diseases have been linked to deficits in homeostasis [7–10] , it is important to know the fundamental properties of homeostatic regulation , its failure modes , and its constraints . One distinguishes two homeostatic mechanisms: synaptic and intrinsic excitability homeostasis [11 , 12] . In case of over-excitement , synaptic homeostasis scales excitatory synapses down and inhibitory synapses up , while intrinsic homeostasis increases the firing threshold of neurons . Intrinsic homeostasis is the subject of this study . Intrinsic homeostasis correlates biophysically to changes in the density of voltage gated ion channels , [13–16] , as well as the ion channel location in the axon hillock [17 , 18] . All homeostatic mechanisms include an activity sensor and a negative feedback that counters deviations of the activity from a desired value . Control theory describes the properties of feedback controllers and the role of its parameters [19] . In engineering one typically strives to bring a system rapidly to its desired state with minimal residual error . It is reasonable to assume that neural homeostasis has to be fairly rapid too in order to be effective , although it should not interfere with the typical timescales of perceptual input or of neural processing ( millisecond to seconds ) . However , intrinsic excitability homeostasis is typically much slower , on the order of many hours to days ( [13 , 15 , 20 , 21] , but see [14] ) . One hypothesis is that this is sufficiently fast to keep up with typical natural perturbations , but an alternative hypothesis , explored here , is that stable control necessitates such slow homeostasis . Note that the speed of homeostasis is the time it takes to reach a new equilibrium after a perturbation and does not rule out that homeostatic compensation can start immediately after the perturbation; it just takes a long time to reach its final value . In computational studies homeostatic parameters are usually adjusted by hand to prevent instability , but a systematic treatment , in particular in networks , is lacking . In a recent study a network with excitatory and inhibitory populations with distinct homeostatic control was studied and with linear stability analysis it was found that instabilities can occur when the inhibitory homeostasis is faster than the excitatory one [22] . However , numerous questions remain: Is homeostatic control consisting of multiple stages equally stable ? How do non-linearities in the input-output relation of the neurons affect results ? Finally , because in that study a separation of time-scales between the neural activity and homeostasis was assumed , only a constraint on the ratio of homeostatic speeds of the two populations was found , but not on their absolute speeds . It raises the question how the homeostatic speed relates to the neural time-constants . In this study we analyze three aspects of the stability conditions for networks of neurons equipped with homeostasis . 1 ) We show that homeostasis can destabilize otherwise stable networks and that , depending on the amount of recurrence , stable homeostatic feedback needs to be slower for networks than for single neurons . 2 ) We analyze how homeostatic stability is affected by non-linearities in the neuron’s input-output relation . In general systems with non-linearities require slower homeostasis than linear analysis predicts . 3 ) We show that having multiple intermediate stages in the feedback loop , common in biological signaling cascades , tends to destabilize control , despite the overall feedback being slower . The results put constraints on the design and interpretation of homeostatic control and help to understand biological homeostasis .
To examine the stability of homeostatic control we first analyze a single neuron with homeostasis , a schematic is shown in Fig 1A . We describe the activity of the neuron as a function of time with a firing rate r1 ( t ) . A common approximation for the firing rate dynamics is τ 1 d r 1 ( t ) d t = - r 1 ( t ) + g ( u ( t ) - θ ( t ) ) ( 1 ) which can be understood as follows: The time-constant τ1 determines how rapidly the firing rate changes in response to changes in the input and how rapidly it decays in the absence of input . We use τ1 = 10 ms . The value of τ1 serves as the time-constant with respect to which all the other time-constants in the system will be defined . As only the ratios between time-constants will matter , the results are straightforwardly adapted to other values of τ1 . The f-I curve g ( ) describes the relation between net input to the neuron and its firing rate . We assume that homeostasis acts effectively as a bias current which shifts the f-I curve , consistent with experimental data [15] . The total input is u ( t ) − θ ( t ) , where u ( t ) is proportional to external input current to the neuron , typically from synaptic input . Crucially , θ ( t ) is the homeostatically controlled firing threshold of the neuron . While physiologically both the threshold current and threshold voltage of neurons are affected by homeostasis [13] , our model comprises both indistinguishably . The homeostatic controller takes its input from averaged activity , rather than reading out activity directly ( see also section Cascaded Homeostatic Control below ) . To obtain the averaged activity r2 ( t ) of the neuron , the firing rate r1 ( t ) is filtered with a linear first order filter with a time-constant τ2 τ 2 d r 2 ( t ) d t = - r 2 ( t ) + r 1 ( t ) ( 2 ) Biophysically , the intra-cellular calcium concentration is a very likely candidate for this sensor [11] in which case τ2 is around 50ms . The last step in the model is to integrate the difference between the average activity and the pre-defined desired activity level rgoal τ 3 d r 3 ( t ) d t = r 2 ( t ) - r g o a l ( 3 ) rgoal was typically set to 1Hz , but its value is inconsequential . The feedback loop is closed by setting the threshold in Eq ( 1 ) equal to this signal , that is θ ( t ) = r3 ( t ) . Thus , if the activity remains high for too long , r2 and r3 increase , increasing the threshold and lowering the firing rate , and vice versa if the activity is below the set-point rgoal for too long . An attentive reader might have noticed that θ is a current , while r3 is a rate . Formally this inconsistency can be resolved by defining θ ( t ) = γr3 ( t ) where γ has dimensions A/Hz , and by giving α ( defined below ) dimensions Hz/A . However , for simplicity we use dimensionless units; this does in no way affect our results . Also note , that while r1 is assumed positive , r3 and θ are not as they are the difference between actual and goal rate and thus can take negative values . Note that in contrast to the earlier equations , Eq ( 3 ) does not have a decay term on the right hand side , i . e . a term of the form −r3 ( t ) . This means that instead of a leaky integrator , it is a perfect integrator which keeps accumulating the error in the rate ( r2 ( t ) − rgoal ) without any decay . Mathematically , this can be seen by re-writing Eq ( 3 ) as r 3 ( t ) = 1 τ 3 ∫ − ∞ t [ r 2 ( t ′ ) − r g o a l ] d t ′ . Perfect integrators are commonly used in engineering solutions such as PID controllers and are very robust . A perfect integrator ensures that , provided the system is stable , the goal value rgoal is eventually always reached , as otherwise r3 ( t ) keeps accumulating . The time-constant τ3 is therefore not strictly a filter time-constant , but it determines how rapidly errors are integrated and thus how quickly homeostasis acts . Although it might appear challenging to build perfect integrators in biology , evidence for them has been found in bacterial chemotaxis [23 , 24] . It is straightforward to extend our theory to a leaky integrator; for small leaks , this does not affect our results . In general the f-I curve is non-linear . To determine the stability to small perturbations around the homeostatic set-point , a linear approximation of the f-I curve is made , g ( x⋆ ) = α ( x − x⋆ ) , where x⋆ = g−1 ( rgoal ) is the total input at the set-point , and α is the slope of the f/I curve at the set-point . An extension to general non-linear f-I curves is presented below . Using the linearization , we can borrow results from linear control theory to examine the stability of the set of differential equations that define the neural and homeostatic dynamics . One needs to solve the differential equations at the equilibrium point and check whether the solutions diverge . Various equivalent approaches have been developed to determine stability of controllers [25] . Here we write the set of first order equations , Eqs ( 1 ) – ( 3 ) in matrix form d d t ( r 1 ( t ) r 2 ( t ) r 3 ( t ) ) = M ( r 1 ( t ) r 2 ( t ) r 3 ( t ) ) + b ( t ) with matrix M = ( - 1 τ 1 0 - α τ 1 1 τ 2 - 1 τ 2 0 0 1 τ 3 0 ) and vector b ( t ) = ( α τ 1 u ( t ) 0 - 1 τ 3 r g o a l ) In this linearized case , the gain α can be absorbed in τ3 . A shallower f-I curve ( α < 1 ) implies a weaker feedback , and is fully equivalent to a proportionally slower τ3; in both cases it takes longer for the system to attain the goal value . In the limit that τ3 ≫ τ1 , τ2 the firing rate settles exponentially with a time-constant τ3/α in response to a perturbation , that is , r 1 ( t ) − r g o a l ∝ e − α t / τ 3 . The theory of differential equations states that the solution to the set of equations is the sum of a particular solution ( which is unimportant for our purposes ) and solutions to the homogeneous equation , which is the equation with b = 0 . With the ‘ansatz’ ri ( t ) = ci eλt , one finds that in order to solve the homogeneous equation , the vector c = ( c1 , c2 , c3 ) must be an eigenvector of M with eigenvalue λ . This means that λ has to solve the characteristic polynomial , det ( M − λI ) = − ( 1 + τ1 λ ) ( 1 + τ2 λ ) τ3 λ − α = 0 . The three eigenvalues of M are in general complex numbers and determine the stability of each mode as follows: If an eigenvalue is real and negative , the corresponding mode is stable as the exponential eλt decays to zero over time . If an eigenvalue is complex and the real part is negative , the corresponding mode decays over time as a damped oscillation . In the context of homeostasis such activity oscillations might be biologically undesirable , in particular when they persist for many cycles . Finally , the solution of the linearized system will diverge if any of the eigenvalues has a positive real part . In practice , some mechanism , such as a squashing or rectifying f-I curve , will restrain the firing rate and strong sustained oscillations in the firing rate will occur . ( For two dimensional systems this can be proven using the Poincare-Bendixson theorem [26] ) . In this case homeostatic control is unstable . Which of these above scenarios occurs depends in our model solely on the ratio between the three τi parameters . In most of what follows , we determine the required value of τ3 for given τ1 and τ2 , i . e . we determine the required homeostatic timeconstant given the neural and calcium timeconstants . Fig 1B shows simulated responses of a single neuron’s firing rate r1 ( t ) , and the threshold variable r3 ( t ) to a step input for various settings of the time-constants . It can be observed that only for extremely short values of τ3 the neuron is unstable ( striped region ) . In this case the firing rate oscillates continuously . In the gray region the neuron is stable but displays damped oscillations after changes in the activity . Stability without oscillation ( white region ) can always be achieved by taking τ3 slow enough . The explicit stability condition follows from the Routh–Hurwitz stability criterion ( see Methods ) . It yields τ 3 > τ 3 0 where τ 3 0 ≡ α τ 1 τ 2 τ 1 + τ 2 The above stability criterion confirms the intuition that slower feedback is more stable than fast feedback . When τ2 is 50ms , τ3 needs to be longer than 8ms to obtain stability ( assuming α = 1 ) . Our main assumption is that the oscillation associated to the instability is to be avoided at all cost . While oscillations by themselves occur in many circumstances in neuroscience and have important functional roles , these particular oscillations here are uncontrollable and can not be stopped . Furthermore , the external input to the neuron has virtually no control over the oscillation’s phase , frequency or amplitude . Turning stimulation on or off hardly affects the neuron’s oscillatory activity , Fig 1B , bottom right . The oscillating state is almost the opposite of homeostasis , as it would be challenging for neurons to code information when oscillating like this . Especially if excitability is regulated through the insertion and removal of ion-channels , the oscillating state is also metabolically costly . To warrant the absence of damped oscillations a similar criterion can be derived ( Methods , Eq ( 13 ) ) and in this example case τ3 needs to be longer than 220ms to avoid damped oscillations . In summary , for single neurons simple homeostatic control is stable even when it is very fast . Therefore , the stability of the homeostatic controller would not appear an issue for homeostasis of intrinsic excitability . Such very fast homeostasis might not even be desirable , because it will filter out components of the input slower than the homeostatic control . For instance in Fig 1B ( top left ) , the neural response equals the stimulus with changes slower than ∼ 100ms filtered out . While the above linear treatment covers the stability to small perturbations , stability to arbitrary perturbations is what ultimately matters . To analyze this the non-linearity of the f/I curve has to be taken into account . Unfortunately , non-linear stability analysis is generally much harder . Furthermore , stability proofs are typically sufficiency proofs , necessity proofs are rare . For the particular system under study , Eqs 1–3 , it can be shown that it is guaranteed to be stable only when for all x ( see Supplementary Information ) 0 < g ˜ ( x ) x < τ 3 τ 3 0 ( 4 ) where g ˜ ( x ) = g ( x + x ⋆ ) − g ( x ⋆ ) = g ( x + x ⋆ ) − r g o a l is the f-I curve re-centered around the set-point ( g ˜ ( 0 ) = 0 ) . The criterion thus becomes τ 3 > [ max g ˜ ( x ) x ] τ 1 τ 2 τ 1 + τ 2 It is known as the Aizerman conjecture [27] , and although not generally true and counter-examples do exist , it is known to hold for this particular 3 dimensional system [28] . The criterion replaces the local slope α with the slope of the line that goes through ( x⋆ , rgoal ) and envelopes the f-I curve . Note that for a linear f-I curve this criterion equals the linear criterion , otherwise it is always more stringent than the linear criterion . We will further explore this criterion below . Next , we analyze the stability of homeostatic control in a network of neurons . For networks the conditions on homeostatic control are more stringent than for single neurons . In Fig 2A the population firing rate of a simulated network is plotted as the strength of the recurrent connections is increased while all other network and homeostasis parameters are fixed ( left to right plot ) . Increasing the recurrent connections in the network leads to strong , persistent oscillations , while , importantly , without homeostasis the network is stable ( top panels , dotted curves ) . To prevent instability the feedback needs to be slower in networks than for single neurons . To analyze this , again first in the limit of small perturbations , we consider a network of N neurons connected with fast synapses via an N × N weight matrix V . In the absence of homeostasis the firing rate dynamics obeys τ 1 d d t r 1 ( t ) = - r 1 ( t ) + G ( V r 1 ( t ) + u ( t ) ) ( 5 ) where r1 ( t ) is a N-dimensional vector containing all firing rates in the network , and u ( t ) is a vector of external input to the neurons in the network and G denotes the relation g ( ) working on each of its elements . The recurrent feedback is contained in the term V r1 ( t ) . Linearized this becomes τ 1 d d t r 1 ( t ) = ( W - I ) r 1 ( t ) + α u ( t ) where we define the gain-scaled weight matrix W = αV , where α is again the slope of the f/I curve at the homeostatic set-point . After the linearization , the dynamics of these networks can be analyzed in terms of the eigen-modes . We denote the eigenvalues of W with wn . There are no a priori restrictions on W . The synapses can be excitatory or inhibitory , and one can impose Dale’s principle ( which will affect the eigenvalue spectrum of the weight matrix [29] , but does not change our results otherwise ) . For clarity we focus in the main text on cases where the eigenvalues are real , in the Methods the generalization to complex eigenvalues is presented . A typical example of a network where the eigenvalues are guaranteed to be real are symmetric networks that are used in many applications , such as noise filtering and evidence accumulation [30] . For such networks we define the largest eigenvalue , wm = max ( wn ) , as the recurrence of the network . The recurrent excitatory connectivity slows down the effective time-constant of a given mode [30 , 31] . This can be seen by writing the equation for each mode as τ 1 1 − w n d r n ( t ) d t = − r n ( t ) + α 1 − w n u n ( t ) , from which the time-constant of a given mode is then identified as τ1/ ( 1 − wn ) . The network time-constant is defined as the time-constant of the slowest mode , i . e . τ1/ ( 1 − wm ) . Without homeostasis the network activity is stable as long as wm < 1 . In the presence of homeostatic regulation , the system becomes 3N-dimensional . It is described by the rate of each neuron r1 , its filtered version r2 , and its threshold r3 . The corresponding linearized differential equation is d d t ( r 1 r 2 r 3 ) = M ( r 1 r 2 r 3 ) + ( α τ 1 u ( t ) 0 - 1 τ 3 r g o a l ) where M is now a block-matrix , given by M = ( 1 τ 1 ( W - I ) 0 - α τ 1 I 1 τ 2 I - 1 τ 2 I 0 0 1 τ 3 I 0 ) ( 6 ) We proceed as above to determine the stability of this system . In analogy with the single neuron case , there are three eigenvalues for the full system per eigenvector of W , so that we obtain 3N eigenvalues . In principle , one should now research the stability of each eigenvector of W . Yet the analysis can be simplified for networks with strictly real eigenvalues . In a network without homeostasis the most critical mode is the one with the largest eigenvalue . This also holds in networks with homeostasis: the network is stable if and only if this mode is stable ( see Methods for proof ) . Thus , rather than analyzing the full network , we only need to analyze the stability of this most critical mode , which is given by a three dimensional system similar to the single neuron system studied above with the pre-factor of r1 ( t ) on the right hand side as only modification , τ 1 d r 1 ( t ) d t = - [ 1 - w m ] r 1 ( t ) + α [ u ( t ) - θ ( t ) ] ( 7 ) The other equations for homeostatic control , Eqs 2 and 3 , remain unchanged . The resulting three dimensional system describes the dynamics of the critical eigenmode and its homeostatic variables . The stability is now determined by the roots of the polynomial ( 1 - w m + τ 1 λ ) ( 1 + τ 2 λ ) τ 3 λ + α = 0 ( 8 ) The network is again stable if all roots of this polynomial have a negative real part . Application of the Routh–Hurwitz criterion ( Methods ) yields the stability condition τ 3 > τ 3 c r i t , where τ 3 c r i t = α 1 - w m [ τ 1 τ 2 τ 1 + ( 1 - w m ) τ 2 ] ( 9 ) In Fig 2B we vary the integration time of the network by changing wm and plot the values for τ3 required for stability . The minimal , critical value of τ3 is shown with the solid black curve . Eq ( 9 ) yields for ( 1 − wm ) τ2 ≫ τ1 that τ3 ≳ τ1/ ( 1 − wm ) 2 , while for ( 1 − wm ) τ2 ≪ τ1 this can be approximated as τ3 ≳ τ2/ ( 1 − wm ) . When , for example , the network has an integration time-constant of 1s , τ3 needs to be at least 4 . 8s to prevent instability . If the network integration time-constant is 10s , this increases to 50s . The linear stability analysis of networks with complex eigenvalues of the weight matrix is analogous , except that rather than only the largest , each eigenvalue needs to be checked ( see Methods ) . The stability to arbitrary perturbations again requires taking the non-linearity of the f/I curve into account . Nonlinear stability analysis of the network ( Eq ( 5 ) supplemented with homeostasis ) is presented in the Supplementary Information based on Lyapunov theory . The τ3 , unfortunately a rather complicated expression , required for stability depends on the maximum slope of the f/I curve . The criterion value for τ3 is always slower than the value found using the linear theory , Eq ( 9 ) , because , first , global stability implies local stability , but not vice versa , and secondly , the non-linear theory only provides a sufficiency condition for stability . We apply the criterion below . The sustained oscillations associated to the instability are detrimental for neural information processing as they are uncontrollable and hinder information coding , yet are energetically expensive . Damped oscillations are less harmful . However , in particular for strongly recurrent networks , damped oscillations can interfere with the desired network response . As an illustration of this we show the response of an ideal leaky integrator , such as might be used for evidence integration in Fig 2C ( gray curve ) . When rapid homeostasis is active , the response shows strong oscillations that occludes the network’s integrative properties ( black curve ) . Only when homeostasis is made so slow that no damped oscillations occur ( dashed curve ) , the response approximates that of the ideal integrator . In recurrent networks the value of τ3 required to ensure homeostasis without damped oscillations is much larger than the value required to prevent persistent oscillation , compare dashed curve to solid curve in Fig 2B . Interestingly , as is shown in the Methods , for long integration times it increases as the square of the integration time ( slope of 2 on the log-log plot ) . For example if the network integration time-constant is 1s , the minimal homeostatic time-constant is 420s to prevent transient oscillations . And if the network integration time-constant is 10s , a realistic value in for instance working memory networks [32] , this values increases to 11hrs . In summary , in particular if an oscillation-free response is required , strongly recurrent networks with long time-constants require homeostasis many orders of magnitudes slower than single neurons and there is a strong dependence on the network time-constant . To examine the generality of the results we included variability and heterogeneity in the model . First , we wondered whether heterogeneity in the time-constants , likely to occur in real neurons , could prevent the synchronous oscillations associated to the instability . Hereto we drew for each neuron the homeostatic time-constants from a gamma-distribution with an adjustable coefficient of variation ( CV ) and a given mean . To quantify the destabilizing effect of homeostasis , we defined the dimensionless critical recurrence strength wc . It is the maximal recurrence for which the network is still stable , possibly with damped oscillations . That is , wc is the value at which the real part of the largest eigenvalue crosses zero . For networks without homeostasis , the critical recurrence is one , but homeostasis limits this to lower values . Stability is again determined by the stability matrix of Eq ( 6 ) , however , in the heterogeneous case the dimension reduction is not possible and the spectrum of the full matrix was examined . When the CV is zero , all neurons have the same set of time-constants and the stability corresponded to that of the homogeneous networks . As the heterogeneity increased , the average maximal allowed recurrence first increased slightly after which it decreased , Fig 3A . Moreover , as can be seen from the error bars , for a given realization of the time-constants , the stability can either be higher or lower than that of the homogeneous network . Hence random heterogeneity of the time-constants does not robustly lead to increased stability . The effect of heterogeneity on the transition between the damped oscillatory and oscillation-free regime is similar . Next , we added noise to the neurons and analyzed how this affected the transition to instability . The noise might potentially have a stabilizing effect by de-synchronizing the population . Gaussian noise with a correlation time of 1ms and a standard deviation equivalent to 0 . 1Hz was added to the input . We measured the fluctuations as the standard deviation of the population firing rate once the system had reached steady state , Fig 3B . These fluctuations comprise both the effect of noise and the periodic oscillations caused by the instability . Without noise , fluctuations are absent when the recurrence is less than the critical amount ( Fig 2A ) , and are strong above this point , Fig 3B ( dashed curve ) . With noise , fluctuations are always present ( solid curve ) and increase close to the transition to instability . Above the transition point the fluctuations are similar to the noise-free model . In a network with homeostasis the resulting fluctuations were always larger than without ( gray line ) . The reason is that in the homeostatic network the noise is continuously exciting a damped resonant system , amplifying the fluctuations . Importantly , the amount of recurrence at which the transition to the unstable regime occurs , does not shift with noise , implying that noise does not increase stability . Rather the opposite happens . Already in the approach to instability ( around a recurrence of 0 . 75 ) , homeostasis increases the fluctuations in the population firing rate ( black curve diverges from gray curve ) . Next , we compared the theory to simulations of networks of spiking neurons ( see Methods ) . The connection strength was such that the network was stable and noise was injected to all neurons to prevent population synchrony . The homeostatic control was implemented exactly as above: the average rate r2 ( t ) was extracted by filtering the spikes ( τ2 = 50ms ) , and this was fed into the integrator as above . The homeostatic target rate was set to 4Hz . In this asynchronous regime , the population firing rate of the spiking network can be reasonably approximated by the rate equation with a non-linear f-I curve ( Eq ( 1 ) ) and recurrent feedback . In order to be able to compare the spiking network to the theory we turned homeostasis off and gave small step stimuli to the network and measured how quickly the firing rate equilibrated as a function of the connection strength , Fig 4A . In the rate model this equilibration time is τ1/ ( 1 − wm ) . A fit to this relation gave τ1 ≈ ( 11 . 5±1 . 5 ) ms and also yielded the proportionality between the synaptic strength and wm , which we calibrated as above so that wm = 1 corresponds to the critical amount of recurrence in the linearized model without homeostasis . As networks close to critical recurrence are slow and difficult to simulate , we used a value of wm = 0 . 6 , so the required homeostatic time-constants are fairly short . The linear stability criterion , Eq ( 9 ) yields that when τ3 ≥ 64ms the network should be stable . However , the simulated network is less stable than the linear criterion predicts . The network shows strong oscillations for such rapid homeostasis , Fig 4B , second plot from below . In simulations a minimal value of τ3 around 240ms was needed to stabilize the network . To include the effect of the non-linearity we first used the Lyapunov-based criterion ( see above and S1 Text ) which yielded τ3 ≥ 1200ms . To see if a tighter bound was possible , we applied the Aizerman criterion to the slowest mode . Note that this is strictly only valid for a 3 dimensional system and not the 3N dimensional system . Thus we assume that the eigenmodes of the system do not or only weakly couple . Under this assumption stability is guaranteed when τ 3 a i z = α β 1 - β w m [ τ 1 τ 2 τ 1 + ( 1 - β w m ) τ 2 ] ( 10 ) where β = max x ( g ˜ ( x ) x ) / α . In other words in the Aizerman criterion , the slope at the origin of g ˜ ( x ) is replaced by the slope of the linear envelope . The parameters α and β are extracted from the f-I curve of the unconnected network , Fig 4A , inset . When applied to our simulations , the criterion leads to a value of τ 3 a i z = 380ms , which is not far from the minimal value found numerically . This indeed leads to stable homeostasis , Fig 4C , top . The above results assumed a simple controller with only three components in the feedback loop , r1 , r2 , and r3 , but homeostatic control of excitability has many intermediate stages , for instance synthesis , transport and insertion of ion-channels are likely involved . Therefore we asked how the stability of homeostatic control changes with longer feedback cascades . Our intuition was that adding more elements to the feedback cascade would slow down the feedback , and therefore would increase stability . However , we found that adding more filters actually de-stabilizes the network . We first simplify our model from three to two filters , and analyze what happens to the critical amount of network recurrence if we add a third filter , Fig 5A . With two filters ( τ1 = 10ms , τ2 = 50ms ) the critical recurrence is one , the same as for a network without homeostasis ( gray curve ) . The addition of a third filter , such that the time-constants are ( τ1 , τ2 , τ3 ) = ( 10 , 50 , τ ) is destabilizing even if the third filter has a time-constant slower than any other time-constant ( dashed curve ) . Only for a very long time-constant it had no detrimental effect . Alternatively , one can add an intermediate filter , such that the time-constants are ( τ1 , τ2 , τ3 ) = ( 10 , τ , 50 ) . Also this is destabilizing ( solid line ) . In this case the destabilizing effect can be minimized by taking τ as short as possible . The filter then has a negligible effect , and the system resembles the two filter system again . More generally , assuming that there is no intermediate feedback between the filters and that each element can be approximated by a linear filter , our formalism can be extended to an arbitrary number of intermediate elements in the feedback loop . Suppose that we have K filters , each with its own time-constant τk . The threshold is taken from the K-th filter , i . e . θ ( t ) = rK ( t ) . We thus have for the linearized system τ 1 d r 1 ( t ) d t = - [ 1 - w m ] r 1 ( t ) + u ( t ) - r K ( t ) τ k d r k ( t ) d t = - r k ( t ) + r k - 1 ( t ) k = 2 … K - 1 τ K d r K ( t ) d t = - r g o a l + r K - 1 ( t ) The corresponding characteristic polynomial in this case is 1 + λ τ K ( 1 - w m + λ τ 1 ) ∏ k = 2 K - 1 ( 1 + λ τ k ) = 0 ( 11 ) This expression is invariant to permutations of the time-constants τ2 , … , τK−1 . The stability is again determined by the real part of the solutions to the polynomial . As analytic results such as Routh-Hurwitz analysis , quickly grow in complexity for an increasing number of filters , we solve the polynomial numerically . As the time-constants or even the number of steps in the homeostatic feedback in neurons is not known , we examined the stability with various hypothetical settings of the additional filters , Fig 5B . When the time-constants were set linearly increasing as τi = 10 , 100 , 200 , 300 , …ms , the stability decreased most strongly as the number of stages K increased ( dashed curve ) . Using τi = 10 , 500 , 500 , … , 500 , 5000 , stability decreased also with the number of filters ( dot-dashed curve ) . When the time-constants were set exponentially as τi = 10 , 20 , 40 , 80 … stability decreased when using only few filters , and leveled off with more filters ( black curve ) . With a stronger exponential increase τi = 10 , 30 , 90 , 270 … the stability reached a minimum for 4 filters and then slightly increased to a constant level ( thick black curve ) . Thus in general addition of filters does not lead to stabilization of the system . This result is not dependent on these particular time-constants , also when for instance τ3 , 4 , …K are orders of magnitude slower than τ1 and τ2 , the destabilization occurs . We wondered what choice of time-constants will be most stable for a given number of filters . Suppose a cascade where the time-constant of the firing rate τ1 and of the threshold setting τK are fixed . In analogy with the three filter network , setting the time-constants of the intermediate stages as short as possible is the most stable configuration . Even adding an intermediate filter with a time-constant much slower than τK will not stabilize the system . The intuition behind these results is that not only the speed of the feedback matters , but its phase delay matters as well . With sufficient filtering the negative homeostatic feedback will be out of phase with the firing rate , amplifying perturbations . This effect is similar to the typically destabilizing effect of delays in control theory . Next we use Eq ( 11 ) to study how network recurrence and cascade depth interact . As an example , consider the case where τ1 = 10ms , τ2 = 20ms , and wm = 0 . 99 . If wm increases to 0 . 995 the required τ3 doubles from 4 . 7 to 9 . 7 s . Alternatively , adding an intermediate filter with a time-constant of 50ms also approximately doubles the required time-constant of the integrator to 9 . 5s . When we increase both wm and increase the number of filters , the required τ3 quadruples to 19 . 5s . Thus the effect of recurrence and cascade length are complementary . One can wonder if stability can be rescued in another way . For instance , it is not unreasonable to assume that biology uses multiple , parallel homeostatic regulators . While a general theory of such systems is lacking , some cases can be incorporated in our framework , for instance if multiple feed-backs use the same error signal , stability is determined by the quickest feedback . An addition of a parallel feedback , even if it is slower can only destabilize the system . The stability can be analyzed using the above techniques , adding the extra controller to the feedback-loop . As a technicality , because the system is invariant to the division of labor between the two feedback loops , the stability matrix gains a zero eigenvalue , which can be safely ignored . The system with parallel controllers is always less stable than the system with a single controller , even if the second controller is slower than the first one , Fig 6 .
We have systematically analyzed instabilities in the neural activity that arise from homeostasis of intrinsic excitability . In the worst case , homeostasis can lead to continuous oscillations of the activity . Homeostasis can also give rise to damped oscillations , which are probably less disastrous to information processing , provided the oscillations do not persist too long . To our knowledge such damped oscillations in the homeostatic response have not been observed experimentally , although averaging of experimental data could have obscured their detection . Nevertheless , we think that they are unlikely to occur in biology because substantial cost is involved in alternating up-down regulation of excitability , and because the homeostatic control can strongly interact with the network activity ( Fig 2C ) . Our control theoretic framework for homeostasis sets constraints on homeostatic control to prevent either form of instability and we have focused on three contributions to the stability: recurrent network interactions , depth of the feedback loop , and non-linearities . First , we find that a typical single neuron model with just a few filters in the feedback loop has no stability issues even when the homeostatic control is very fast . However , this is no longer true when network interactions are included . The stronger the recurrence of the network , the slower the feedback needs to be . Networks with time-constants on the order of seconds have been proposed to explain sensory evidence integration , decision making and motor control [32–34] . For homeostasis to be oscillation-free , the minimal homeostatic time-constant scales quadratically with the network time-constant . Thus in particular for networks with slow dynamics , the required homeostasis can become of the order of hours , a value comparable to experimentally observed homeostatic action [13 , 15 , 20 , 21] . Stability typically decreases further when the number of stages in the feedback loop increases , Fig 5 . This effect complements the effect of the recurrence , so that for recurrent networks consisting of neurons with long homeostatic cascades , even slower homeostasis is required . The instability can not be prevented by including heterogeneity or adding noise to the system and is also found in spiking network simulations . The above results are mainly based on the linearized system , which describes stability to small perturbations . In addition , we have derived the condition for stability to arbitrary size perturbations in the case that the f-I curve is non-linear . The non-linear f-I curve limits the minimal homeostatic time-constant even further . Ideally , one would like to know the stability requirements for any given non-linear homeostatic controller . However , only in a very limited number of cases extensions of mathematical results to either multiple non-linearities in the control loop or to higher dimensional systems ( i . e . with longer feedback cascades ) are known . These are topics of current control theory research . Stability of homeostatic control has been the main consideration in this study . This is of course of utmost importance biologically , but it is unlikely to be the only criterion . There can also be cases where rapid acting homeostasis is needed . For instance , one might want to minimize periods of prolonged hyperactivity , while in a recent study fast synaptic homeostasis was required to counter synaptic plasticity [35] . It suggests that homeostatic control is constrained “from below and from above” , and therefore more finely tuned than previously thought . Unfortunately data on the time-course of the homeostasis of intrinsic excitability , its mediators and regulation cascade is limited , hindering a direct comparison of data to our analysis . Nevertheless , a number of predictions follows from this work: we predict homeostasis to be slower in brain regions with strong recurrent connections and long network integration times . Secondly , we predict that intermediate steps in the homeostatic feedback cascade are rapid so as to prevent instability . A recent complementary study examined homeostatic control for a network with separate excitatory and inhibitory populations and a shallow feedback loop ( K = 2 ) and found as the only requirement for stability that the homeostasis of excitatory neurons is at least as fast as that of inhibitory neurons [22] . When excitation and inhibition are subject to equally fast homeostasis , the system is identical to the one studied here . As for these shallow feedback loops the homeostasis is always stable ( our Fig 5A ) , no constraint on the speed of homeostasis relative to the neural and network timescales arises in that study . It should be possible to use our framework to extend those two population results to deeper feedback cascades . Other targets for extension and application of this study include excitatory/inhibitory balanced networks , controllers with parallel slow and fast components , as well as models that include dynamical synapses . Also the interaction with ‘Hebbian’ modification of the intrinsic excitability [36] will be of interest . Finally , these results might be important for other regulatory feedback systems such as synaptic homeostasis and spike frequency adaptation .
In the main text we state that stability of a homeostatic network is determined by the stability of the mode with the largest eigenvalue . Here we prove that if the reduced linear ( 3D ) model based on the largest eigenvalue is stable , then so is the full ( 3N dimensional ) linearized network model . Given the interaction matrix M of the full network , Eq ( 6 ) , it is easy to show that the eigenvectors of the matrix M have the form ( e n α n e n β n e n ) , where en is an eigenvector of the W matrix , and αn and βn are complex numbers . This means that the filtered firing rates ( the vectors r2 and r3 ) follow the firing rates r1 with a phase lag and arbitrary amplitude . We assume that N × N matrix W is symmetric so that it is diagonizable by an orthogonal matrix , that is W = UT DU , where D is a diagonal matrix with the eigenvalues wn on the diagonal and UUT = I . We analyze M in the eigenspace of W using the matrix U3N = U ⊗ I3 , where ⊗ is the Kronecker product . In these coordinates M ¯ = U 3 N M U 3 N T and equals M ¯ = ( 1 τ 1 ( D - I ) 0 - 1 τ 1 I 1 τ 2 I - 1 τ 2 I 0 0 1 τ 3 I 0 ) . In these coordinates , there is no interaction between the various eigenmodes . The stability of each mode is given by Eq ( 9 ) . Because the factor ( 1 − wn ) is positive and minimal for wn = wm , stability of the eigenmode with eigenvalue wm implies stability for all other modes for which wn ≤ wm . The stability condition is found from the Routh–Hurwitz stability criterion [25] . It states that the third order polynomial ∑ i = 0 3 c i λ i = 0 has exclusively negative roots when 1 ) all the coefficients ci are larger than zero , and 2 ) c0 c3 < c1 c2 . Applied to homeostatic control this yields Eq ( 9 ) . The analysis can be extended to networks with non-symmetric weight matrices . Symmetry of W implies that the eigenvalues of the matrix W are real . For non-symmetric W , the eigenvalues are no longer guaranteed to be real but can be complex . The Routh-Hurwitz criterion needs now to be applied after splitting the real and imaginary parts of the polynomial . The conditions that guarantee negative real parts for the solutions of the polynomial λ3 + c1 λ2 + c2 λ + c3 = 0 with complex coefficients ci are [37]: 1 ) ℜ ( c1 ) > 0 , 2 ) ℜ ( c 1 ) ℜ ( c 1 c ‾ 2 − c 3 ) − ℑ ( c 2 ) 2 > 0 , and 3 ) [ ℜ ( c 1 ) ℜ ( c 1 c ‾ 3 ) − ℜ ( c 3 ) 2 ] [ ℜ ( c 1 ) ℜ ( c 1 c ‾ 2 − c 3 ) − ℑ ( c 2 ) 2 ] − [ ℜ ( c 1 ) ℑ ( c ‾ 1 c 3 ) − ℜ ( c 3 ) ℑ ( c 2 ) ] 2 > 0 , where c ‾ denotes the complex conjugate of c , and ℜ and ℑ the real and imaginary parts . In this case one has c1 = 1/τ2 + ( 1 − wn ) /τ1 , c2 = ( 1 − wn ) /τ1 τ2 , c3 = 1/τ1 τ2 τ3 , where wn is the complex eigenvalue . Splitting the real and imaginary parts as wn = wr + iwi , these conditions combine to the condition τ 3 ≥ τ 3 c c with τ 3 c c = 1 1 - w r τ 1 τ 2 [ τ 1 + ( 1 - w r ) τ 2 ] + 1 2 τ 2 3 w i 2 [ 1 + 1 + 4 τ 1 ( 1 - w r ) / ( τ 2 w i 2 ) ] [ τ 1 + ( 1 - w r ) τ 2 ] 2 + w i 2 τ 2 2 ( 12 ) In contrast to the case of symmetric W , these conditions have to be checked for all N eigenvalues of W . By taking the limit of infinite wi it can be shown that stability is guaranteed for any complex wn when τ3 > τ2/ ( 1 − wr ) , which is more stringent than the condition given in Eq ( 9 ) . Under this condition any network , including non-symmetric ones , is guaranteed to be stable to small pertubations . To guarantee an oscillation-free response of the network , the eigenvalues need to be negative and real . For a given wn this implies that all the solutions of the polynomial P ( λ ) = ( 1 - w n + τ 1 λ ) ( 1 + τ 2 λ ) τ 3 λ + 1 have to be real . As in our analysis above , the largest eigenvalue of W is the most critical one so that we only need to study the case wn = wm . The polynomial is negative for large , negative λ and positive for large , positive λ . For all solutions to be real , the polynomial has to dip down after the first zero-crossing and cross zero again , after which it crosses the x-axis a final time . The condition on the minimum of the dip , given by P′ ( λc ) = 0 and P′′ ( λc ) > 0 , is that it should be below zero , i . e . P ( λc ) < 0 . This yields the condition τ 3 ≥ τ 3 c o with τ 3 c o = 1 ( 1 - w m ) 2 ( τ 1 - τ 2 ′ ) 2 [ ( τ 1 - 2 τ 2 ′ ) ( 2 τ 1 - τ 2 ′ ) ( τ 1 + τ 2 ′ ) + 2 ( τ 1 2 - τ 1 τ 2 + τ 2 2 ) 3 / 2 ] ( 13 ) where we defined τ 2 ′ = ( 1 − w m ) τ 2 . In the limit of strong recurrence τ 3 c o = 4 τ 1 ( τ 1 1 − w m ) 2 , which implies that the required time-constant τ3 scales quadratically with the network time-constant , τ1/1 − wm . A population of 16000 linear integrate-and-fire neurons was coupled with a 2% connection probability via excitatory synapses modeled as exponentially decaying conductances ( 5ms synaptic time-constant ) . It is possible to add inhibitory connections to the network , but as long as the network remains in the mean-driven regime this should not affect the results . The membrane voltage of each neuron obeyed τ m e m d V ( t ) d t = − V ( t ) + V r e s t + R I ( t ) , where tmem = 20ms , Vrest = −60mV and R = 1MΩ . In addition , upon reaching the threshold ( Vthr = −50mV ) the voltage reset ( Vreset = Vrest , 5ms refractory period ) . The current I consisted of recurrent input , external drive and homeostatic bias , I ( t ) = ge ( t ) ( V ( t ) − Ee ) +I ( t ) − hr3 ( t ) . The factor h converts the filtered firing rate r3 to a current and sets the strength of the homeostatic control . It was set to 1 pA/Hz . The homeostatic control was implemented as in the rate based networks: the average rate r2 ( t ) was extracted by filtering the spikes ( τ2 = 50ms ) , and this was fed into the integrator . The homeostatic target rate was set to 4Hz . The external current I ( t ) contains both stimulation and a Gaussian white noise term ( σ = 75pA ) to prevent population synchrony . | Despite their apparent robustness many biological system work best in controlled environments , the tightly regulated mammalian body temperature being a good example . Biological homeostatic control systems , not unlike those used in engineering , ensure that the right operating conditions are met . Similarly , neurons appear to adjust the amount of activity they produce to be neither too high nor too low by , among other ways , regulating their excitability . However , for no apparent reason the neural homeostatic processes are very slow , taking hours or even days to regulate the neuron . Here we use results from mathematical control theory to examine under which conditions such slow control is necessary to prevent instabilities that lead to strong , sustained oscillations in the activity . Our results lead to a deeper understanding of neural homeostasis and can help the design of artificial neural systems . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| []
| 2015 | Stability of Neuronal Networks with Homeostatic Regulation |
Residents of resource-poor tropical countries carry heavy burdens of concurrent parasitic infections , leading to high rates of morbidity and mortality . This study was undertaken to help identify the social and environmental determinants of multiple parasite infection in one such community . Residents of Kingwede , Kenya aged 8 years and older were tested for presence and intensity of S . haematobium and Plasmodium spp . infections in a cross-sectional , household-based , community survey . Using General Estimating Equation ( GEE ) models , social and environmental determinants associated with patterns of co-infection were identified , with age being one of the most important factors . Children had 9 . 3 times the odds of co-infection compared to adults ( 95%CI = 5 . 3–16 . 3 ) . Even after controlling for age , socio-economic position , and other correlates of co-infection , intense concomitant infections with the two parasites were found to cluster in a subset of individuals: the odds of heavy vs . light S . haematobium infection increased with increasing Plasmodium infection intensity suggesting the importance of unmeasured biological factors in determining intensity of co-infection . Children in this community are more likely to be infected with multiple parasites than are adults and should therefore be targeted for prevention and control interventions . More importantly , heavy infections with multiple parasite species appear to cluster within a subset of individuals . Further studies focusing on these most vulnerable people are warranted .
Parasitic infections account for a large proportion of the burden of disease in Kenya , with far-reaching effects on the nation's health and economy . Malaria , for example , represents 30–50% of all outpatient visits to health facilities , causing more than 20% of deaths in children less than 5 years old ( yo ) [1] . Neglected tropical diseases such as urinary schistosomiasis are important causes of morbidity in Kenya as well . Urinary schistosomiasis is estimated to affect roughly one-quarter of the Kenyan population leading to anemia , impaired growth , development and cognition [2] and other adverse outcomes . The parasites that cause these diseases are co-endemic in coastal Kenya subjecting the local population to a substantial , possibly synergistic disease burden [3] . Evidence exists that morbidities are likely to be compounded in individuals harboring multiple parasites , as compared to those with single infections . Co-infections with helminths and Plasmodium species , for example , increase various negative health effects , including organomegaly [4] , low birth weight [5] , and anemia [6] , [7] . The morbidities associated with co-infections also are likely to depend on parasite loads [6] , [8] , [9] as seen in single-species Plasmodium infections [10] , [11] and S . haematobium infections [12] . Understanding the complexities of the pathogen-host landscape in settings endemic for multiple human parasites is essential for mitigating morbidities . Identifying interspecies associations could advance intervention by targeting efforts that have the most prevention and treatment benefits . Despite extensive research on these individual infections , surprisingly little is known about the distributions , causes and effects of co-infections ( polyparasitism ) . At the population level , the distribution of polyparasitism depends on risk factors for individual infections and the extent to which these factors are shared across species . Factors known to influence the distribution of single species infection with Plasmodium spp . and S . haematobium include both social and environmental conditions . Some examples include poverty , access to safe water and sanitation , access to health care , water use behaviors and use of bednets [13]–[16] . Socioeconomic and ecologic parameters are sometimes difficult to separate as they often overlap in space . Co-infections may be spatially clustered if households are located near habitats suitable for transmission and if land use is socially and spatially patterned [17]–[19] . Alternatively , clustering of infections may be caused by biological synergism within the host . Evidence for this is mixed , however , with some studies showing increased susceptibility to clinical malaria in persons with helminth infection [20] , [21] , some showing protective effects of helminth infections [22] and others finding no effect [23] , [24] . To improve understanding of causes and effects of polyparasitism , we undertook studies of infection by Plasmodium spp . , and Schistosoma haematobium among members of a community in coastal south-eastern Kenya . The specific research objectives were to describe the prevalence and intensity of single and multiple species infections , identify individual and household-level correlates of single and multiple infections , and determine associations between infection prevalence and intensities after controlling for contextual effects .
For this cross-sectional study , a complete demographical profile of Kingwede , a rural village in coastal Kenya located ∼50 km southwest of Mombasa ( see map , Figure 1 ) , was compiled in December 2005 . Trained interviewers identified every house in the study area and collected information ( sex , age , relationship to household head , and years of residence ) on each person who slept in that house the previous night . Subsequently , every individual ≥8 years old ( yo ) was invited to participate in the study during May through July 2006 . Kingwede residents are predominantly subsistence farmers and fishermen of the Digo tribe . Electricity and access to piped water are virtually nonexistent in the village . Pumped , potable water is available at 8 locations in the village . A seasonal stream and ponds are used for washing clothes , bathing and swimming . The study site is bordered on the northwest by the paved road to Tanzania , on the southeast by the Indian Ocean ( 3 km ) and comprises a ∼6 sq . km area . Participants were asked to provide a midday urine sample on two consecutive days and one finger-prick blood sample . Presence of S . haematobium eggs was determined by microscopic examination of filtered midday urine . Two 10 ml aliquots of each stirred urine sample were filtered using 12-µm pore Nucleopore filters ( Nucleopore , Pleasanton , CA ) and mounted on microscope slides for examination [25] . Positive S . haematobium samples were defined as having ≥1 egg in at least one of the two samples . Finger-prick blood ( ∼200 µL ) that had been collected in potassium ethylene- diaminetetraacetic acid ( K-EDTA ) -coated Vacutainer tubes and stored at −20°C was further evaluated for presence and quantity of circulating parasites at Case Western Reserve University in Cleveland , OH . Plasmodium spp . was detected in these preparations using polymerase chain reaction/ligase detection reaction fluorescent microsphere-based assay ( PCR/LDR-FMA ) as previously described [26]–[28] . Briefly , DNA was extracted , amplified and added to a multiplex , species-specific ligase detection reaction ( LDR ) where species-specific primers hybridized to target sequences and were subsequently labeled with oligonucleotide probes with fluorescent capacity . A Bio-Plex array reader ( Bio-Rad Laboratories , Hercules , CA ) was used for detection of fluorescence in a species-specific manner . Samples were considered positive if they had a median fluorescent intensity ( MFI ) >2 standard deviations above those of negative controls ( MFI above 205 , 260 , or 220 for P . falciparum ( Pf ) , P . malariae ( Pm ) and P . ovale ( Po ) , respectively ) . Of note , no Plasmodium vivax infection was detected in this study . Once biological measures were obtained , trained interviewers conducted questionnaires with all willing participants . Different questionnaires were used for adults and for children . These questionnaires included queries aimed at assessing effect modifiers or confounders for infection status , including socio-economic position ( SEP ) , knowledge , attitudes and practices ( KAP ) regarding malaria and schistosomiasis , mosquito avoidance behaviors , and water contact patterns . Household-level information ( e . g . ownership of assets ) was extracted from questionnaire responses of the household head or most senior household member , and was applied to other household members in analyses . All questionnaire data relied on self-reporting by participants or head-of-household responses . Other household-level information was collected by the interviewers who observed house quality ( e . g . roofing and construction material , presence of window screens , etc . ) . The location of each house entrance ( latitude , longitude ) was measured using a handheld GPS device ( Garmin eTrex Summit , Olathe , KS ) . A 1 km resolution IKONOS image of the study area was obtained from GeoEye ( www . geoeye . com , Dulles , VA ) for use in spatial and environmental analyses . ArcGIS v . 9 . 1 ( ESRI , Redlands , CA , USA ) was used to create spatial variables such as Euclidean distances from each house to water sources . Following standard participant recruitment procedure used by other researchers in this area , prior to the community-wide parasitological survey , awareness of the study aims , risk , and potential benefits was raised through meetings with district and village leaders . These were followed by community-based , village-wide information meetings addressing concerns and questions . A village-wide demographic survey was then conducted to enumerate and code houses , and establish the age , sex , and number of individuals in each household . Written informed consent was obtained from adults and child assent accompanied by written parental consent was obtained for each child for participation in the parasitological survey and before administration of questionnaires . Approval for this study was granted by the University of Michigan Institutional Review Board ( IRB# H05-00008982-I ) and the Ethical Review Committee of the Kenya Medical Research Institute ( KEMRI ) . Intensity of infection was defined as arithmetic mean eggs per 10 mL of the replicate urine filtrations for S . haematobium . For Plasmodium spp . infections , a standardized measure of MFI was created . All values below the aforementioned cut-off values were set to zero . Subsequently , the MFI were divided by the maximum observed value for a range of 0 to 1 . This method was used separately for each Plasmodium species . A summed value was then created to represent burden of any Plasmodium species . The same method was used to create standardized S . haematobium intensity scores . Ordered categorical variables also were created based on the density distribution of each parasite . Intensities of each infection were divided into five , approximately equal categories based on natural cut-points in the distributions ( 0–4; no infection to heaviest ) . Finally , binary variables were created to define heavy infections as follows: S . haematobium ≥100 eggs per 10 mL urine [12] , [29] ( 15% of individuals testing positive for S . haematobium ) ; and Plasmodium spp . ≥0 . 55 standardized score ( selected to include the two most heavily infected categories representing 26 . 9% of Plasmodium-infected individuals ) . All data were double entered in Microsoft Access and analyzed using SAS 9 . 1 ( SAS Institute , Inc . , Cary , NC ) and SPSS ( v . 20 , IBM SPSS , New York , NY ) software . The outcome of co-infection was defined in several different ways for these analyses: A binary variable ( Y/N ) for Plasmodium spp . - S . haematobium co-infection was used to identify important covariates; a multinomial variable ( no infection , S . haematobium only , Plasmodium spp . only , S . haematobium and Plasmodium spp . ) was used to examine associations between parasites . For analyses of infection intensity outcome , variables included heavy Plasmodium spp . infection ( Y/N ) and heavy S . haematobium infection ( Y/N ) . Covariates of primary interest in these analyses were age , sex , knowledge of malaria ( MKAP ) , knowledge of schistosomiasis ( SKAP ) , individual educational attainment , regular income ( Y/N ) , use of bednets ( Y/N ) , night outdoor activities , water contact behaviors , and recent malaria diagnosis . Household-level variables included household socio-economic position ( SEP ) and distance to the local stream . KAP questions ( N = 3 for both malaria and schistosomiasis ) were used to create a scale ( 0–3 ) for each infection based on the number of correct responses . Individual educational attainment was measured by highest class completed . Five categories of educational status were created based on quintiles of distributions . Education was ignored in subsequent analyses due to co-linearity with age in this study population . Bed net use was treated as a binary variable and did not distinguish between treated and untreated nets . A binary variable was created to measure effective treatment of malaria with medication in the past month . Water contact behaviors were assessed by combining responses to questionnaire data such that participants reporting swimming , fishing , bathing , washing dishes or washing clothes in water sources potentially infected with S . haematobium were considered exposed . Household SEP was assessed with an asset index constructed using principal components analysis of questionnaire responses [30] , [31] . Assets measured include electricity , radio , television , bicycle , motor vehicle , land ownership , domestic animals and toilet . The score also included information on crowding , quality of house construction , and numbers of full and part time workers in a household . Missing data necessary for construction of the household SEP score were first imputed using the IVEware ( Imputation and variance estimation software ) SAS macro [32] . Each individual was assigned an SEP score corresponding to their household and this household SEP score was treated as a continuous variable . General estimating equation ( GEE ) models with exchangeable correlation matrix structures and logistic distributions were used to estimate the association between odds of infection , both single and co-infection , and other social and environmental variables . Each infection outcome was assessed separately , thus in models of Plasmodium infection the referent group was individuals without Plasmodium infection regardless of S . haematobium infection status , and vice versa . In models of co-infection the referent group was individuals with single infections or no infections . This analytic approach accounts for household clustering in the data and estimates fixed effects of variables averaged across households . To identify individual and household-level variables associated with polyparasitism , all individual and household-level variables significantly associated with co-infection at α = 0 . 1 were included in multiple logistic GEE models . Although GEE models can be written for multinomial distributions using SAS 9 . 1 , these models only accommodate ordinal outcome variables; thus associations with covariates assume a stepwise relationship between the infection categories . For simplicity in interpreting findings , we employed logistic GEE models for this section of the analyses . Backward elimination , in which the least significant effect as identified by Wald tests is removed from the model stepwise , was then used to select models best predicting the outcomes . Datasets for this analysis were limited to include only participants with complete biological and questionnaire data devoid of any missing responses . Continuous variables were centered on the grand mean . Intensity of infection variables were similarly identified using logistic GEE models . In this case , models were restricted to infected individuals using a binary outcome of heavy versus light infection for each parasite and for co-infection . This was done in order to isolate factors related to the quantity of parasites an individual might carry from those influencing whether or not infection occurs . To identify synergy or antagonism between infections , potential confounders , identified in the logistic GEE models of co-infection previously described , were included in multinomial GEE models . The outcome variable in these models was infection status with four potential values: 0 = uninfected; 1 = only S . haematobium; 2 = only Plasmodium spp . ; 3 = co-infected . Co-infection was set as the referent category . Associations were considered significant at an α = 0 . 05 for the Type 3 likelihood ratio test of the co-infection regression coefficient in the full model , controlling for all other variables . To examine associations between intense infections of the two species , logistic GEE models of heavy single infections included a variable for the intensity of the second infection , treated as an independent , ordered categorical variable ( standardized intensity score ) with 5 categories ( 0 to 4 ) . These analyses were restricted to infected individuals .
A total of 1 , 854 persons 8 years of age and older from 460 households was identified in the study area . Approximately half ( 935 ) of the eligible individuals , representing 310 different households , agreed to participate in the study by consenting to contribute samples and answer questions . The proportions of the eligible population contributing parasitological samples for each infection and those contributing questionnaire data varied ( Figure 2 ) . Complete parasitological and at least partial questionnaire data were received for 766 ( 41 . 3% ) individuals from 252 ( 54 . 8% ) households . By further restricting the dataset to those with no missing questionnaire responses for the variables needed in subsequent analyses 561 persons from 226 households remained . This study population represented the total eligible population in its age distribution but adult males were underrepresented ( 30 . 8% of participating adults were male , 47 . 1% of the eligible adult population was male ) . Non-participants were not otherwise significantly different from participants in terms of age or household size , nor were the participant households significantly clustered within the village ( data not shown ) . When the study population was stratified by age , characteristics of children ( 8 to 17 yo ) differed greatly from those of adults ( Table 1 ) . Adults had higher malaria and schistosomiasis KAP scores than did children , with more adults reporting a regular income , using bednets and going outside at night . Since only a few children reported having a regular income , this variable was not used in subsequent child analyses . Children were more likely than adults to report water contact with potentially infected water sources . Recent consumption of antimalarials was similar between children and adults , as were household SEP scores and distances of households to the local stream . This population experienced high prevalence of infection by both parasites , with over 75% of participating children carrying Plasmodium spp . and over 40% harboring S . haematobium ( Table 1 ) . Co-infection was seen in 31 . 8% of children . In adults , infection prevalences were somewhat lower , with 34% , 14 . 5% and 5% carrying Plasmodium , S . haematobium and co-infection , respectively . Children were 6 . 6 times more likely than adults to be infected with Plasmodium spp . ( 95% C . I . = 4 . 40–9 . 78 ) , 4 . 1 times more likely to have S . haematobium ( 95% C . I . = 2 . 77–6 . 04 ) and 9 . 3 times more likely to carry Plasmodium-S . haematobium co-infections ( 95% C . I . = 5 . 29–16 . 27 ) . Age trends in single and multiple infection prevalences are illustrated in Figure 3 . Correlates of co-infection also differed between children and adults ( Tables 2 and 3 ) . In children , the odds of co-infection with Plasmodium spp . and S . haematobium were lower for those using bednets , engaged in outdoor activities at night , and living farther from the stream , yet higher in those who reported potentially risky water contact . In adults , co-infection odds decreased with age and with increasing household SEP . Among participants infected with Plasmodium parasites , over a quarter carried heavy infections ( Table 1 ) . Fifteen percent of participants with S . haematobium infections were heavily infected ( 100 eggs/10 mL ) . As with infection prevalence , the intensity of infection was much heavier in children than in adults ( Plasmodium spp . ( OR = 8 . 09 , 95% C . I . = 4 . 41–14 . 81 ) ; S . haematobium OR = 3 . 11 , 95% C . I . = 1 . 31–7 . 39 ) . After restricting analyses to infected individuals , children were still more likely than adults to harbor heavy Plasmodium spp . infections ( OR = 3 . 79 , 95% C . I . = 1 . 97–7 . 29 ) whereas this was not the case for S . haematobium ( Table 1 ) . Simple regression analyses indicated that odds of Plasmodium spp . infection were higher in individuals infected with S . haematobium compared to those without ( OR = 1 . 97 , 95% C . I . = 1 . 34–2 . 89 ) . Multivariable , multinomial logistic models confirmed this association even after adjustment for shared contextual variables among children , but this association was not seen in adults ( Table 4 ) . Children were less likely to carry single S . haematobium infections than to be co-infected with both parasites ( OR = 0 . 60 , 95% C . I . = 0 . 39–0 . 93 ) . Similarly , the significant association between the intensity of Plasmodium spp . infection and S . haematobium infection observed in simple regression analyses ( OR = 3 . 15 , 95% C . I . = 1 . 26–7 . 89 ) persisted after adjustment for confounders in children; as noted in Table 5 , children with heavy Plasmodium infection were more likely than those with light infection to have concomitant heavy S . haematobium infection , controlling for relevant covariates ( OR = 1 . 39 , 95% C . I . = 0 . 99–1 . 97 ) . No evidence of association in odds or intensities of these infections was seen in adults ( Tables 4 and 5 ) .
Results of this study reveal several key findings . Participating Kingwede residents had high prevalence of infection with S . haematobium and Plasmodium spp . parasites . Co-infection was also common and was associated with both individual- and household-level factors including young age and low socio-economic position . Interestingly , co-infection was more common than single S . haematobium infection in children , after controlling for other variables . Similarly , intense infections with both parasites appeared to cluster in a subset of children . While non-participation may have biased some of estimates of association , we believe that because individual per-household participation was reasonably good ( 62% ) , our investigation illustrates the importance of context in studies of infectious diseases . We found that contextual factors ( household SEP ) contributed significantly to variation in an individual's parasite infection profile , including , for children , proximity to the main local water source . These findings compare with other studies of single-species infection studies in which risk of malaria and schistosomiasis was higher in households proximate to water bodies [33]–[35] and are in accord with those that associate these infections with lower SEP [36]–[39] . Our results also support recent efforts to reconsider broader social and ecological contexts in studies of infectious diseases in lieu of traditional biomedical models [40] . Finally , the identification of risk factors common to infection by both parasites argues for a broader , integrated approach to control and prevention [41] and suggests potential targets for these efforts . Aside from contextual variables , age was found to be the most important shared risk factor of infection , with both higher prevalence and higher intensity of single and concomitant infection in school-aged children as compared to adults . The observed negative associations between age and co-infection are consistent with other published reports from endemic regions on clinical malaria [42] and S . haematobium infection [43] . Higher prevalence at younger ages is likely due to immunological naïveté governing susceptibility to both infections [44]–[46] , a variable which was not measured in this study . The strength and consistency of the age effect on infection led to stratification of all analyses , with separate models for children and for adults . Our results support current trends in disease prevention and treatment efforts focusing on school-aged children . A key caveat may be that the many asymptomatic Plasmodium spp . infections in adults could be an important source for continued transmission to children and should therefore not be ignored in intervention efforts . Age-stratified , multivariable logistic GEE analyses of Plasmodium spp . and S . haematobium prevalence revealed no evidence of positive association between two parasites for the combined presence of infection ( measured as we did , by PCR and urine filtration ) after adjustment for relevant SEP , behavioral and environmental variables ( Tables 2 and 3 ) . The apparent lack of association between these two parasite species using prevalence data has been supported by previous research [47] and could be explained by dissimilar risk factors , distributions , and modes of transmission for the two infections . Comparability of our results with those of other malaria-helminth polyparasitism studies is limited by differences in outcome measures ( asymptomatic infection vs . clinical disease ) [48]–[50] and helminth species of interest ( most other co-infection studies investigate a variety of soil-transmitted helminth ( STH ) species rather than S . haematobium ) [reviewed in 41] . However , our cross-sectional design may have limited our estimation of continuing infection risk , particularly for malaria , given annual weather related fluctuations in exposure . In analyses of infection intensities , positive associations were observed between the intensities of the two parasites among co-infected children ( Tables 5 and 6 ) , thus highlighting the importance of host factors in determining parasite loads , and also suggesting a synergistic relationship . Although some studies have reported antagonistic relationships between these parasites ( i . e . lower P . falciparum intensities in children harboring light S . haematobium infections vis-à-vis those not co-infected [51] , [52] ) , many other investigations support our findings ( reviewed in [53] ) . The potential of a synergistic relationship between these parasites is important from a clinical perspective , as well as epidemiologically . Clinically , the morbidity associated with co-infection is likely to depend on parasite loads [6] , [8] , [9] as seen in single-species Plasmodium infections [10] , [11] and S . haematobium infections [12] . If heavy infections cluster in a subset of individuals , as suggested by our results , identification and targeting of this subset for multi-disease prevention and treatment interventions could be highly cost effective in that it would serve to reduce overall pathogen burdens . We hypothesize that immunological parameters , unmeasured in this study , are partially responsible for the observed positive association between parasite intensities . In that case , an improved understanding of the complex immunological milieu governing infection profiles would be needed to evaluate this possibility . Failure to differentiate between co-infection and co-morbidity outcomes has confounded discussion of results from other similar studies . To illustrate , data suggest that P . falciparum may protect against S . haematobium infection by promoting protective antibody development , but Plasmodium infection can also increase inflammatory factors associated with morbidity [54] , [55] . It is also possible that seasonal or climatic variations or other unmeasured variables influenced our results . This study was cross-sectional , thus natural temporal variations in infection prevalence and intensity were not considered . Due to limited resources , we did not control for other infections such as STH or HIV that could affect susceptibility to , or intensity of Plasmodium spp . or S . haematobium infections . Concomitant infections also could have effects on host immune response , thereby influencing S . haematobium-Plasmodium spp . associations . Although STH are likely to be ubiquitous , S . mansoni is not endemic to this region and HIV rates are fairly low ( 7% prevalence in rural Kenya and 7 . 9% prevalence in Coast Province overall [56] ) , suggesting that confounding likely was minimal . Genetic factors and nutritional status represent other potentially important unmeasured confounders . Our use of marginal models should have accounted for or attenuated any genetic effects shared within families by assessing clustering of individuals within households . Furthermore , another study of S . haematobium from the same geographic region concluded that heritability in host susceptibility is low and unlikely to play a major role in determining individual risk for infection [57] . Potential confounding effects of nutritional status should not be large assuming similar access to nutrition within households . This also should have been partially accounted for by the inclusion of household SEP variables in analyses . Of particular note , in our study , outdoor night-time activity was associated with lower odds of co-infection or malaria infection among children , and with reduced odds of heavy Plasmodium infection in adults . This finding may be related to the endophilic feeding preference of anophelines within houses , with reduced risk among those who are out of the house for part of the night . Future research on the epidemiology of polyparasitism could benefit from the inclusion of several key parameters not fully evaluated in this study: i ) A wider range of spatial and temporal scales and study sites could reduce potential chance associations and improve understanding of the climatic , seasonal and environmental factors that influence parasite distributions and interactions; ii ) Malaria investigations would greatly benefit from more sensitive , standardized diagnostic tools for quantification of species-specific Plasmodium parasites . Although work of other researchers has shown good correlation between median fluorescent intensity ( MFI ) values and parasitemia , the relationship between the two is not linear toward the upper and lower limits [28] , [58] , [59] . Validation of our use of MFI as a quantitative measure is needed; iii ) finally , in order to fully understand the implications of observed associations between parasitic species in human hosts , a better understanding is needed of relevant immunological mechanisms [41] , [53] , [60] . However , the need for improved scientific knowledge about the biology , and epidemiology of polyparasitism should not take precedence over what is already known; parasites cause significant morbidity and more accessible , effective treatment and prevention is urgently required . | Parasitic diseases such as malaria and schistosomiasis account for a large proportion of global morbidity and mortality by contributing to malnutrition , developmental delays , loss of productivity , negative birth outcomes , disfigurement , physical handicap and social stigma . We studied these two infections in a community-wide , household-based study among residents of a coastal Kenyan village . With the use of newer PCR methods to detect sub-clinical malaria infections , and with the aid of statistical methods that could adjust for the household-level clustering of exposure factors , we found evidence that people of low socio-economic position are more likely to experience co-infection by these parasites . In addition , among children , we found that heavy infections clustered within a subset of individuals independent of their water contact , night activity , bednets use , and household distance to water , thus providing indirect evidence for biological interaction between these parasites in human hosts . If confirmed , these findings indicate that interventions focused on prevention and control of such co-infections could be concentrated on the most vulnerable groups of people in order to maximize benefits where resources are limited . Further studies of immunological and behavioral risk are warranted . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"social",
"epidemiology",
"infectious",
"diseases",
"schistosomiasis",
"public",
"health",
"and",
"epidemiology",
"tropical",
"diseases",
"(non-neglected)",
"epidemiology",
"environmental",
"epidemiology",
"infectious",
"disease",
"epidemiology",
"neglected",
"tropical",
"diseases",
"spatial",
"epidemiology",
"economic",
"epidemiology",
"malaria"
]
| 2012 | Partnering Parasites: Evidence of Synergism between Heavy Schistosoma haematobium and Plasmodium Species Infections in Kenyan Children |
Mycobacterium tuberculosis infection presents across a spectrum in humans , from latent infection to active tuberculosis . Among those with latent tuberculosis , it is now recognized that there is also a spectrum of infection and this likely contributes to the variable risk of reactivation tuberculosis . Here , functional imaging with 18F-fluorodeoxygluose positron emission tomography and computed tomography ( PET CT ) of cynomolgus macaques with latent M . tuberculosis infection was used to characterize the features of reactivation after tumor necrosis factor ( TNF ) neutralization and determine which imaging characteristics before TNF neutralization distinguish reactivation risk . PET CT was performed on latently infected macaques ( n = 26 ) before and during the course of TNF neutralization and a separate set of latently infected controls ( n = 25 ) . Reactivation occurred in 50% of the latently infected animals receiving TNF neutralizing antibody defined as development of at least one new granuloma in adjacent or distant locations including extrapulmonary sites . Increased lung inflammation measured by PET and the presence of extrapulmonary involvement before TNF neutralization predicted reactivation with 92% sensitivity and specificity . To define the biologic features associated with risk of reactivation , we used these PET CT parameters to identify latently infected animals at high risk for reactivation . High risk animals had higher cumulative lung bacterial burden and higher maximum lesional bacterial burdens , and more T cells producing IL-2 , IL-10 and IL-17 in lung granulomas as compared to low risk macaques . In total , these data support that risk of reactivation is associated with lung inflammation and higher bacterial burden in macaques with latent Mtb infection .
The vast majority of people infected with Mycobacterium tuberculosis ( Mtb ) develop asymptomatic , latent infection ( LTBI ) . It is increasingly recognized that there is a spectrum of LTBI in humans , and this spectrum may correlate with the risk of reactivation [1] . Although reactivation risk is estimated at 10% per lifetime in HIV-negative LTBI humans , this is a population level estimate . Instead , it seems more likely that a small percentage of those with LTBI are at higher risk of reactivation . However , it has been challenging to identify the small fraction of the more than 2 billion latently infected humans who are at greatest risk of reactivation , so that therapy can be targeted to that population . As in humans , LTBI in macaques is a stable , asymptomatic infection without clinical signs [2] . Reactivation of LTBI can be triggered in macaques by immune suppression due to SIV infection , TNF neutralization and CD4 depletion [3–6] , but variable rates of reactivation are observed , similar to humans . We hypothesize that the spectrum of LTBI is associated with susceptibility to reactivation [1 , 2] . Here we develop criteria based on 18F-fluorodeoxyglucose ( FDG ) positron emission tomography coupled with computed tomography ( PET CT ) imaging of macaques with LTBI to predict reactivation risk due to TNF neutralization . These criteria were then applied to latently infected macaques ( without TNF neutralization ) to identify biologic features that correlate with higher risk of reactivation . Macaques at high reactivation risk had greater cumulative lung bacterial burden , higher bacterial burden within an individual granuloma , more Mtb-infected mediastinal lymph nodes , and more T cells producing IL-2 , IL-10 and IL-17 in lung granulomas compared to low risk macaques . Our results support the model of a spectrum of latency , suggesting that the extent and quality of bacterial control as well as lung inflammation in latency determines risk of reactivation after TNF neutralization .
We have previously published criteria for determining whether cynomolgus macaques with M . tuberculosis infection are “active” or “latent” by 6 months post-infection , based on clinical and microbiologic tests , as in humans [2 , 7] . These clinical classifications were confirmed at necropsy , where those classified as active TB had significantly more pathology and bacterial burden than those classified as latent [2] . In this study , our aim was to determine whether we could identify latently infected macaques that would be more susceptible to reactivation . To do this , we employed serial FDG PET CT imaging , prior to and during neutralization of TNF , which we have shown previously can induce reactivation in macaques [5] . A cohort of cynomolgus macaques with LTBI ( n = 26 ) was PET CT imaged at least 6 months post-infection , immediately prior to being randomly assigned to receive either TNF neutralizing antibody for 5–8 weeks or no treatment . Each macaque was evaluated for reactivation which was strictly defined here as dissemination , determined by the appearance of at least one new granuloma in lungs or extrapulmonary sites by PET CT during anti-TNF antibody treatment ( Fig 1 ) . Of 26 animals with TNF neutralization , 50% ( n = 13 ) developed new lesions . At necropsy , macaques that developed new granulomas during TNF neutralization had greater disease pathology and higher total bacterial burden in lungs ( Fig 2A and 2B ) as well as within individual granulomas and lymph nodes ( S1 Fig ) . Animals that developed reactivation had a significantly smaller proportion of sterile ( or greater proportion with Mtb growth ) among granulomas and mediastinal lymph nodes compared to animals that did not reactivate ( Fig 2C and 2D ) . Thus , these data support the use of dissemination , the formation of new lesions in lungs or extrapulmonary sites , as a primary metric of reactivation . While we defined reactivation in terms of bacterial dissemination , we postulated that we might also see evidence of loss of bacterial control among pre-existing lesions from the animals that reactivated . We examined lesion specific changes in metabolic activity ( FDG avidity of each granuloma by PET , reported as standard uptake value , SUV ) and/or size ( by CT ) during TNF neutralization . Granulomas were classified as “stable” if they remained similar in SUV ( change < 5 units ) or size ( change < 1mm ) or “dynamic” if they increased in SUV ( ≥ 5 units ) or size ( ≥1mm ) ( S2 Fig ) . At least 1 dynamic lesion was observed during TNF neutralization in 69% ( 9 of 13 ) of reactivated monkeys compared to only 31% ( 4 of 13 ) among non-reactivated animals . Dynamic lesions were less likely to be sterile and had significantly higher bacterial burdens ( measured as colony forming units , CFU ) compared to stable lesions ( Fig 2E ) among all TNF-neutralized animals . The frequency of sterile lesions among new , stable and dynamic lesions was statistically different ( Fig 2E ) with the lower proportion of sterile lesions among the dynamic and newly developed granulomas . The total number of granulomas per monkey among reactivated ( median = 12 , IQR25 = 8 , IQR75 = 24 . 5 ) and non-reactivated ( median = 8 , IQR25 = 4 . 5 , IQR75 = 19 . 5 ) animals was similar ( Mann-Whitney , p = 0 . 3 ) . These data suggest that the increased bacterial burden observed in reactivation is not solely driven by the number of new lung granulomas but likely a combination of granuloma types and MLN burden . We also compared the ratio of live Mtb CFU to chromosomal equivalents ( CEQ ) ( the cumulative burden of live and dead Mtb ) to estimate bacterial killing [8] in dynamic and stable lesions . Dynamic granulomas had higher CFU/CEQ ratios ( i . e . , less bacterial killing ) than stable granulomas among all animals undergoing TNF neutralization ( Fig 2F ) . Granulomas from reactivated animals had higher CFU/CEQ ratios compared to non-reactivated animals and LTBI controls ( that did not receive TNF antibody ) ( S3 Fig ) . Importantly , however , many lesions in reactivated animals did not increase in metabolic activity or size or display reduced killing after TNF neutralization . This supports our previous data that demonstrates marked heterogeneity of lesions within an individual animal [8–11] . We next sought to identify PET CT characteristics of macaques prior to TNF neutralization that are predictive of reactivation risk . The number of lung granulomas observed before TNF neutralization was similar between animals that reactivated and those that did not ( Fig 3A ) . We have previously shown that overall lung inflammation ( total lung FDG avidity ) detected by PET is loosely associated with lung bacterial burden in macaques with active TB , and decreased dramatically with anti-TB drug treatment [10 , 12] . In this study , total lung FDG avidity immediately prior to anti-TNF treatment was significantly higher in animals that would later reactivate ( Fig 3B ) compared to those that did not . We then sought to define the distinguishing characteristics of individual lung granulomas prior to anti-TNF treatment that correlated with reactivation risk . Animals that would develop reactivation had a higher proportion of FDG avid ( defined as SUV ≥ 5 ) granulomas ( 67 . 2% ) compared to non-reactivator animals ( 30 . 8% ) ( Fisher’s Exact , p<0 . 0001 ) before TNF neutralization . Comparing the single most FDG avid or the largest size granuloma in each animal prior to anti-TNF treatment showed that those that would develop reactivation had a granuloma that was larger or had higher FDG avidity ( Fig 3C and 3D ) . Although dynamic lesions had higher FDG avidity ( median SUV 8 . 8 , IQR 25–75: 5 . 0 , 13 . 9 ) compared to stable lesions ( median SUV 4 . 5 , IQR25-75: 2 . 9 , 7 . 5 , Mann-Whitney , p<0 . 0001 ) before TNF neutralization , we were unable to predict which lesions would become dynamic as only 23 . 7% of the lesions with greatest SUV and 31 . 6% of lesions of maximum size were dynamic lesions after TNF neutralization . FDG avid mediastinal lymph nodes ( MLN ) seen on PET CT are usually associated with Mtb involvement . Previous data from our lab and from human studies suggest that infected MLN are a potential source of reactivation [5 , 6 , 13] . The cumulative FDG avidity of hot MLNs was significantly higher among animals that would later reactivate ( reactivators cumulative SUV median = 9 . 6 , IQR25-75: 5 . 8 , 31 . 5 vs . non-reactivators cumulative SUV median = 5 . 5 , IQR25-75: 0 , 13 . 9 , Mann-Whitney , p = 0 . 04 ) . Lastly , the number of extrapulmonary sites of infection before TNF neutralization was significantly higher in animals that would later develop reactivation compared to those that would not ( Fig 3E ) . To assess which PET CT variables could distinguish reactivators from non-reactivators before TNF neutralization , we ran several different simple logistic regression models ( and a contingency analysis ) to narrow down the best predictors using the following variables: total lung FDG activity , number of “hot” lymph nodes , total number of lymph nodes , total SUV of lymph nodes , number of lesions , and the presence of extrapulmonary lesions . We chose total lung FDG activity and the presence of extrapulmonary lesions as best predictors based on goodness of fit . We then used recursive partitioning , splitting the data into a decision tree to define an optimal cut-off ( 947 . 2 SUV ) for total lung FDG activity . Combining the presence of extrapulmonary involvement and total lung FDG avidity resulted in favorable receiver operator curve results ( area under the curve = 0 . 94 ) and a high sensitivity ( 92 . 3% ) and specificity ( 92 . 3% ) ( Fig 3F ) . It was not feasible to administer TNF neutralizing antibody to another large set of latently infected macaques to validate our predictions or identify biologic features associated with reactivation risk . Therefore we leveraged a set of latent control macaques ( N = 25 ) that were necropsied concurrently with the TNF neutralized group , categorizing them being as at high or low risk for reactivation based on our PET CT defined parameters ( i . e . , total lung FDG avidity and extrapulmonary involvement ) using the scan prior to necropsy . Animals who had evidence of extrapulmonary involvement on scan or a total lung FDG avidity of greater than or equal to 947 . 2 were classified as high risk . We then analyzed lesions from these animals to investigate the bacterial and immunological factors associated with risk of reactivation . While the median CFU per granuloma was the same between high and low risk animals ( S4 Fig ) , a wide range of CFU per granuloma was observed within each individual monkey ( S5 Fig ) . Because low risk animals appeared to have a lower peak bacterial burden compared to high risk , we then compared the maximum CFU per single granuloma within an in individual monkey to limit the bias . Interestingly , the maximum CFU per single granuloma within an animal was also higher among high risk compared to low risk animals ( Fig 4A , S6 Fig ) . The total lung CFU ( cumulative CFU of all lesions in the lung ) was also greater among high-risk LTBI animals ( Fig 4B , S6B Fig ) . High-risk LTBI animals also had a trend toward greater CFU per MLN ( Fig 4C ) , which was surprising given the large range of CFU per MLN on an individual animal level ( S7 Fig ) . High risk animals also had a smaller proportion of sterile MLN ( i . e . , greater proportion of MLN with Mtb growth ) compared to low risk animals ( Fig 4D ) . Together these data indicate that individual lesional characteristics ( i . e . , high bacterial burden in one granuloma or lymph node ) are associated with high risk of reactivation . We then examined Mtb-specific T cell cytokine production within individual lung granulomas and blood of high- ( n = 10 ) and low-risk ( n = 10 ) LTBI control animals ( without TNF neutralization ) . Granulomas from high-risk animals had higher frequencies of IL-17 , IL-10 or IL-2 producing CD3+ T cells as compared to granulomas from low risk animals ( Fig 5 , S9 Fig ) . While most T cells from granulomas were single cytokine producers , as we previously described [9] , we found that high-risk animals had a higher percentage of granuloma T cells producing more than one cytokine . There were no differences in frequencies of Mtb specific cytokine production by CD4+ and CD8+ T cells or by memory subsets in peripheral blood ( S9 , S10 and S11 Figs ) . These data suggest that granulomas from high-risk LTBI animals are more immune stimulated , possibly due to higher bacterial activity , although the specific factors driving this are unknown .
In this study , we strictly defined reactivation following TNF neutralization based on dissemination ( formation of at least one new granuloma ) and validated that definition at necropsy , where reactivated macaques had higher bacterial burden than those that did not reactivate . Together the data in this study support the hypothesis that the spectrum of latency has implications for risk of reactivation . Here we provide evidence that lung inflammation and/or evidence of extrapulmonary involvement detected by PET CT is associated with reactivation risk following TNF neutralization in macaques . In addition , reactivation risk is correlated with at least one granuloma of larger size or higher inflammation ( measured by PET ) . This unique set of data provided the opportunity to investigate individual granulomas in macaques predicted to be at high or low risk of reactivation . Using the parameters we developed based on PET CT characteristics of LTBI macaques that did or did not reactivate following TNF neutralization , we predicted the risk of reactivation of 25 latent control macaques . Neutralizing TNF would change the bacterial and immunologic features in these animals , which would prevent investigation of these factors in risk of reactivation . Therefore , instead of testing our prediction by treating the animals with anti-TNF treatment , we compared several factors in our predicted high or low risk animals without further intervention . Animals predicted at high risk had higher total lung and lymph node bacterial burden . In addition , an individual granuloma in high risk animals had a high bacterial burden , suggesting that a single poorly contained granuloma can contribute to reactivation . T cell cytokine production in granulomas was higher in high risk compared to low risk macaques in the absence of TNF neutralization . This could be due to more bacterial replication and antigen production in high risk animals . Alternatively , the combination of cytokines in certain granulomas from high risk animals may provide a less stable host immune environment . Further studies of immune responses in granulomas from low or high risk animals is necessary to differentiate cause and effect of cytokine responses in predisposing animals to reactivation . The data from this study suggests that only one or a few granulomas need to fail in bacterial containment to lead to dissemination and reactivation . Not all granulomas were equally affected by TNF neutralization , suggesting that reactivation and dissemination can occur from as few as one unstable granuloma . For example , TNF neutralization led to dynamic granulomas ( increasing in size or FDG avidity ) , but this was restricted to a subset of granulomas in reactivating monkeys . This is consistent with the independent and dynamic nature of lung granulomas in this model [8–11] . However , given the lack of current technology for tracking individual bacilli , it is not possible to confirm that dynamic lesions , or those with higher bacterial burden , are the source of dissemination . A limitation of this study is that we are unable at this time to identify direct causes of increased bacterial burden or instability of lesions , which are both associated with reactivation . The ability of mediastinal lymph nodes ( MLN ) to control infection during clinical latency also appears to contribute to risk of reactivation . In LTBI , a Ghon complex refers to the combination of a lung granuloma and an involved lymph node , suggesting lymph node involvement is common in humans . Studies from the pre-antibiotic era also demonstrate substantial lymph node involvement in humans soon after infection [14] . Similarly , mediastinal lymph nodes have been detected by PET CT in humans with Mtb infection [15–18] . Thus , the role of lymph nodes in susceptibility to reactivation is likely important [19] and should be targeted in the development of drugs to treat latent infection . We previously published an association between extent of CD4 depletion in MLN and reactivation in latently infected macaques treated with CD4-depleting antibody [6] . In this current study , the latent controls predicted to be at high risk of reactivation a greater proportion of MLNs with Mtb growth compared to those that were at low risk . However , in a previous study , macaques vaccinated with BCG plus the protein fusion H56 vaccine were protected against TNF-neutralization induced reactivation [20] . Examination of our data from that study shows that protection against reactivation was associated with significantly fewer Mtb positive MLN ( S12 Fig ) . Thus , it is likely that the MLN play an important role in reactivation risk . Even less is known about the presence of extrapulmonary sites of infection during LTBI . While it occurs in humans [21 , 22] , the actual prevalence has not been well described . We speculate that the events that result in extrapulmonary infection being established are due to poor initial control and early dissemination , which is then associated with reactivation risk . In summary , we have provided evidence that the spectrum of clinically defined LTBI , specifically that associated with inflammation detected by PET CT and the presence of extrapulmonary disease , is associated with reactivation risk . It is likely that this occurs in humans with LTBI and similar lung lesions have been described in LTBI humans by PET CT [16 , 23–25] . Importantly , this is the first assay that can functionally distinguish those at high and low risk for reactivation induced by TNF neutralization . We recognize that using PET CT to stratify reactivation risk is not feasible in most human settings . However , the use of PET CT in this well characterized animal model provides an opportunity to identify potential biomarkers in blood , including transcriptional signatures , which may correlate with reactivation risk . Prioritizing treatment to those patients at increased risk of reactivation ( especially those with HIV infection ) is a more efficient strategy in our current efforts to eradicate TB , as most programs are unable to provide treatment to all LTBI patients .
Adult ( ≥ 4 years of age ) cynomolgus macaques ( Macacca fasicularis ) ( Valley Biosystems , West Sacramento , CA ) were screened with standard tests for co-morbidities prior to infection with Mtb as previously published[26] . Animals were maintained in a Biosafety Level 3 facility for primates after M . tuberculosis infection . Cynomolgus macaques were infected with low dose ( ~25 CFU ) of M . tuberculosis ( Erdman strain ) via bronchoscopic instillation into a lower lung lobe and subsequent serial clinical , microbiologic and immunologic parameters were followed until outcome was determined as previously described [2 , 7] . Once animals were classified as latent , a subset was randomized to receive TNF neutralizing agent Adalimumb ( Humira , Abbott Labs , Abbott Park , IL ) at 4 mg/kg/dose subcutaneously every 7–10 days [5] . In general , PET-CT scans were performed at baseline before TNF neutralization and every 2 weeks after treatment until 5–8 weeks . A pre-necropsy scan was performed on all animals to facilitate harvesting scan-identified lesions . Serial analysis of these lesions before and during treatment was performed ( see below ) . At necropsy , animals were maximally bled and gross pathology was assessed using our previously published quantitative scoring system in which a number is given for the size , number and pattern of granulomas in each lung lobe , mediastinal lymph node and extrapulmonary sites ( e . g . , liver , spleen ) . Harvested lung granulomas are characterized , measured and processed into single cell suspension for bacterial burden and flow cytometry as previously reported [2 , 9] . All animal protocols and procedures were approved by the University of Pittsburgh’s Institutional Animal Care and Use Committee ( protocol assurance number A3187-01 . ) Our specific protocol approval numbers for this project are 0808244 , 0906877 , 1011342 , 1105870 and 11080037 . The IACUC adheres to national guidelines established in the Animal Welfare Act ( 7 U . S . C . Sections 2131–2159 ) and the Guide for the Care and Use of Laboratory Animals ( 8th Edition ) as mandated by the U . S . Public Health Service Policy . At predetermined time points , animals were sedated , intubated and imaged by 2-deoxy-2-18F-D-deoxyglucose ( FDG ) PET ( microPET Focus 220 preclinical PET scanner , Siemens Molecular Solutions , Knoxville , TN ) and CT ( Neurologica Corp , Danvers , MA ) imaging within our biosafety level 3 facility as previously described [10 , 12 , 27] . Lesions were identified by two analysts ( M . T . C . and P . M . ) and size was measured by CT . FDG avidity was measured by drawing a region of interest ( ROI ) in the axial view and SUVs ( standard uptake volume , SUV = counts/ ( injected activity/body weight ) , normalized to muscle to reduce variability between scans , were calculated using OsiriX ( Pixmeo , Geneva , Switzerland ) as previously published [27] . The total lung FDG avidity was analyzed and calculated using Osirix viewer , an open-source PACS workstation and DICOM viewer . The whole lung was segmented on CT by using the Growing region algorithm on the OsiriX viewer to create a ROI of normal lung ( Hounsfield units between -1024 and -200 ) . The closing tool was used to include individual nodules and other pulmonary disease . The ROI was transferred to the co-registered PET scan and manually edited to ensure all pulmonary disease was included . All extrapulmonary structures and disease , including mediastinal lymph nodes , were excluded . Voxels outside the ROI were set to zero and voxels within the ROI with an SUV higher than normal lung ( SUV ≥ 2 . 3 ) were isolated . These ROIs ( capturing all SUV ≥ 2 . 3 within the lung ) were exported into a spreadsheet using the OsiriX “Export ROI” plugin . Finally , the sum from the pixels of each slice ( from the exported ROI ) was calculated to represent the measurement of “Total Lung FDG Avidity” . Granuloma specific changes observed before and during TNF neutralization were performed . “Stable” lesions were defined as lesions that maintained the same size and FDG avidity before and after TNF neutralization whereas “dynamic” lesions were those that that increased in size by at least 1mm or FDG avidity by at least 5 SUV . “New” lesions were not present baseline but appeared during the course of TNF neutralization . Following necropsy , tissue sample homogenates were stored in PBS at -80°C . At processing , M . tuberculosis genomes were extracted with phenol-chloroform as previously described [8] . In brief , blinded samples were re-suspended in 1 mL Tris-EDTA buffer , pH 8 . 0 , with 300 μl of 70°C UltraPure phenol:chloroform:isoamyl alcohol ( 25:24:1 ) ( Invitrogen ) and 250 μl of 0 . 1 mm zirconia-silica beads ( BioSpec Products , Inc . ) . Tubes were mixed by inversion , incubated for 10 min , and twice vortexed for 4 min with a 1 min break at highest speed using a 24-tube vortex adaptor ( MO BIO Laboratories , Inc . ) . Following a 10 min centrifugation at 14 , 000 RPM at 4°C , the aqueous layer of each sample was extracted and placed in a fresh tube with 50 μl of 5M sodium chloride . The phenol:chloroform:isoamyl alcohol extraction was repeated with 250 μl and a 30 min incubation at room temperature . After the incubation , samples were centrifuged once more at 14 , 000 RPM for 30 min at 4°C to separate off the aqueous phase . One volume of isopropanol and 1/10 volume of 3M sodium acetate ( pH 5 . 2 ) was then added to each extraction to precipitate genomic DNA with an overnight incubation at -20°C . Each DNA pellet was washed with 70% ethanol and centrifuged for 30 min as before . Each pellet was then left to air-dry to remove excess ethanol and subsequently re-suspended in sterile nuclease free water ( Ambion ) . DNA purity and concentration was measured using the Spectramax 190 spectrophotometer ( Molecular Devices ) . Quantification of chromosomal equivalents ( CEQ ) of M . tuberculosis was performed using real-time PCR of a single copy gene , Mtb sigF , with a previously described primer-probe combination [8] . The primer and probes for this target were purchased together in a pre-mixed PrimeTime qPCR assay ( Integrated DNA Technologies ) . The sequences are as follows , 5’3’: probe–FAM-TCG GAC TTC GTC TCC TTC-Iowa Black , sigF Fwd–GCG GGT CGG GCT GGT CAA C , and sigF Rvs–CCT CGC CCA TGA TGG TAG GAA C . Real-time PCR was preformed in duplicate on the iQ5 Multicolor Real-Time PCR Detection System ( Bio-Rad Laboratories , Inc . ) and the 384well-capable 7900HT Fast Real Time PCR System ( Applied Biosystems ) with TaqMan Universal Master Mix II ( Life Technologies ) . Precise determination of CEQ was derived from a standard curve of serially diluted M . tuberculosis genomic DNA prepared from liquid culture for each qPCR run . Real time PCR efficiency for each run was maintained between 90% and 110% . While the quantification of both live Mtb and chromosomal equivalents are estimates , the ratios reflect a relative estimate of bacterial killing as published [8] . Accuracy of these estimates both in the detection of live bacteria and chromosomal equivalents may be limited by potential clumping of the bacteria resulting in underestimates of CFUs and minor variations in PCR amplification based on sample-specific differences in protein contamination or PCR inhibitors from blood . Intracellular cytokine analyses were performed on individual granulomas harvested at necropsy and on PBMC at predetermine time points ( 6 months post infection ) . As previously described [9] single cell suspension of individual granulomas or PBMC were stimulated ex vivo with peptide pools of Mtb specific antigens ESAT-6 and CFP-10 ( 10 μg/ml of every peptide ) or controls in the presence of Brefeldin A ( Golgiplug: BD biosciences ) for 3 . 5 hours ( for granulomas ) or 6 hours ( for PBMCs ) at 37°C with 5% CO2 . For PBMC , Brefeldin A was added after 1 hour stimulation with Mtb antigens or controls . Positive control included stimulation with phorbol dibutyrate ( PDBu ) and ionomycin and negative controls included a media only control and an isotype controls ( only for intracellular cytokine markers ) for all PBMC samples and for granulomas whenever additional cells were available . For flow cytometry , cells from granulomas were initially stained for viability marker ( Invitrogen ) followed by cell surface marker CD3 ( clone SP34-2; BD Pharmingen ) for T cells . Cell surface markers for PBMC T cells included CD4 ( clone L200 , BD Horizon ) and CD8 ( clone SK1: BD biosciences ) and markers for T cell memory subsets included CD45RA ( clone 5H9 , BD biosciences ) and CD27 ( clone O323 , eBioscience ) . Intracellular cytokine staining panel for both granulomas and PBMC included Th1 pro-inflammatory cytokines: IFN-γ ( clone B27 ) , IL-2 ( clone: MQ1-17H12 ) , TNF ( clone: MAB11 ) ; Th17 cytokine: IL-17 ( clone eBio64CAP17 ) and regulatory cytokine: IL-10 ( clone JES3-9D7 ) as previously described [9] . Data acquisition was performed using an LSR II ( BD ) and analyzed using FlowJo Software v . 9 . 7 ( Treestar Inc , Ashland , OR ) . For all PBMC T cells , the non-specific T cell response from the negative control ( media ) was subtracted from the Mtb specific antigen stimulated responses . MIATA guidelines were followed for sample collection and staining procedure for PBMC samples . The gating strategies used for the evaluation of granulomas and PBMC are described in S8 and S9 Figs . Intracellular cytokine data from granulomas in latent control animals ( Fig 5 ) was previously published [9] but not analyzed based on risk of reactivation . PBMC T cell cytokine data ( S10 and S11 Figs ) from a subset of the animals was used for computational modeling and has been published [28] . Single cell suspension of each harvested site ( i . e . , lung granulomas , complex pathologies , grossly normal lung , MLN , liver , spleen ) was plated on 7H11 plates ( minimum detection level of M . tuberculosis burden was estimated at <10 Colony Forming Units per granuloma ) as previously described [2 , 8] . Specific bacterial burden of each site ( granuloma , complex pathologies , or MLN ) was calculated as the product of the bacterial burden on a per gram basis and the total mass of the tissue site . Total thoracic burden was calculated as the sum of all M . tuberculosis growth from the lung ( includes all individual granulomas and more complex pathologies such as consolidations , TB pneumonia , coalescing granulomas and clusters ) and mediastinal lymph nodes . Total lung burden was calculated as the sum of all M . tuberculosis growth from lung lesions without mediastinal lymph nodes . The bacterial burden data were then transformed by adding 1 and reported as CFU . Normal distribution of the data was assessed for each continuous variable using the D’Agostino-Pearson Omnibus Test . For statistical comparison , pair-wise analysis of continuous data was performed by Student's T test for normally distributed data and Mann-Whitney test for nonparametric data . For analyses in which more than two groups were compared , Kruskall-Wallis test was performed with Dunn’s multiple comparisons as a post-hoc test of non-normally distributed data . Pair-wise analysis for matched data was performed using the Wilcoxon rank-sum test . P-values below 0 . 10 were reported specifically in figures and text . To assess which PET CT variables could distinguish reactivators from non-reactivators before TNF neutralization , we ran several different simple logistic regression models ( and a contingency analysis ) to narrow down the best predictors using the following variables: total lung FDG activity , number of “hot” lymph nodes , total number of lymph nodes , total SUV of lymph nodes , number of lesions , and the presence of extrapulmonary lesions . We chose total lung FDG activity and the presence of extrapulmonary lesions as best predictors based on goodness of fit and then used recursive partitioning ( a decision tree ) to evaluate a cut-off for total lung FDG activity . A receiver operating characteristic ( ROC ) curve was plotted in order to graphically represent the sensitivity and specificity of the combination of these two predictors . Statistical analysis was performed using Prism 6 . 0 ( Graphpad Software , Inc ) . The ROC curve was plotted in JMP Pro 10 . 2 ( SAS Institute Inc . ) . Given the variability in the number of granulomas per animals , methods were developed to minimize the potential for bias among animals that had a greater number of lesions ( i . e . , granulomas and MLN ) for analysis . We derived a number of representative samples per monkey by first calculating the median number of samples ( for which we had bacterial burden ) per monkey per group so that one monkey could not contribute more than the median number of samples per monkey . | Asymptomatic infection with Mycobacterium tuberculosis , often called latent tuberculosis , affects more than 2 billion people . Reactivation of latent infection to active TB occurs in only a minority of those infected , yet can lead to deadly disease and transmission . Here we show , using a non-human primate model , that imaging using PET/CT can identify certain features that are associated with a higher risk of reactivation . These factors include overall lung inflammation , individual granulomas in the lung with higher bacterial burden , and a site of infection outside the lungs . Using these parameters may allow discovery of peripheral biomarkers regarding risk of reactivation from latent TB . Such biomarkers could identify those people who would benefit most from treatment of latent TB to prevent reactivation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"blood",
"cells",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"diagnostic",
"radiology",
"granulomas",
"immunology",
"vertebrates",
"neuroscience",
"animals",
"mammals",
"positron",
"emission",
"tomography",
"primates",
"developmental",
"biology",
"lymph",
"nodes",
"lymphatic",
"system",
"molecular",
"development",
"bacteria",
"neuroimaging",
"old",
"world",
"monkeys",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"imaging",
"techniques",
"monkeys",
"animal",
"cells",
"t",
"cells",
"actinobacteria",
"tomography",
"computed",
"axial",
"tomography",
"macaque",
"immune",
"system",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"cell",
"biology",
"anatomy",
"mycobacterium",
"tuberculosis",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"amniotes",
"organisms"
]
| 2016 | PET CT Identifies Reactivation Risk in Cynomolgus Macaques with Latent M. tuberculosis |
Virus-specific CD8+ T cells ( TCD8+ ) are initially triggered by peptide-MHC Class I complexes on the surface of professional antigen presenting cells ( pAPC ) . Peptide-MHC complexes are produced by two spatially distinct pathways during virus infection . Endogenous antigens synthesized within virus-infected pAPC are presented via the direct-presentation pathway . Many viruses have developed strategies to subvert direct presentation . When direct presentation is blocked , the cross-presentation pathway , in which antigen is transferred from virus-infected cells to uninfected pAPC , is thought to compensate and allow the generation of effector TCD8+ . Direct presentation of vaccinia virus ( VACV ) antigens driven by late promoters does not occur , as an abortive infection of pAPC prevents production of these late antigens . This lack of direct presentation results in a greatly diminished or ablated TCD8+ response to late antigens . We demonstrate that late poxvirus antigens do not enter the cross-presentation pathway , even when identical antigens driven by early promoters access this pathway efficiently . The mechanism mediating this novel means of viral modulation of antigen presentation involves the sequestration of late antigens within virus factories . Early antigens and cellular antigens are cross-presented from virus-infected cells , as are late antigens that are targeted to compartments outside of the virus factories . This virus-mediated blockade specifically targets the cross-presentation pathway , since late antigen that is not cross-presented efficiently enters the MHC Class II presentation pathway . These data are the first to describe an evasion mechanism employed by pathogens to prevent entry into the cross-presentation pathway . In the absence of direct presentation , this evasion mechanism leads to a complete ablation of the TCD8+ response and a potential replicative advantage for the virus . Such mechanisms of viral modulation of antigen presentation must also be taken into account during the rational design of antiviral vaccines .
CD8+ T cells ( TCD8+ ) play important roles in host elimination of pathogens , tumors and transplanted tissues . Virus-specific TCD8+ recognize major histocompatibility complex ( MHC ) class I molecules bound to peptides derived from viral proteins [1] . These peptide-MHC complexes can be generated via two spatially distinct pathways . Virus-infected cells present peptides derived primarily from a subset of viral proteins that are rapidly degraded in a process known as direct presentation [2] . Alternatively , long-lived protein substrates may be transferred from virus-infected cells to pAPC where they are processed and presented by uninfected cells via the cross-presentation pathway [3] . The extent to which the direct or cross-presentation pathways contribute to the induction of virus-specific TCD8+ in vivo remains controversial [4] . Many pathogens have evolved mechanisms to modulate or evade the direct-presentation pathway [5] , implying that such mechanisms may confer a survival advantage . Cross presentation is generally thought to compensate when direct presentation is blocked , allowing the generation of specific TCD8+ targeting such pathogens [5] . Here we delineate a unique mechanism of viral immune evasion whereby viral antigen is prevented from entering the cross-presentation pathway . We investigated the pathways used for presentation of vaccinia virus ( VACV ) antigens driven by late promoters . Recombinant antigens driven by VACV late promoters , which are active only following DNA replication , stimulate poor or undetectable TCD8+ responses as compared with the response to identical antigens driven by early VACV promoters [6] . This reduced response occurs despite production of much larger quantities of late promoter-driven antigen both in vitro and in vivo . The inability of late VACV promoter-driven antigen to stimulate TCD8+ responses has been correlated to an abortive in vitro infection of pAPC in which late antigens are not produced and so direct presentation cannot occur [7] . Here , we demonstrate that despite the availability of the cross-presentation pathway for initiation of an antiviral TCD8+ response the late VACV promoter driven antigen cannot enter the cross-presentation pathway . We provide evidence of a mechanism that is dependent upon sequestration of antigen during the poxvirus life cycle and which is specific for the cross-presentation pathway within pAPC . These data are the first to describe an evasion mechanism of the cross-presentation pathway that in the absence of the direct-presentation pathway leads to a complete ablation of the TCD8+ response and a likely replicative advantage for the virus .
In order to directly study the effects of driving antigen expression with early or late VACV promoters following infection , we used recombinant viruses in which the early p7 . 5 or late p11 promoter drive expression of a model antigen . We used β galactosidase ( β-gal ) as a model antigen as it contains well-defined MHC class I binding determinants and its activity can be readily measured by enzymatic methods even when present in low quantities . We measured proliferation of adoptively transferred BG1 TCR transgenic TCD8+ ( specific for β-gal96–103-Kb complexes ) [8] in response to immunization with VACV expressing β-gal driven by the p7 . 5 ( rVACV-β-gal-Early ) or p11 ( rVACV-β-gal-Late ) promoters . The BG1 TCD8+ did not proliferate ( Fig . 1A ) or acquire effector activity ( Fig . 1B ) upon immunization with rVACV-β-gal-Late and did not accumulate above background levels following immunization with a control VACV ( data not shown ) . Proliferation of BG1 TCD8+ in mice immunized with rVACV-β-gal-Late could be stimulated following subsequent immunization with adenovirus encoding β-gal ( data not shown ) . Thus , late promoter-driven β-gal does not stimulate TCD8+ responses , and the lack of a TCD8+ response does not result from tolerance induced by high dose late promoter-driven antigen . The reduced immunogenicity of recombinant antigens driven by late VACV promoters has been correlated to a lack of activity of these promoters in pAPC , such as macrophages [9] and dendritic cells [7] in vitro . To determine whether late VACV promoters are functional in various cell types we measured β-gal production in a fibroblast cell line or in bone marrow-derived dendritic cells ( BMDC ) infected with either rVACV-β-gal-Early or rVACV-β-gal-Late using a chromogenic β-gal substrate . Our limit of detection using a chromogenic β-gal substrate is 10−8 mg/mL of β-gal ( Fig . S1 ) . Figure 2A demonstrates typical expression of β-gal from each virus in fibroblasts . rVACV-β-gal-Early produced a linear accumulation of β-gal almost immediately following infection , while β-gal from rVACV-β-gal-Late is not detectable until >3 h post infection . β-gal produced from rVACV-β-gal-Late rapidly accumulates in much greater quantities than that from rVACV-β-gal-Early , with equivalent levels of β-gal present after 5 h of infection . In contrast to β-gal production in fibroblasts , expression of β-gal from rVACV-β-gal-Late was undetectable in BMDC ( Fig . 2B ) while β-gal production from rVACV-β-gal-Early occurred rapidly after infection . As our limit of detection was 10−8 mg/mL we can conclude that β-gal production was lower than 10 attograms/cell ( 10−18 g/cell ) in BMDC . DC are phenotypically and functionally specialized in vivo beyond the phenotype of BMDC . The major subsets of DC in vivo include CD11b+ CD8α− , CD11b− CD8α+ “lymphoid-resident” DC and B220+ plasmacytoid DC . We infected DC purified from the spleens of wild-type mice with VACV expressing EGFP-OVA driven by early or late promoters and examined expression of EGFP-OVA in each of these DC subsets . Expression of eGFP from VACV-eGFP-OVA-Late was not detectable above background levels in infected plasmacytoid DC ( CD11c+ , B220+ ) , CD11b+ CD8α− DC , or CD11b− CD8α+ DC while each DC subset readily expressed eGFP from eGFP-OVA-Early ( Fig . 2C ) . Thus , VACV undergoes an abortive infection in all DC subsets such that VACV late promoter-driven antigens are not expressed following infection . To extend these observations in vivo we infected mice intradermally with rVACV-β-gal-Early or rVACV-β-gal-Late and then visualized β-gal production at the site of infection or in the draining lymph node . Twelve h after infection , β-gal production was readily detectable from either virus at the site of infection ( Fig . 3A ) . However , production of β-gal could only be detected in the draining lymph node after infection with rVACV-β-gal-Early ( Fig . 3B , C ) . We have previously observed that all of the VACV infected cells in a lymph node are macrophages or DC at 12 h post infection [10] indicating that late promoter-driven antigen is undetectable in infected pAPC in vivo . The primary substrates for production of peptides in the direct-presentation pathway are rapidly degraded proteins that may be defective [2] . Such proteins are unlikely to acquire the secondary structure required to become enzymatically active and so may not be detected in our assays . To ensure that β-gal from rVACV-β-gal-Late is not directly presented by virus-infected BMDC , we infected BMDC or fibroblasts expressing H2-Kb and measured antigen presentation to primary β-gal96–103-specific TCD8+ . Infected fibroblasts stimulated interferon-γ production in TCD8+ regardless of whether the early or late promoter drove β-gal production ( Fig . 3D ) . VACV-infected BMDC triggered interferon-γ by β-gal96–103-specific TCD8+ only when infected with rVACV-β-gal-Early ( Fig . 3E ) even when the infection was allowed to proceed for >12 h ( data not shown ) . Thus , direct presentation of β-gal driven by a late promoter did not occur in infected pAPC . Under conditions where the direct-presentation pathway is blocked in vivo , the cross-presentation pathway is thought to compensate and allow generation of TCD8+ [11] , [12] . However , this compensatory mechanism does not occur with late promoter-driven VACV β-gal ( Fig . 1 ) , despite the accumulation of large quantities of antigen that should increase the efficiency of cross presentation [13] . This observation has been interpreted as a functional irrelevance of cross presentation in the induction of virus-specific TCD8+ [14] , but could also be explained by an inability of late promoter-driven antigen to enter the cross-presentation pathway , a hitherto undescribed phenomenon . To examine cross presentation of β-gal driven by the early or late promoters , we infected SV40 transformed cells that lack β2-microglobulin ( TAg-β2mneg ) and are therefore direct presentation-incompetent . At 5 h post-infection , a time point at which equivalent levels of β-gal are expressed ( Fig . 2A ) , the cells were treated with psoralen and UVC to halt both protein production and potential virus spread [15] . We measured the ability of these cells to stimulate proliferation and effector function of adoptively transferred BG1 TCD8+ following in vivo immunization . Under these conditions , initiation of a TCD8+ response can only occur following antigen presentation via the cross-presentation pathway . TAg-β2mneg cells infected with rVACV-β-gal-Early efficiently triggered proliferation of BG1 TCD8+ ( Fig . 4B ) but those infected with rVACV-β-gal-Late failed to stimulate proliferation ( Fig . 4C ) or effector function at levels above those found following immunization with TAg-β2mneg cells infected with a control VACV ( Fig . 4D ) . Similar data were obtained after infection with rVACV-β-gal-Late for up to 11 h ( data not shown ) , a time point at which p11-driven β-gal is present in enormous excess compared to p7 . 5 driven β-gal ( Fig . 2A ) . Infection with rVACV-β-gal-Early allowed access to the cross-presentation pathway in vivo as soon as 1 h post-infection ( Fig . 4E–G ) indicating that antigen was not limiting even when present at low intracellular concentrations . These data clearly indicate that late promoter-driven VACV β-gal is not accessible to the cross-presentation pathway even when present in very large quantities . We have previously demonstrated that cellular protein synthesis , which is rapidly halted following VACV infection , is not required for antigen donation [8] . Nonetheless , it is possible that VACV infection may block donation of all cellular antigen . To investigate this possibility , we exploited the expression of the SV40 T antigen ( TAg ) as a cellular protein in TAg-β2mneg cells . We measured proliferation of adoptively transferred BG1 and SV40 TAg Site I-specific TCR transgenic T cells [16] simultaneously in mice immunized with TAg-β2mneg cells infected with rVACV-β-gal-Early or rVACV-β-gal-Late . As before , rVACV-β-gal-Late infected TAg-β2mneg cells failed to induce proliferation of BG1 TCD8+ ( Fig . 5C ) but in the same recipient mice proliferation of Site I TAg TCD8+ occurred efficiently ( Fig . 5F ) . The entry of cellular antigen into the cross-presentation pathway is therefore not blocked by VACV infection . It is possible that VACV encoded proteins produced after infection can bind to newly synthesized cellular antigen and prevent entry into the cross-presentation pathway . However , as TAg is constitutively expressed in TAg-β2mneg cells the existing cellular pool of antigen could be resistant to such a mechanism of inhibition of cross presentation . Ideally , to examine this possibility one would initiate transcription of a cellular antigen after VACV infection , but as VACV is so adept at shutting down host protein synthesis the initiation of transcription of a cellular gene following VACV infection is technically challenging . Therefore we introduced soluble antigen into TAg-β2mneg cells after 5 h of VACV infection and measured the response to this antigen in vivo . Again , VACV infection did not inhibit the donation of β-gal ( Fig . 5G–I ) or OVA ( not shown ) introduced into infected cells . These data indicate that VACV does not globally suppress the availability of antigen to enter the cross-presentation pathway in vivo but utilizes a specialized mechanism to prevent the access of its own antigens to the cross-presentation pathway . Katsafanas and Moss recently described that soluble proteins driven by intermediate and late promoters are concentrated within cytosolic virus factories following coordinated transcription and translation within these domains [17] . Virus factories are rough endoplasmic reticulum-bound perinuclear organelles in which VACV replication and early assembly of viral particles occurs [18] . There is a possibility that the specialized structure of these compartments in which late antigens are synthesized could prevent access to the cross-presentation pathway . VACV-infected TAg-β2mneg cells were visualized to determine the localization of β-gal relative to virus factories labeled with DAPI and the VACV double stranded RNA binding protein E3L ( Fig . 6 ) . β-gal from rVACV-β-gal-Early was distributed throughout the cytosol of the cell ( Fig . 6C , D ) , and only 1 . 3% ( +/−0 . 2 ) of pixels staining for β-gal were localized within virus factories . In contrast , β-gal from rVACV-β-gal-Late was localized only to the perinuclear virus factories ( Fig . 6G , H ) , with greater than 83% ( +/−4 . 8% ) of pixels staining for β-gal being localized within virus factories . An altered distribution of antigen thereby correlates with an inability of that antigen to enter the cross-presentation pathway , and sequestration of newly synthesized antigen within VACV virus factories likely facilitates this process . To test whether sequestration of antigen within virus factories is essential for the blockade in cross presentation we used recombinant VACV expressing the model antigen HSV-1 glycoprotein B ( gB ) driven by the p11 promoter ( rVACV-gB-Late ) [19] . The egress of some late VACV proteins from virus factories is required for viral replication . Targeting of such proteins to the secretory pathway allows proteins to leave the virus factories , so we surmised that similar sequences within the gB protein might allow this protein to exit the factories . Figure 7A–D demonstrates that , in contrast to β-gal driven by a late VACV promoter , gB driven by the identical p11 promoter distributes across many cellular membranes and is not confined to VACV factories . The ability of gB to leave virus factories did not allow direct presentation of the gB498–505 peptide by pAPC , as BMDC infected with rVACV-gB-Late did not activate a gB-specific T cell hybridoma ( Fig . 7E ) . However , proliferation of adoptively transferred gB-specific TCR transgenic TCD8+ could be detected following immunization with rVACV-gB-Late ( Fig . 7F ) . As direct presentation was blocked in pAPC , the proliferation likely resulted from cross presentation of gB-derived peptides . To test whether gB restricted to the cross-presentation pathway was immunogenic in vivo we immunized mice with TAg-β2mneg cells infected with VACV-gB-Late for 5 h . In contrast to the results observed with β-gal that was sequestered within VACV factories , TAg-β2mneg cells infected with VACV-gB-Late did stimulate proliferation of gB-specific TCR transgenic TCD8+ ( Fig . 7H ) . Thus , antigen that can leave VACV factories is available for cross presentation but antigen that remains sequestered within these factories is blocked from entering the pathway . Having gained a mechanistic insight into the means by which VACV acts within the virus infected cell to prevent access of late antigen to the cross-priming pathway we sought to investigate at what point the blockade of cross presentation occurred within pAPC . In order to preserve the in vivo nature of our studies we examined presentation of early or late promoter-driven β-gal by the MHC Class II presentation pathway . MHC Class II-mediated presentation of exogenous antigens shares many common components with the MHC Class I-restricted cross-presentation pathway so a differential ability to enter this pathway would give a strong indication of the point at which cross presentation is blocked . In order to study MHC Class II-restricted presentation of β-gal in vivo we constructed a transgenic mouse ( BG2 ) bearing a T cell receptor specific for a β-gal peptide presented in complex with MHC Class II . The majority of CD4 cells in the resulting mice expressed the Vα11 chain from the transgene ( Fig . 8B ) and produced IL-2 , IFN-γ and TNF-α in response to peptide sequences corresponding to residues 725–735 from β-gal [20] ( Table 1 ) . The TCD4+ from the transgenic mice also proliferated following adoptive transfer into a wild-type mouse that was then infected with rVACV-β-gal-Early ( Fig . 8C ) . MHC Class II-restricted presentation can occur through a number of pathways , including presentation of endogenously synthesized antigen [21] . Early antigen may enter this pathway , but late antigen is not synthesized within pAPC ( Fig . 1 ) and so will not be presented from endogenous sources . To ensure that we were directly comparing MHC Class II-restricted presentation of β-gal driven by early or late promoters we adoptively transferred both BG1 . SJL TCD8+ and BG2 . SJL TCD4+ into mice and then immunized with TAg-β2mneg cells infected with rVACV-β-gal-Early , rVACV-β-gal-Late or control rVACV as above . We readily detected MHC Class I- and MHC Class II-restricted responses following immunization with rVACV-β-gal-Early or with TAg-β2mneg cells infected with rVACV-β-gal-Early ( Fig . 8D , F , H , J ) . As previously shown we did not observe an MHC Class I-restricted response following immunization with rVACV-β-gal-Late ( Fig . 8I ) or cells infected with rVACV-β-gal-Late ( Fig . 8K ) but we did detect an MHC Class II-restricted response under both of these circumstances ( Fig . 8E , G ) . Sequestration of antigen , therefore , specifically blocks components of the cross-priming pathway but not the MHC Class II presentation pathway .
The data presented here demonstrate three significant points . First , we show that cross presentation is an important compensatory mechanism of antigen presentation which when blocked results in a complete ablation of the TCD8+ response . If a virus inhibits the direct-presentation pathway in vivo the resulting TCD8+ response is often unchanged [12] , [22] . In contrast , we have demonstrated that if entry to the cross-presentation pathway is blocked when the direct-presentation pathway is unavailable , the TCD8+ response for the affected antigens is undetectable . Second , although many studies have described the modulation of the direct-presentation pathway , this is the first to describe a viral strategy to evade the cross-presentation pathway . Third , our data demonstrate that the blockade in cross presentation occurs because a number of viral antigens are sequestered within virus factories indicating that the subcellular localization of antigen may prevent access to the cross-presentation pathway . This observation has far reaching implications , as an altered localization of cellular antigens that are normally sequestered from the cross-presentation pathway may allow the induction of TCD8+-mediated autoimmunity . The blockade in cross presentation is specific for the cross-presentation pathway , as MHC Class II-restricted presentation of exogenous late antigen is unaffected . A previous study has examined the impact of altered cellular localization upon donation of antigen during cross presentation . The authors found that cellular localization could affect the efficiency of cross presentation [23] but the study could not rule out an effect of altered antigen stability , a known factor in the effectiveness of cross presentation [3] . Thus , prior investigations have not produced direct evidence to indicate that alteration of the cellular localization of an antigen can enhance or prevent its entry to the cross-presentation pathway . VACV infection alters vesicular trafficking within infected cells and induces the formation of specialized structures such as virus factories . VACV virus factories are cytoplasmic structures that are bound by rough ER . The ER membrane surrounding VACV virus factories is not continuous , however , and “holes” to the cytosol do exist [18] . Intermediate and late VACV proteins are transcribed and translated within virus factories [17] and require specialized signals to leave these structures [24] . The rules governing exit from VACV virus factories remain to be fully characterized . In our current study identical antigens with different cellular localizations are presented differently , with cross presentation of those sequestered within viral factories being completely ablated whereas those that are localized to the cytosol are available for cross presentation . This could indicate that alteration of the localization of cellular antigen may also prevent the entry of antigen into the cross-presentation pathway and subsets of the cellular proteome could be unavailable to the cross-presentation pathway . Point mutations in motifs responsible for the targeting of protein to compartments that sequester antigen from the cross-presentation pathway would render these antigens immunogenic , potentially producing TCD8+-mediated autoimmunity via the cross-presentation pathway . The blockade in cross presentation is specific , as the MHC Class II pathway that shares many components with the cross-presentation pathway is unaffected . Thus , pAPC-mediated internalization and degradation of late antigens sequestered within virus factories is likely unaltered . As MHC Class I-restricted direct presentation of late antigens sequestered within virus factories readily occurs this strongly indicates that the mechanism involved targets a specific component of the cross-presentation pathway . The unique component of the cross-presentation pathway involves release of antigen from within an endosomal/lysosomal compartment into the cytosol [25] , [26] , a process that may involve the retrotranslocation machinery involved in ER-associated degradation [27] . Human Cytomegalovirus alters ER-associated degradation to increase the degradation of MHC Class I heavy chains within infected cells , so the manipulation of this degradative pathway by viruses is possible [28] . Cross presentation of β-gal derived from VACV-β-gal-Early requires the TAP transporter ( data not shown ) , and thus retrotranslocation into the cytosol . This process of release of antigen into the cytosol represents the likely mechanism responsible for blockade of the cross-presentation pathway . Our studies have utilized model antigens expressed by VACV but the observations made can readily be extended to native VACV antigens . A number of studies have mapped MHC class I-restricted antigenic determinants from VACV proteins restricted by either mouse [29] , [30] , [31] or human MHC molecules [32] , [33] , [34] . The source of the mapped determinants reveals that the majority of peptides recognized are derived from early VACV gene products . In contrast , the majority of MHC Class II-restricted determinants are found within late VACV gene products [35] . A small number of peptides recognized by TCD8+ are found in late genes . All of these immunogenic late VACV genes contain N-terminal signal sequences or hydrophobic transmembrane domains and are components of the intracellular mature virus , intracellular enveloped virus , or extracellular enveloped virus membranes that would leave virus factories . The remainder of the determinants mapped within late VACV gene products are present within proteins that may associate with other VACV proteins ( e . g . A10L that associates with A4L [36] ) to facilitate their exit from factories . These data validate our hypothesis that late VACV proteins that remain within virus factories are not immunogenic whereas those that can leave can generate TCD8+ responses , likely via the cross-presentation pathway . Peptides derived from late gene products can enter the direct-presentation pathway , irrespective of whether the protein from which they are derived cannot exit the virus factory ( Fig . 3D ) . However , late VACV gene products are not produced within infected pAPC , and so any immunogenicity in the TCD8+ compartment likely results via the cross-presentation pathway . VACV is closely related to the cowpox virus [37] , which has been demonstrated to inhibit direct presentation by inhibiting movement of peptide-loaded MHC Class I molecules out of the ER [38] , [39] . It is not beyond the realm of possibility that a common ancestor of cowpox virus and VACV inhibited MHC Class I-restricted presentation of the majority of virus proteins . If egress of a particular late protein was required for virus replication then presentation of that antigen via the cross-presentation pathway could be evolutionarily tolerated . However , VACV has clearly gone to significant lengths to prevent access of other antigens to the cross-presentation pathway producing a newly discovered mechanism of evasion of the adaptive immune response .
Female C57BL/6 mice were purchased from Charles River Laboratories ( Wilmington , MA ) . OT-1 TCR RAG1−/− transgenic mice [40] , [41] were obtained from the NIAID Exchange Program ( Line 4175 ) . gBT-1 . 3 mice were a kind gift from Dr . Frank Carbone ( University of Melbourne , Victoria , Australia ) [42] . B6 . SJL-Ptprca/BoAiTac mice were purchased from Taconic Farms ( Germantown , NY ) and bred to both OT-1 TCR and BG1 TCR mice to produce OT-1 . SJL and BG1 . SJL offspring , respectively . SV40 Site I TCR mice were a kind gift from Dr . Satvir Tevethia ( Milton S . Hershey Medical Center , Hershey , PA ) [16] . All mice were maintained under specific pathogen-free conditions at the M . S . Hershey Medical Center . All studies were approved by the Penn State College of Medicine Institutional Animal Care and Use Committee . BG2 mice that express a T cell receptor on TCD4+ specific for an MHC class II-I-Ab-restricted epitope of β-gal on a C57BL/6 background were generated . Total RNA was isolated from an I-Ab-restricted , β-gal specific TCD4+ clone and the α and β TCR were amplified by a 5′-Rapid Amplification of cDNA Ends ( 5′ RACE , Invitrogen , Carlsbad , CA ) using constant region anti-sense primers a1 ( 5′-GGCTACTTTCAGCAGGAGGA-3′ ) and b1 ( 5′-AGGCCTCTGCACTCATGTTC-3′ ) , respectively . 5′-RACE products were amplified with nested TCR alpha and beta constant region primers a2 ( 5′-GGGACTCAAAGTCGGTGAAC-3′ ) and b2 ( 5′-CCACGTGGTCAGGGAAGAAG-3′ ) and cloned into pCR4TOPO TA sequencing vectors ( Invitrogen ) . Genomic cloning PCR primers were designed based upon the method previously described [43] . The genomic variable domains were validated by sequencing , subcloned into TCR cassette vectors kindly provided by Dr . Diane Mathis ( Harvard ) , and coinjected into fertilized C57BL/6 embryos ( SAIC , Frederick , MD ) yielding TCR transgenic founder mice . Mice were bred with B6 . SJL mice and maintained as heterozygotes . Transgene expression monitored by PCR or by staining of blood cells . For PCR , tail samples from 3–4 week old mice were employed for genotyping of BG2 mice using the red Extract-N-Amp Tissue PCR kit ( Sigma , St . Louis , MO ) . Primers used are as follows: BG2 Alpha F1: ACAACCCGGGATTCCACAG; BG2 Alpha R1: GTATAGCGGCCGCCTCCTAGTGCAATGGT; BG2 Beta F1: TATCTCGAGTCCTGCCGTGACCCTACTATG; BG2 Beta R1: CAGCCGCGGAACCCAACACAAAAACTATAC . Transgene expression was monitored by flow cytometry following staining with anti-PE-Vα11 ( Clone RR8-1 ) and anti-PE-Cy5-CD4 ( Clone L3T4 ) antibodies . To map the BG2 determinant , transgenic T cells were incubated with splenocytes in the presence of overlapping peptides ( 1 µM ) or whole βgal ( 50 µg/ml ) . Supernatants were collected for cytokine analysis 48 h post-stimulation using the CBA kit from BD Biosciences ( San Jose , CA ) . Only the peptides shown in Table 1 stimulated cytokine production by the BG2 cells . VACV ( Western Reserve strain ) , rVACV-β-gal-Late , rVACV-β-gal-Early , rVACV-gB-Late , rVACV-OVA , rVACV-gB498–505 , rVACV-CD4 [44] and recombinant adenovirus expressing β-gal ( Ad-β-gal ) were a kind gift from Dr . Jon Yewdell and Dr . Jack Bennink ( Laboratory of Viral Diseases , NIAID , Bethesda , MD ) . VACV expressing the β-gal96–103 peptide ( rVACV-β-gal96–103 ) targeted to the endoplasmic reticulum ( ER ) with a signal sequence derived from the adenovirus E3/19k protein was previous published [45] . The plasmid pRB21 expressing the full length vp37 VACV ORF with the p7 . 5 early/late promoter was a kind gift from Dr . Bernard Moss ( Laboratory of Viral Diseases , NIAID , Bethesda , MD ) [46] . The peGFP-C1 plasmid expressing full-length OVA ( peGFP-C1-OVA1–385 ) was a kind gift from Dr . Kenneth Rock ( Department of Pathology , University of Massachusetts Medical School , Worcester , MA ) [23] . For construction of VACV-eGFP-OVA-Late pRB21 backbone DNA was ligated with eGFP-OVA using T4 DNA Ligase ( Invitrogen ) . Following ligation , plasmid DNA was sequenced to ensure that the vp37 , p7 . 5 early/late promoter , and eGFP-OVA1–385 sequences were correct . To make rVACV-eGFP-OVA-Late the p11 promoter was inserted in place of the p7 . 5 promoter . rVACV-eGFP-OVA-Early and rVACV-eGFP-OVA-Late were generated by infecting transfected BSC-1 cells infected with VACV-vRB12 at an MOI of 1 using the CellPhect Transfection Kit ( GE Healthcare , Buckinghamshire , UK ) . As VACV-vRB12 contains the flanking sequences of vp37 , homologous recombination occurred to allow virus spread [46] . The resulting rVACV were plaque purified three times prior to characterization . The resulting rVACV-eGFP-OVA-Early and rVACV-eGFP-OVA-Late produced green fluorescence upon infection of WT3 cells and sequencing revealed the presence of the correct promoter and OVA sequences in DNA purified from virions . All media were purchased from Invitrogen . WT3 [47] , TAg-β2mneg [15] and L929 fibroblasts that stably express Kb ( L-Kb ) were maintained in Dulbecco's Modified Eagle Media containing 10% fetal bovine serum ( FBS ) supplemented with penicillin/streptomycin and 2 mM L-glutamine . E22 cells ( the H2b EL4 thymoma transfected with β-gal ) [45] were maintained in RPMI 1640 , 5% FBS , penicillin/streptomycin , 2 mM L-glutamine and 400 mg/ml G418 . The gB498–505-specific LacZ T cell hybridoma , 2E2 , was a kind gift from Dr . Frank Carbone ( University of Melbourne , Victoria , Australia ) and was maintained in RPMI 1640 , 5% FBS , penicillin/streptomycin , 2 mM L-glutamine . Bone marrow-derived dendritic cells ( BMDC ) were generated as previously described [48] . C57BL/6 mice were inoculated i . d . with approximately 5×105 Flt3 ligand expressing B16 tumor cells . Two weeks later the spleens from immunized mice were harvested , microdissected , and incubated in 1 mg/mL Collagenase D ( Roche Diagnostics , Indianapolis , IN ) at 37°C for 20 min . Following lysis of red blood cells the remaining cells were incubated with Pan-DC microbeads ( Miltenyi Biotec , Auburn , CA ) and positively sorted . Purified DC were infected with rVACV-eGFP-OVA1–385-Early or rVACV-eGFP-OVA1–385-Late at an MOI of 10 for a duration of 7 hours in the presence or absence of cytosine arabinoside and analyzed by flow cytometry for the expression of eGFP . Live mononuclear splenocytes from mice immunized 30 d previously with 1×106 pfu Ad-β-gal were harvested by centrifugation over a Lymphocyte Separation Medium ( LSM ) cushion ( BioWhittaker , Walkersville , MD ) , washed once and resuspended at 1×107 cells per well in RPMI 1640 with 10% FBS , 1% non-essential amino acids , penicillin/streptomycin , 2 mM L-glutamine , and 7 . 5 U/ml of IL-2 ( Peprotech , Rocky Hill , NJ ) . Cells were stimulated weekly with 2 . 5×105 irradiated E22 cells per well . Spleens and lymph nodes were removed , homogenized to produce a single cell suspension , and mononuclear cells isolated as above . Where indicated , cells were labeled with 5 µM 5- ( and-6 ) carboxyfluorescein diacetate , succinimidyl ester ( CFDA-SE , Invitrogen ) for 10 min at 37°C and washed once prior to injection . Approximately 4×106 TAg-β2mneg cells were suspended in phosphate buffered saline ( PBS ) containing 1 mg/mL ovalbumin ( OVA ) or 1 mg/mL β-gal with 10 mM MgCl2 and incubated on ice for 10 minutes . The cells were then electroporated in disposable cuvettes ( Bio-Rad , Hercules , CA ) on a Bio-Rad gene pulser at 0 . 25 kV or 0 . 45 kV with a capacitance of 250 uFD . Following electroporation , cells were incubated on ice for an additional 10 min and washed three times with 10% Iscoves Modified Dulbecco's Medium ( IMDM ) . Cells were irradiated at 20 , 000 rad prior to injection . For in vivo immunization , mice were infected i . v . with 1×107 pfu of VACV or were injected i . p . with TAg-β2mneg that were either infected with VACV or electroporated with antigen as described above . TAg-β2mneg were infected with VACV at a multiplicity of infection of 10 and then treated with psoralen and ultraviolet light ( UV-C ) as previously described [3] . As VACV will not infect all cells , in some experiments TAg-β2mneg were infected with rVACV-CD4 , and infected cells were sorted using anti-CD4 microbeads ( Miltenyi Biotech ) . Mononuclear cells isolated from splenocytes or TCD8+ lines were washed twice after isolation over an LSM cushion and plated in triplicate into individual wells of a 96 well plate ( 3×106 cells per well ) . Cells were stimulated with 10−6 β-gal96–103 peptide for 2 h at 37°C or were incubated with BMDC infected with VACV as indicated . After 2 h of stimulation , 10 µg/mL Brefeldin A ( BFA , Sigma , St . Louis , MO ) was added and the cells were incubated for another 4 h . TCD8+ were then assayed for production of IFN-γ by flow cytometry . BMDC were incubated with anti-CD11c microbeads ( Miltenyi Biotech ) and positively sorted . Purified DC were infected with VACV ( MOI = 20 ) for a duration of 7 h in the presence or absence of cytosine arabinoside . Infected BMDC were then incubated with β-gal96–103-specific T cells generated as outlined above , and activation of the T cells was determined either by intracellular cytokine staining , or by activation of the LacZ hybridoma 2E2 using the chlorophenol red β-D-galactopyranoside ( CPRG ) substrate of β-gal as outlined below . For all assays , cells were incubated on ice with Fc block containing 20% normal mouse serum ( Sigma ) for 20 min prior to staining . For intracellular cytokine staining analysis , all antibodies were purchased from BD Biosciences except where noted . Cells were stained with anti-CD8 PE-Cy5 ( Clone 53-6 . 7 ) , washed once with PBS , and fixed with 1% paraformaldehyde ( PFA ) . Fixed cells were then stained with anti-IFN-γ-FITC ( Clone XMG1 . 2 ) in 0 . 5% saponin , washed , and analyzed . Antibodies used to identify OT-1 . SJL or BG1 . SJL cells were anti-CD45 . 1-PE ( Clone A20 ) . Antibodies used to identify gBT-I . 3 cells were anti-Vα2-PE ( Clone B20 . 1 ) . For SV40 Site I and BG1 . SJL double adoptive transfers , cells were stained in triplicate with anti-CD8-PE-Cy7 ( Clone 53-6 . 7 ) and anti-Vβ7-PE ( Clone TR310 ) for SV40 site I TCR cells and anti-CD45 . 1-PE-Cy5 ( eBioscience , San Diego , CA , Clone A20 ) for BG1 . SJL TCR cells . For BG2 and BG1 double adoptive transfer cells were stained with anti-CD45 . 1-PE to identify adoptively transferred cells and with anti-CD8-Alexa Fluor 750 and anti-PE-Cy5-CD4 ( Clone L3T4 ) to distinguish the two cell populations . Antibodies used to distinguish DC subsets were anti-CD11c-PE ( eBioscience , Clone N418 ) , anti-CD8α-PerCP-Cy5 . 5 ( Clone 53-6 . 7 ) , anti-CD11b-Alexa Fluor 750 ( eBioscience , Clone M1/70 ) , anti-CD45R/B220-Alexa Fluor 647 ( eBioscience , Clone RA3682 ) , anti-CD90 . 2-Biotin ( eBioscience , Clone 53-2 . 1 ) , anti-NK1 . 1-Biotin ( eBioscience , Clone PK 136 ) , anti-CD19-Biotin ( eBioscience , Clone 1D3 ) , and PE-Cy7 Conjugated Streptavidin . DC subsets were distinguished based on the expression of CD11c ( CD11c+ , CD8+ , CD11b− , B220− ) ( CD11c+ , CD8− , CD11b+ , B220− ) ( CD11c+ , B220+ ) and the lack of expression of CD90 . 2 , NK1 . 1 , and CD19 . To measure expression of β-gal , cells were infected with VACV for 1–12 h at a MOI of 10 in IMDM . Activity of β-gal in cells was determined using either of the β-gal substrates , o-nitrophenol β-D-galactoside ( ONPG ) or CPRG . Briefly , for the ONPG assay , approximately 3–5×105 cells were lysed with 150 µL 1% Igepal ( Sigma , St . Louis , MO ) and 10 µl aliquots incubated with 150 µL 1 mg/mL ONPG substrate in Z buffer ( 0 . 06 M Na2HPO4 , 0 . 04 M NaH2PO4 , 0 . 01 M KCl , 0 . 001 M MgSO4 , 40 mM β-mercaptoethanol ) for 10 min at 37°C . After 10 minutes the reaction was stopped by addition of 50 µL Na2CO3 . β-gal activity was measured using a micro-plate reader ( Dynex , Chantilly , VA ) at 405 nm wavelength . For the CPRG assay 1×105 cells per well were washed twice in cold PBS and incubated with 0 . 15 mM CPRG , 10 mM phosphate buffer , 1 mM MgCl2 , and 0 . 1255% Igepal . Upon color change , 50 µL of stop buffer ( 300 mM glycine , 15 mM EDTA , 10 M NaOH ) was added , and absorbance measured at a wavelength of 595 nm , with 630 nm as a reference wavelength . To measure localization of virally expressed recombinant antigen , TAg-β2mneg cells were plated in 8 well Permanox chamber slides ( Nalge Nunc International , Rochester , NY ) and allowed to adhere overnight . Cells were infected at a MOI of 20 with VACV for 5 h and then fixed for 15 min with 4% PFA . Cells were permeabilized with 0 . 2% Triton X-100 ( Bio-Rad ) and blocked with 20% goat serum ( Sigma ) for 20 min . Infected cells were stained with primary antibodies as follows in 10% goat serum: Unconjugated polyclonal rabbit anti-β-gal IgG antibody ( AbCam , Cambridge , MA ) , mouse anti-vaccinia E3L ( TW2 . 3 supernatant ) [49] , unconjugated mouse anti-gB IgG antibody ( Virusys , Sykesville , MD ) or polyclonal rabbit anti-vaccinia IgG-FITC antibody ( Biogenesis , Kingston , NH ) . Secondary antibodies used were goat anti-rabbit IgG-Alexa Fluor 647 , goat anti-mouse IgG-Alexa Fluor 647 , and goat anti-mouse IgG-Alexa Fluor 488 ( all from Invitrogen ) . The slides were overlaid with ProLong Gold antifade reagent with 4′-6-diamidino-2-phenylindole ( DAPI ) ( Invitrogen ) and allowed to cure overnight . Mice were infected i . d . in each ear with rVACV-β-gal-Early or rVACV-β-gal-Late . Twelve h post-infection , ears were removed and fixed in 2% PFA/0 . 2% gluteraldehyde . Cervical lymph nodes were frozen in Tissue-Tek OCT Compound ( Fisher Scientific , Pittsburgh , PA ) , sections ( 15 µm ) cut using a Bright Cryostat ( Hacker Instruments , Winnsboro , SC ) and then fixed with 10% buffered formalin phosphate . β-gal expression was visualized using 5-bromo-4-chloro-3-indolyl-β-D galactopyranoside ( X-gal , 0 . 25 mg/ml ) in 2 mµ potassium ferrocyanide , 5 mM ferricyanide and 2 mM MgCl2 in PBS following overnight incubation at 37°C . All images of infected cells , murine ear and lymph node sections were acquired on an Olympus IX81 deconvolution microscope ( Olympus , Center Valley , PA ) using Slidebook 4 . 0 software ( Intelligent Imaging Innovations , Denver , CO ) or Q Capture software ( QImaging , Burnaby , BC , Canada ) . Colocalization was measured using the Colocalization Plugin for ImageJ analysis software ( NIH ) . | Understanding the pathways by which protective immunity is mediated against viral pathogens is essential to allow the design of effective vaccines . No effective vaccine has been designed to activate killer cells of the immune system expressing CD8 , although CD8+ T cells are the most effective cells at modulating anti-viral immunity . We have studied the process that activates the CD8+ T cell to better understand how the cells are triggered so future vaccines might readily activate these cells . CD8+ T cells are activated following recognition of small peptides derived from a virus that binds to a cell surface MHC molecule . Many viruses have evolved to prevent the presentation of these peptide-MHC complexes to CD8+ T cells . However , the immune system avoids these viral “evasion” mechanisms by allowing virus-derived peptides to be generated from viral proteins that are taken up by uninfected cells , a process termed “cross presentation” . We have shown that a poxvirus can specifically prevent the presentation of its proteins by uninfected cells , the first demonstration of evasion of cross presentation . This knowledge is vital in the use of certain viral vectors during vaccine design and adds to the numerous ways in which viruses can evade the immune system . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"virology/immune",
"evasion",
"immunology/antigen",
"processing",
"and",
"recognition",
"immunology/immunity",
"to",
"infections",
"virology/host",
"antiviral",
"responses"
]
| 2009 | Viral Sequestration of Antigen Subverts Cross Presentation to CD8+ T Cells |
As part of the Nucleotide Excision Repair ( NER ) process , the endonuclease XPG is involved in repair of helix-distorting DNA lesions , but the protein has also been implicated in several other DNA repair systems , complicating genotype-phenotype relationship in XPG patients . Defects in XPG can cause either the cancer-prone condition xeroderma pigmentosum ( XP ) alone , or XP combined with the severe neurodevelopmental disorder Cockayne Syndrome ( CS ) , or the infantile lethal cerebro-oculo-facio-skeletal ( COFS ) syndrome , characterized by dramatic growth failure , progressive neurodevelopmental abnormalities and greatly reduced life expectancy . Here , we present a novel ( conditional ) Xpg−/− mouse model which -in a C57BL6/FVB F1 hybrid genetic background- displays many progeroid features , including cessation of growth , loss of subcutaneous fat , kyphosis , osteoporosis , retinal photoreceptor loss , liver aging , extensive neurodegeneration , and a short lifespan of 4–5 months . We show that deletion of XPG specifically in the liver reproduces the progeroid features in the liver , yet abolishes the effect on growth or lifespan . In addition , specific XPG deletion in neurons and glia of the forebrain creates a progressive neurodegenerative phenotype that shows many characteristics of human XPG deficiency . Our findings therefore exclude that both the liver as well as the neurological phenotype are a secondary consequence of derailment in other cell types , organs or tissues ( e . g . vascular abnormalities ) and support a cell-autonomous origin caused by the DNA repair defect itself . In addition they allow the dissection of the complex aging process in tissue- and cell-type-specific components . Moreover , our data highlight the critical importance of genetic background in mouse aging studies , establish the Xpg−/− mouse as a valid model for the severe form of human XPG patients and segmental accelerated aging , and strengthen the link between DNA damage and aging .
If DNA damage , either inflicted from exogenous or endogenous sources , cannot be repaired , this has detrimental consequences for an organism ranging from transcription blocks , permanent cell cycle arrest and mutations , to cell death . In the end , this unrepaired DNA damage contributes to the onset and progression of the aging process , as well as to cancer [1]–[3] . Cells are equipped with a set of elaborate DNA repair mechanisms integrated into a complex DNA damage response machinery that jointly attempt to fix the unrepaired DNA [4] . One such DNA repair mechanism is the Nucleotide Excision Repair ( NER ) pathway that removes a wide category of helix-distorting lesions , such as those induced by UV and bulky chemical adducts , in a tightly coordinated process involving over 30 proteins [5]–[7] . NER can be divided into two subpathways based on the mode of damage recognition . The Global Genome ( GG- ) NER subpathway specifically involves the XPC and XPE protein complexes , and probes the entire genome for lesions that disrupt base-pairing [5] , [7]–[9] . Transcription-Coupled ( TC- ) NER , on the other hand , detects helix-distorting lesions that stall transcription in the transcribed strand of expressed genes , and hence enables resumption of transcription . TC-NER is independent of XPC and XPE and specifically involves proteins such as CSA , CSB and UVSSA [8] , [10] , [11] . After lesion recognition , the subsequent ‘cut-and-patch’ core repair reaction encompasses local opening of the DNA helix and lesion verification , performed by the TFIIH complex together with XPA . Both correctly position the structure-specific endonucleases ERCC1/XPF and XPG for strand-specific excision of the lesion as part of a 22–30 bp oligonucleotide [5] , [7] , [12] . Finally , the gap is filled by repair synthesis and closed by ligation [5] , [7] , [12] . Multiple NER proteins have been attributed additional roles , both in DNA repair pathways other than NER , and in transcription regulation . For instance , TFIIH is an essential component of the general transcription machinery [13] , [14] , but also other NER factors , including XPG and CSB , have been implicated in transcription regulation [15]–[18] . The 5′ endonuclease ERCC1/XPF participates in the repair of interstrand crosslinks [19] , [20] and subpathways of DNA double-strand break repair [21] . XPC , CSB , and XPG have been individually implicated in promoting base excision repair ( BER ) of oxidative DNA damage [22]–[30] . TFIIH and XPG are , together with CSB , thought to be involved in the early steps of Transcription-Coupled Repair ( TCR ) , and XPG interacts directly with both CSB and RNA Polymerase II [31] . Although still controversial , there are accumulating reports that TCR not only directs NER to blocked transcription but may also recruit BER for preferential repair of oxidative DNA damage in transcribed strands [32]–[34] . Such a mechanism might be related to the roles of both CSB and XPG in promoting BER more globally . If correct , it could explain the much greater consequences for the organism of TCR defects compared to defects in NER alone ( see below ) [35] . A number of rare , autosomal recessive disorders resulting from mutations in NER genes underscore the importance of genome maintenance for the prevention of cancer as well as aging [3] . NER-associated diseases are characterized by sun ( UV ) hypersensitivity and include xeroderma pigmentosum ( XP ) , UV-sensitivity syndrome ( UVSS ) , Cockayne syndrome ( CS ) , cerebro-oculo-facio-skeletal ( COFS ) syndrome , XPF-ERCC1 ( XFE ) progeroid syndrome , trichothiodystrophy ( TTD ) and disorders that combine the symptoms of these syndromes , including XP/CS [35]–[39] . XP originates from defects in GG-NER or total NER activity and is characterized by an over 2000-fold increased risk of cancer in sun-exposed skin and , to a much lesser extent , in internal organs [36] . XP patients may also develop progressive neurological symptoms and neuronal degeneration depending on the severity of the total NER deficiency [36] , [39] , [40] . UVSS is characterized by skin UV hypersensitivity without actually developing skin cancer . UVSS results from the selective loss of TC-NER function as a consequence of mutations in the proteins involved in detection of UV-induced transcription-blocking DNA lesions , i . e . UVSSA , CSA , and CSB [11] , [35] , [41]–[44] . Mutations in CSA and CSB , however , generally cause CS , a heterogeneous multisystem disorder that , in addition to UV-sensitivity , is characterized by severe growth failure and cachexia , accelerated aging features , short lifespan , and progressive sensori-neuronal abnormalities [38] , [45] . The severe symptoms of CS cannot be explained by the loss of TC-NER function as they do not occur in fully NER-deficient XP patients and TC-NER deficient UVSS patients . Therefore , CS symptoms have been linked to additional , yet incompletely , defined functions of CSA and CSB in DNA repair , transcription regulation , other processes , or a combination of deficiencies [3] , [46] , [47] . The same applies for mutations in the down-stream NER factors XPB , XPD , XPF , ERCC1 and XPG that cause combined XP/CS , or severe developmental/degenerative multisystem disorders such as COFS and XFE that share multiple features with severe CS forms [35] , [48]–[51] . Thus CS symptoms can result from mutations in multiple proteins that operate together in NER , but the symptoms caused by these mutations cannot be explained by NER deficiency alone , raising questions about the identities of these non-NER activities underlying CS symptoms and the extent to which different symptoms reflect deficits of different cellular processes [35] , [46] . Mutations in the structure-specific NER 3′-endonuclease XPG are rare , with less than 20 patients and 25 mutant alleles described so far [52]–[55] , and cause a spectrum of disease phenotypes varying from XP to XP/CS and COFS [53] . Point mutations that selectively eliminate XPG nuclease activity cause XP , while C-terminal truncations , destabilizing point mutations , and mutations that abolish the interaction between XPG and the basal transcription factor TFIIH cause XP/CS and COFS [52]–[58] . These data support the notion that a deficient function of XPG outside NER is responsible for the severe CS symptoms [15] , [52] , [53] , [55]–[57] . For most NER disorders , mouse mutants have been generated that mimic the genetic defect found in patients , and to various extents reproduce XP and CS-like features as well as the progeroid hallmarks found in the corresponding human syndrome [59]–[62] . Accordingly , Xpg-null ( Xpg−/− ) mice were found to develop a severe phenotype characterized by growth deficiency and very short lifespan , resembling severe XP/CS [63] . In contrast , Xpg mutant mice carrying amino acid substitutions that selectively abolish the nuclease function of XPG ( XpgE791A and XpgD811A ) show severe UV-sensitivity but normal lifespan , hence , reproducing the XP phenotype [64] , [65] . In addition , a mutant XPG construct containing a C-terminal truncation lacking the last 360 amino acids that was made to mimic the genotype of some XP-G/CS patients , developed a growth deficiency and short-living phenotype resembling that of Xpg−/− mice , albeit somewhat milder [65] . Yet another C-terminal truncation mutant lacking the last 180 amino acids showed a normal lifespan , but produced a CS-like growth-deficient short-living phenotype after crossing with Xpa−/− mice that are already fully NER-deficient [65] , [66] . Significantly , the same conversion of a normal lifespan into a short-living mouse model is observed after crossing CSA- or CSB-deficient CS mice with total NER- ( Xpa−/− ) or GG-NER ( Xpc−/− ) deficient mouse models [67]– , but not by crossing NER-deficient XpgD811A with Xpa−/− mice [66] . Together these data indicate a deleterious synergistic interaction between NER deficiency and loss of non-NER activities that underlie CS . Furthermore , they show that the C-terminus of XPG could play a role in the CS symptoms , and that the XPG-deficient Xpg−/− mice may reproduce the phenotype of Xpa−/−Csb−/− , Xpc−/−Csb−/− or Xpa−/−Csa−/− double mutant mice [62] , [66]–[69] . In previous analyses we clearly observed progeroid characteristics in many NER mutant mouse models including Xpa/Csb , Xpb , Xpd , and Ercc1 mutants [51] , [68] , [70]–[72] , yet the occurrence of progeroid features in Xpg−/− mice has hitherto been poorly established , mostly due to their very short lifespan . Since we are particularly interested in the effect of Xpg deletion on organ-specific aging , we generated a conditional Xpg mutant . As genetic background can have a significant effect on phenotype development , we first re-examined the pathological characteristics of Xpg−/− mice in a C57BL6/FVB F1 hybrid background , as was previously described for Ercc1 mutant mice [51] , [72] . In this hybrid background Xpg−/− mice lived longer and presented progeroid features including cachexia and osteoporosis with pronounced degenerative phenotypes in both liver and brain . We next studied the effect of liver- and forebrain-specific inactivation of Xpg , showing that the observed phenotypes are indeed due to lack of XPG protein . Together our data show that , consistent with previous data in ERCC1- and CSB/XPA-deficient mice , Xpg−/− mice develop a multisystem progeroid degenerative phenotype .
To generate a Cre-inducible Xpg knockout allele we flanked the third exon of Xpg with LoxP sites ( Figure 1A ) . Deletion of this exon causes a frame shift and a premature translational stop immediately after exon 2 at amino acid residue 89 ( instead of the full-length 1170 ) . After transfection to 129 ES cells and selection of properly targeted clones ( Figure 1B and Materials and Methods ) , two independent transfected clones were used to generate germ-line transmitting chimeras ( Figure 1C ) . Heterozygous males , carrying the conditional Xpg allele , were crossed to females ubiquitously expressing Flp for excision of the Neomycin cassette and to yield mice that are heterozygous for the floxed Xpg ( Xpgf ) allele . Xpgf/+ mice were backcrossed and maintained in FVB/N background . To generate Xpg mice carrying a knockout allele ( Xpg− , Figure 1A ) , Xpgf/+ mice were crossed to Cag-Cre mice , which ubiquitously express Cre recombinase from germline [73] , yielding heterozygous Xpg+/− animals ( Figure 1C ) . Xpg+/− animals were>10 times backcrossed into C57BL6 or FVB/N genetic backgrounds . Unless otherwise stated , experiments were performed with Xpg−/− mice in the C57BL6/FVB F1 hybrid background obtained from intercrossing C57BL6 Xpg+/−×FVB/N Xpg+/− animals to minimalize background specific effects ( see below ) . The presence of a premature stop codon in the Xpg− allele was confirmed by sequencing Xpg cDNA from liver of Xpg−/− mice ( Figure S1A ) . Accordingly , Western immunoblot analysis with an antibody raised against the central spacer region ( R-domain ) of XPG shows the absence of XPG protein product in mouse dermal fibroblasts ( MDFs ) isolated from Xpg−/− mice ( Figure 1D ) . Next , we tested DNA repair deficiency of Xpg−/− MDFs . In accordance with complete NER deficiency , Xpg−/− MDFs showed an almost 10-fold hypersensitivity to UV , similar to fully NER-defective MDFs derived from Xpa−/− mice ( Figure 1E ) [74] . In addition , Xpg−/− MDFs were hypersensitive to treatment with Illudin S ( Figure 1E ) , consistent with the loss of TC-NER function [75] , and were deficient in UV-induced unscheduled DNA synthesis in line with loss of GG-NER activity ( Figure 1F ) . Also , recovery of RNA synthesis after UV exposure was almost completely abolished in Xpg−/− MDFs , further demonstrating loss of TC-NER activity ( Figure 1G ) . Xpg−/− MDFs showed no increased sensitivity to potassium bromate ( KBrO3 ) which causes oxidative DNA lesions , and a minimal increased sensitivity to the cross-linking agent cisplatin ( Figure S1B ) . In view of a significant effect of genetic background on embryonic lethality and lifespan in ERCC1-deficient mice [51] , [72] , we examined whether a similar genetic background effect occurred in Xpg−/− mice , by comparing birth frequencies and lifespan of Xpg−/− mice in a C57BL6 , FVB/N or a C57BL6/FVB F1 hybrid background . In C57BL6 background birth frequencies were below Mendelian expectations ( ∼8% , Table 1 ) , whereas in the FVB/N and C57BL6/FVB F1 hybrid background birth frequencies were Mendelian and near-Mendelian , respectively ( Table 1 ) . Also the lifespan of Xpg−/− animals was strongly dependent on genetic background , with C57BL6 Xpg−/− mice showing a lifespan of 3 weeks , and Xpg−/− animals in FVB/N and C57BL6/FVB F1 hybrid background living for 15–18 weeks ( Figure 2A ) . Further analysis of C57BL6/FVB F1 Xpg−/− mice showed that they had the same size and weight as wild type and heterozygote littermates at late embryonic stage ( E17 . 5; Figure 2B and C ) . However , after birth , Xpg−/− mice showed reduced growth and weight gain compared to controls , and stopped growing at 6–8 weeks when their body weight was about 65–70% of that of wild type littermates ( Figure 2C ) . From 10–11 weeks , body weights declined and the Xpg−/− mice became progressively cachectic ( Figure 2C and D ) . At 14 weeks all Xpg−/− mice were severely runted ( Figure 2D ) , and the mice died a few weeks thereafter between 15–18 weeks of age ( Figure 2A and E ) . The growth deficiency was paralleled by the development of kyphosis ( Figure 2D and F ) . In addition , Xpg−/− mice progressively developed neurological symptoms , including clasping of the hind-limbs when lifted by their tails ( Figure S2A ) , and at a later time point fine tremors ( Figure 2E ) . Accelerating rotarod and grip strength tests in 14-week old Xpg−/− mice revealed severe motor deficits and muscle weakness at this age ( Figure S2B and C ) . Computed tomography ( CT ) confirmed severe kyphosis in Xpg−/− mice at 16 weeks of age ( Figure 2F ) . To further examine skeletal abnormalities and the occurrence of osteoporosis as observed in other NER-deficient mouse models [68] , [76]–[79] , we measured several bone parameters using femoral bones . Analysis of bone strength revealed decreased strength of the Xpg−/− femoral bone at 14–16 weeks ( Figure 2G ) . Micro-CT analysis showed that the thickness of the trabeculae and cortex of the femoral bones was significantly smaller in Xpg−/− compared to wild type mice at 14–17 weeks , but not yet at 7 weeks ( Figure 2H ) . Overall , these data indicate a progressive increase of age-related features such as osteoporosis . Weight loss and reduced size of Xpg−/− mice was associated with reduced weight of internal organs ( Figure S3A ) and with a strong reduction in the amount of subcutaneous fat ( Figure S3B ) . The Xpg−/− mice previously reported by Harada et al . [63] showed developmental abnormalities of the gastro-intestinal tract . These gastro-intestinal abnormalities were proposed to be a major contributor of the post-natal growth failure and short lifespan ( <3 weeks ) of their animals [63] . However , in contrast to their data , the gastro-intestinal tract of our Xpg−/− mice had a normal size and macroscopic appearance , and showed a normal histological appearance in HE-stained sections ( Figure 3A ) . Furthermore , staining for the proliferative cell marker Ki-67 indicated that the number of proliferative cells in the intestinal epithelium was similar between Xpg−/− and wild type mice ( Figure 3A ) . In accord with normal function of the gastro-intestinal tract we found that food intake per gram body weight was similar between wild type and Xpg−/− animals ( Figure S3C ) . The liver is a central organ in many aspects of metabolic control , including regulation of circulating glucose levels and detoxification , and it plays a key role in regulation of IGF1-somatotrophic axis signaling . Previous studies have shown that ERCC1/XPF-deficient mice develop multiple liver abnormalities , in particular anisokaryosis resulting from polyploidy , and intranuclear inclusions [72] , [80]–[83] . Analysis of HE-stained liver sections of our Xpg−/− mice revealed mild anisokaryosis , and increased mean nuclear size at 14 weeks , but not at 4 weeks of age ( Figure 3B ) . In addition , sporadically , hepatocytes had intranuclear inclusions . These liver nuclear changes are a well characterized phenomenon in the aging liver , and indicate that Xpg−/− mice show features of accelerated aging in the liver similar to Ercc1Δ/− mice [72] , [82] . Liver cells of ERCC1-deficient and other progeroid NER-deficient mouse mutants display changes in gene expression that encompass a downregulation of catabolic and oxidative metabolism and an upregulation of antioxidant and stress defense pathways , suggestive of a compensatory survival response to cope with increased DNA damage [51] , [68] , [84] , [85] . To determine whether Xpg−/− liver cells also display a ‘survival-like’ stress response , we determined expression levels of selected antioxidant and somatotrophic genes by real-time PCR . Indeed , mRNA levels from a subset of antioxidant effector genes , including Nqo1 , Srxn1 , Gstt2 and Gsta1 , were significantly increased in liver homogenates of young ( 7-week old ) Xpg−/− animals compared to controls . At 14 weeks of age , we observed a similar significant increased expression of Nqo1 and Gsta1 while mRNA from the other antioxidant effector genes tested showed unaltered expression ( Figure 3C ) . Expression levels of Nrf2 , which is a potent inducer of the antioxidant response element ( ARE ) , were unaltered , in line with the fact that NRF2-activation is largely achieved by post-translational mechanisms [86] . As increased expression of antioxidant genes could be an indication of increased genotoxic stress , we also checked the expression of the p53-responsive kinase inhibitor p21 , a master regulator of cell survival and death [87] , which is generally increased after DNA damage and was previously shown to be elevated in livers of Ercc1 mutant mice [88] . Expression levels of p21 doubled at the age of 7 weeks and were massively increased at the age of 14 weeks , indicative of genotoxic stress caused by the absence of XPG . To determine changes in somatotrophic gene expression we examined mRNA levels of Ghr , Igf1r , Igf1 and Igfbp3 . We found a two-fold suppression of Ghr and Igf1r mRNA expression at 7 weeks , and a significant downregulation of Ghr mRNA levels at 14 weeks of age ( Figure 3D ) . Together the data indicate that the Xpg−/− liver in part reproduces gene expression changes observed in other short-living NER-deficient mice , which we refer to as a survival-like stress response . Finally , consistent with other NER-deficient progeroid mice [51] , [85] , we found significantly reduced steady-state blood glucose levels in Xpg−/− mice ( Figure 3E ) . The occurrence of neurological abnormalities and impaired motor behavior in Xpg−/− mice ( Figure 2E and S2 ) , as well as the abundant neurodegenerative features in ERCC1-deficient and combined XP/CS mouse models [89]–[91] , prompted us to investigate the central nervous systems of Xpg−/− animals for neurodegenerative changes . Macroscopically , the brains and spinal cords of Xpg−/− mice showed a normal appearance , albeit somewhat smaller . In addition , the gross histological organization analyzed in thionin-stained sections appeared normal in all brain regions . As a first step to examine the occurrence of neurodegenerative changes , we examined the brains of 4- and 14-week old Xpg−/− mice immunohistologically for glial acidic filament protein ( GFAP ) expression , which outlines reactive astrocytosis in response to neuronal injury . A mild increase in GFAP immunostaining occurred in patches in multiple nervous system areas at 4 weeks ( Figure 4A and S4A ) . Instead , at 14 weeks , Xpg−/− mice showed a prominent ubiquitous increase in GFAP staining throughout the entire central nervous system including spinal cord , indicative of widespread astrocytosis ( Figure 4A and S4A ) . Double-labelling of GFAP and the microglia cell marker Iba-1 showed that the increased GFAP staining was paralleled by microglia activation , characterized by increased Iba-1 immunoreactivity and the transformation of resting microglia cells into activated cells with thicker processes and larger cell bodies ( Figure S4B and C ) . Next , to determine whether Xpg−/− central nervous system cells experience genotoxic stress , we studied the expression of the transcription factor p53 , which is activated by multiple types of DNA damage and is expressed in neurons and macroglia of many NER-deficient mouse models including mice defective in Ercc1 , Csa or Csb [89]–[91] . Immunohistochemistry revealed p53-positive cells in all central nervous system regions . Analysis of the p53 density in neocortex and cerebellum indicated an increase in number of p53-positive cells in brains of 14-week old compared to 4-week old Xpg−/− mice ( Figure 4B ) . Similar to our findings in other NER mutant mice [89]–[91] , double labelling of p53 with neuronal ( NeuN ) and astrocytic ( GFAP , S100β ) markers , indicated that these p53-positive cells include neurons , astrocytes ( GFAP+ or S100β+; Figure S4D ) , and oligodendrocytes . Although not systematically investigated , we also noted that , as in other NER-deficient mice , in neocortex and cerebellar cortex the majority of p53-positive cells were neurons , while in spinal cord a large proportion of p53-positive cells were astrocytes ( Figure S4E ) . To obtain evidence for the occurrence of neuronal death , we analyzed calbindin staining in cerebellar cortex where it outlines Purkinje cells and enables easy detection of the degeneration of these cells [90]–[92] . Calbindin staining revealed degeneration and loss of Purkinje cells in 14-week old Xpg−/− mice ( Figure 4C ) . Also , calbindin staining revealed sporadic Purkinje cells with abnormal dendritic morphologies and , more frequently , Purkinje cells with swellings in their proximal axon ( Figure 4C ) . Axonal swellings ( also designated torpedoes or axonal spheroids ) are a common feature in neurodegenerative disorders and aging [93] , that is also well documented for Purkinje cell axons of CS and XP/CS patients [94] . The presence of axonal pathology indicates that many surviving Purkinje cells in 14-week old Xpg−/− mice display compromised health . Notably , few small axonal swellings occurred in Purkinje cells axons in 4-week old Xpg−/− mice . To further examine the extent to which neurons in Xpg−/− mice show compromised health , we examined the morphology of the Golgi apparatus in motor neurons . In a previous study in Ercc1 mutant mice we noted that motor neurons displayed a variety of morphological abnormalities of the Golgi apparatus , and we proposed that these abnormalities reflect a heterogeneity of cellular deficits resulting from stochastic DNA damage [90] . Immunostaining for the cis-Golgi marker GM130 showed that motor neurons in Xpg−/− mice developed the same heterogeneity in morphological abnormalities of the Golgi apparatus as observed in Ercc1Δ/− mice . Double labelling of GM130 and p53 indicated that only a small subset of neurons with abnormal Golgi apparatus is p53 positive . This variability in p53 expression further illustrates the heterogeneity of degenerative events that may occur in Xpg−/− neurons ( Figure S4F ) . TUNEL staining to determine the amount of apoptotic cells showed a significant increase in both the cerebrum and the cerebellum at 4 as well as 14 weeks of age ( Figure 4D ) . Finally , real-time PCR in Xpg−/− cerebellum revealed an upregulation of the p53-responsive kinase inhibitor p21 consistent with the activation of genotoxic stress pathways ( Figure 4E ) , and increased expression of several oxidative stress response genes ( Figure 4E ) . In addition to the brain and spinal cord , we also investigated the retina , as retinal degeneration is a frequent symptom of CS and XP/CS patients [38] , that is also reproduced in CSA- and CSB-deficient mice [23] . TUNEL staining revealed cell death in both the inner and outer nuclear layers of the retina of 4- and 14-week old Xpg−/− mice ( Figure 4F ) . Hence , Xpg−/− mice display loss of photoreceptor cells as well as degeneration of the retinal circuitry . Together these data indicate the occurrence of widespread progressive degenerative changes in Xpg−/− nervous system , strongly resembling the phenotype of ERCC1-deficient mice . Transgenic expression of ERCC1 in the liver has been shown to alleviate growth deficiency and to extend lifespan of ERCC1-deficient mice [83] , suggesting that liver abnormalities are an important determinant of the reduced lifespan of these mice . To determine the importance of liver pathology in the runting and the reduced lifespan of our Xpg−/− mice , we generated mice with liver-specific inactivation of the Xpg gene by crossing Xpgf/+ mice carrying the floxed Xpg allele with heterozygous Xpg+/− mice that also express the albumin-Cre recombinase transgene that drives Cre expression specifically in hepatocytes [95] to yield Xpgf/−/Alb-Cre+ mice , hereafter designated Alb-Xpg mice . This Alb-Xpg mouse also has the advantage that it allows to study the effect of liver XPG-deficiency in the absence of abnormalities in other tissues . A cohort of Alb-Xpg mice was allowed to reach the age of one year . All Alb-Xpg mice displayed normal growth and weight gain , and none of them died prematurely ( Figure 5A ) . Livers of Alb-Xpg mice had an increased size compared to wild type , while brain , kidney and spleen displayed unaltered size and weight ( Figure S5A ) . Albumin and glucose blood levels were the same as in control mice ( Figure S5B and C ) . Histological analysis revealed anisokaryosis with karyomegaly in the liver of Alb-Xpg mice analyzed at 26 and 52 weeks ( Figure 5B ) . The observed karyomegaly was more prominent than that observed in 14-week old Xpg−/− mice , and cells with intranuclear inclusions were more frequent . In addition , we identified p53-positive cells , increased cell death and increased cell proliferation in Alb-Xpg liver consistent with a progeroid degenerative phenotype ( Figure 5C–E ) . Furthermore , real-time PCR showed that livers of Alb-Xpg mice displayed a massive induction of the DNA damage response gene p21 as well as increased expression of several antioxidant effector genes ( Figure 5F ) . We also observed a trend of reduced expression of Ghr and Igf1r ( Figure 5G ) , hence , reproducing gene expression changes determined in livers of Xpg−/− mice . Activation of the Nrf2 antioxidant response genes , reduction of the IGF1 axis , and increased proliferation shown by Ki67-staining are all consistent with liver regeneration after tissue damage [96] . As an additional control we showed that the expression of these genes is unaltered in livers from Emx1-Xpg mice that are XPG-deficient in the dorsal forebrain ( see below; Figure S5D and E ) . Together the data from the Alb-Xpg mice show that progeroid and gene expression changes in the liver triggered by the absence of XPG are not sufficient to explain the runted short-living phenotype of Xpg−/− mice . Our neuropathological analyses of Xpg−/− mice uncovered severe neurodegenerative changes at 14 weeks of age , compatible with neurological and motor deficits in these mice . Importantly , the presence of p53 in glia and abundant astrocytosis and microgliosis in the white matter indicate that abnormalities in Xpg−/− mice are not limited to neurons , but also involve glia cells . This is consistent with our findings in CS mouse models [62] , [91] , and with the neuropathological changes found in CS and XP/CS patients that is dominated by white matter pathology , in addition to neuronal , glial and vascular pathology , and , in severe cases , developmental abnormalities [39] , [94] , [97]–[99] . In previous studies , using a Cre-lox approach with CamKIIα-Cre and L7-Cre transgenic mice that drive Cre expression in post-mitotic forebrain neurons and Purkinje cells , respectively , we showed that neuron-specific deficiency of ERCC1 or combined deficiency of XPA and CSB was sufficient to trigger stochastic degeneration of these neuronal populations [89] , [91] , [92] . These studies showed that neurodegenerative changes in ubiquitous ERCC1- and XPA/CSB-deficient mice are not a consequence of developmental abnormalities , vascular problems , or degenerative changes in other organs . Furthermore , these neuron-specific mice enabled us to follow the degenerative process beyond the normal lifespan of the short-living ubiquitous ERCC1- and XPA/CSB-deficient mice [89] , [91] , [92] . In the present study we therefore used a similar approach , but with an Emx1-Cre transgenic line that drives Cre expression in progenitor cells of the dorsal telencephalon , to achieve inactivation of the Xpg gene not only in excitatory neurons of the neocortex and hippocampus , but also of astrocytes and oligodendrocytes in these brain areas [100] . Analysis of a cohort of Emx1-Xpg mice that were allowed to age for one year revealed no early death and showed that body weights were indistinguishable from that of wild type littermates until the age of 30 weeks . At older age the mean weight gain of Emx1-Xpg mice was significantly smaller than in control littermates ( Figure 6A ) . This difference in weight was associated with a proportional reduced weight of internal organs ( Figure S6A ) . Basal blood glucose concentrations were the same as in controls , indicating that reduced weight of old Emx1-Xpg mice is not a consequence of reduced energetic intake ( Figure S6B ) . To obtain a crude impression of the development of neurological symptoms , we examined the time of onset of clasping of the hind limbs when animals were lifted by their tails . This abnormality developed in Emx1-Xpg between 14–24 weeks of age and was not observed in control littermates ( Figure 6B ) . The Emx1-Xpg mice did not develop tremors and deficits in accelerating rotarod performance ( Figure S6C ) . Both the absence of these motor deficits and the delayed onset of clasping in comparison to the Xpg−/− mice can be explained by the selective inactivation of Xpg in neocortex and hippocampus , avoiding the bulk of circuitries controlling motor behavior in mice [101] . Macroscopic inspection of brains of Emx1-Xpg mice at 26- and 52-week already revealed that the neocortex was considerably smaller . Histological analysis confirmed that the neocortex was thinner , and showed that also the hippocampus was dramatically smaller , while other brain regions were unaltered ( Figure S6D ) . GFAP immunohistochemistry showed a very strong increase in GFAP staining indicative of astrocytosis in both cortex and hippocampus ( Figure 6C ) , and no changes in other brain regions . As in the Xpg−/− nervous system , astrocytosis was paralleled by microgliosis , identified by immunohistology with an Iba-1 antibody . Staining for Mac2 ( also known as galectin-3 ) , to outline phagocytosing microglia cells [102] , revealed very high levels of Mac2-positive cells in the corpus callosum and the fimbria fornix of Emx1-Xpg mice ( Figure 6D and S6D , E ) , and a moderate amount of phagocytosing microglia in the neocortex and hippocampus ( Figure 6D ) . Remarkably , the presence of Mac2 could not be explained by axonal degeneration of cortical and hippocampal neurons solely , as we did not observe this phenomenon in our CamKIIα-Ercc1 mice [89] . Furthermore , the capsula interna of Emx1-Xpg mice , which contains the descending corticofugal axons and shows severe axonal degeneration , did not show this dramatic increase of Mac2 staining ( Figure 6D and S6D ) . Hence , the presence of high levels of Mac2 labelling may reflect the same phenomenon that we observed in our CS mice , i . e . the presence of phagocytosing microglia in the absence of axonal degeneration [91] . Accordingly , we also found an upregulation of Hsp25 expression in astrocytes in the corpus callosum and fimbria fornix ( Figure S6F ) , a phenomenon that we also observed in the white matter of CS mouse models [91] . Furthermore , analysis of p53 expression in Emx1-Xpg showed that multiple p53-positive cells populated the fimbria-fornix and the corpus callosum ( Figure 6E ) . Double-labelling of p53 with NeuN or the glia marker GFAP and S100β showed that some p53-positve cells were neurons ( NeuN ) , but a large proportion were astrocytes ( GFAP+ , S100β+ ) , or oligodendrocytes as identified on the basis of nuclear morphology ( Figure 6F ) . Together the data indicate that Emx1-Xpg mice develop a combined neuro- and gliopathy of cortex and hippocampus .
The main goal of the present study was to determine the extent to which XPG deficiency in mice results in multisystem progeroid degenerative changes as observed in CS and XP/CS patients , as well as in other NER-deficient mouse models including Xpa/Csb , Xpd , and Ercc1 mutants [48] , [51] , [68] , [69] , [72] , [77] , [107] . Our data show that , although seemingly normal at late embryonic stage , and showing only mild degenerative features at 4 weeks of age , C57BL6/FVB Xpg−/− at 14–16 weeks showed abundant degenerative changes in multiple tissues , indicative of accelerated aging . These degenerative/accelerated aging features included kyphosis , cachexia , osteoporosis , liver aging , and abundant nervous system degenerative changes . In addition , the data from our liver- and dorsal forebrain-specific Xpg mice showed that a much more severe tissue-specific degenerative phenotype could be obtained in these mice because of survival beyond the maximal lifespan of Xpg−/− mice . Hence , the data indicate that Xpg−/− mice indeed develop a multisystem premature aging phenotype reminiscent of the phenotypes of some of our other NER-deficient mouse models , in particular the Ercc1Δ/− mice , which show a slightly longer lifespan and comparable pathological changes in liver and nervous system ( see below ) . Moreover , the Xpg−/− mouse phenotype shares many features with XP-G/CS patients [17] , [94] , [108] including a cachectic ‘frail’ appearance with loss of subcutaneous fat and signs of osteoporosis ( Table 2 ) . Previously we have shown that transcription-blocking lesions trigger a “survival response” involving somatotroph attenuation and increased resistance to oxidative stress , which resembles the response triggered by dietary restriction and is associated with delaying many aspects of aging and increasing lifespan [51] , [68] , [84] . This mechanism has been observed in response to direct but persistent DNA damage as well as during the course of aging , both in natural aging as well as in specific progeroid mouse models , and is activated in long-lived mutant mice and centenarians [51] , [68] , [109] . A central hallmark of this survival response is the downregulation of genes involved in the GH/IGF axis , combined with upregulation of genes involved in the antioxidant response . Using real-time PCR for selected genes of these pathways in the present study , we show that Xpg−/− mice display survival-response-like changes in gene expression in liver cells , including a decrease in Ghr mRNA levels and a robust increase in expression of the antioxidant defense effector genes Nqo1 , Srxn1 , Gstt2 , and Gsta1 . Similar changes were observed in the liver of liver-specific XPG-deficient mice . Consistent with previous observations concerning the survival response , we found significantly reduced circulating glucose levels in the Xpg−/− mouse . However , we found no change in glucose in the liver-specific mouse , indicating that reduced glucose levels are not necessarily a consequence of gene expression changes in the liver . We did not observe significantly lower levels of Igf1 and Igf1r , as previously found for several other progeroid DNA repair mutants [51] , [84] , although we did find a trend of reduced expression . Increased expression of antioxidant genes was also observed in the central nervous system ( i . e . cerebellum ) of Xpg−/− mice , but the identity and the degree of changed expression is somewhat different . For instance , Nrf2 shows a relatively increased expression in cerebellum but not in liver , while Nqo1 shows a very large relative increase in liver , and a modest relative increase in cerebellum . These differences may be explained by altered stress responses in different cell types [110] . In addition , in the nervous system , changes in gene expression may result from the death and disappearance of neurons and reactive proliferation of glial cells and as a consequence , reduced and increased expression of neuronal and glial genes , respectively . The liver of Xpg−/− mice showed several characteristics also found in aging liver , i . e . increased nuclear size and the presence of nuclear inclusions . These changes were much more prominent and associated with additional degenerative changes in liver-specific XPG-deficient mice . However , despite prominent liver pathology , the liver-specific Xpg−/− mice at 52 weeks did not show altered weight , nor altered glucose and albumin levels suggestive of metabolic problems . Although at this point we cannot exclude that our liver-specific mutants will develop health problems in their second year of life and might develop a shorter than normal lifespan , it is safe to conclude that liver problems on their own cannot be the main culprit of the reduced lifespan and the small cachectic appearance of Xpg−/− mice . In contrast to our data , findings in Ercc1 mutant mice show that transgenic overexpression of ERCC1 in the liver alleviates growth deficiency and extends lifespan [83] , indicating that liver abnormalities are an important determinant of the reduced lifespan of these mice . A possible explanation is that liver pathology is more prominent in Ercc1 mutant mice , putatively as a consequence of additional defects in interstrand crosslink and double-strand break repair [48] . An alternative explanation for the differences in results could be that the impact of liver pathology on survival depends on the severity of pathology in other organs . This hypothesis is testable by studying survival of liver-specific ERCC1-deficient mice , and vice versa by generating liver-corrected Xpg−/− mice using a transgenic ‘rescue’ strategy as reported by Selfridge et al . [83] . A possible conclusion from these studies could be that Ercc1 and Xpg mutant mice die prematurely as a consequence of a synergistic deleterious effect of multi-organ failure . In view of the severe neurological and neurodegenerative deficits in 14–16 week old Xpg−/− mice , yet an alternative scenario is that Xpg−/− mice die as a consequence of nervous system abnormalities ( see below ) . This scenario is realistic for a subset of human NER-deficient patients , in particular those developing XP with neurological abnormalities [36] , [40] . Interestingly , our dorsal forebrain-specific XPG-deficient mice also showed reduced weight-gain after 30 weeks of age . At this age these mice displayed severe neuronal degeneration in neocortex and hippocampus , as a result of which the mice would be expected to have severe cognitive deficits , whereas basal motor functions remain unaltered . Weight loss was not associated with altered glucose levels or gene expression changes in the liver ( Figures S5 and S6 ) . We noted the same reduced weight-gain in our forebrain neuron-specific XPA/CSB-deficient mouse model [91] . The data suggest that severe disruption of cortical or hippocampal circuitries may result in the weight loss , but the precise mechanism remains to be defined . Our data show that Xpg−/− mice , within their relatively short lifespan ( 16–18 weeks ) , develop neurological abnormalities of increasing severity in association with degenerative changes throughout the nervous system . These nervous system degenerative changes seemingly are more severe than those previously reported for another line of Xpg−/− mice [111] . The differences in nervous system neurodegenerative changes can be explained by the very short lifespan ( 3 weeks ) of previously reported Xpg−/− mice , since our C57BL6 Xpg−/− mice , which show the same short lifespan , also displayed a low level of nervous system pathology at the end of life ( see above and Figure S7 ) . Together with our demonstration that dorsal forebrain-specific XPG-deficient mice , allowed to age for one year , display very severe neurodegenerative changes , our data indicate that age is a key determinant in the development of neurodegenerative changes in Xpg−/− mice . Our data also indicate that XPG deficiency does not result in obvious neurodevelopmental abnormalities , although subtle developmental abnormalities cannot be excluded at this point . The neurodegenerative changes in our Xpg−/− mice strongly resemble those of incomplete ERCC1-deficient Ercc1Δ/− mice that have a slightly longer lifespan ( 24–30 weeks ) . Our studies with Ercc1Δ/− mice and neuron-specific Ercc1 knock-out animals indicate that ERCC1-deficient neurons in time stochastically accumulate structural and functional deficits to eventually die and disappear [89] , [90] , [92] , [112] . A stochastic accumulation of deficits also appears to occur in Xpg−/− neurons . Thus , the widespread distribution of astrocytosis and microgliosis , as well as p53 and TUNEL staining , indicates that degenerative changes in the Xpg−/− nervous system affect all neuronal populations . The increased size and frequency of axonal spheroids in Purkinje cells , and a diversity of Golgi apparatus morphological abnormalities in motor neurons , on the other hand , illustrate that Xpg−/− neurons may asynchronously accumulate a variety of degenerative features over time . This widespread and asynchronous accumulation of cellular damage in Xpg−/− neurons is consistent with a model in which neurons are afflicted by stochastic DNA lesions that deregulate gene expression , as we have proposed for Ercc1Δ/− mice [89] . A prime role for genotoxic stress in causing the degenerative phenotype in the Xpg−/− nervous system is further suggested by the presence of stochastically distributed p53-positive cells and increased expression of the DNA damage responsive p21 gene . Together the degenerative changes in the Xpg−/− nervous system favor a pathogenic model involving deficient DNA repair in the same way as proposed for ERCC1-deficient mice [48] , [89] . The central role of DNA damage in the Xpg−/− nervous system raises questions about the identity of the DNA lesions involved . Importantly , the absence of a significant neurodegenerative phenotype in entirely NER-deficient Xpa−/− mice has led to the conclusion that DNA lesions that accumulate in the nervous system in the absence of NER are not sufficient to trigger neuronal degeneration within the normal lifespan of mice [74] , [91] , [113] . In the case of ERCC1-deficient mice , it therefore has been proposed that the degenerative nervous system changes result from combined deficiencies in NER and other DNA repair pathways , i . e . interstrand crosslink repair and double-strand break repair [48] , [89] . It is possible that a similar mechanism may contribute to the severe neurodegenerative phenotype caused by XPG deficiency , as XPG has been reported to play a role in repair of crosslinks induced by mitomycin C [114] . An alternative possibility is that XPG deficiency reproduces the severe degenerative phenotype resulting from crossing NER-deficient Xpa−/− with the Csb−/− or Csa−/− CS mouse models [67]–[69] , [91] , [106] . In this scenario XPG deficiency combines the deleterious synergistic interaction between NER deficiency and loss of the yet incompletely understood non-NER activities that underlie CS . CS proteins operate together at the interface of DNA repair and transcription regulation , and several mechanisms have been put forward to explain CS symptoms [46] . XPG can interact directly with both CSB and RNA Polymerase II [31] , and may be implicated in the repair of transcription blocking oxidative lesions , for instance by recruiting base excision repair factors [32]–[34] . Accordingly , cells from CS patients including XP-G/CS cells have been found to display increased vulnerability to inducers of oxidative DNA lesions [55] , [115] , [116] . Importantly , Soltys et al . showed oxidative damage sensitivity of XP-G/CS , but not of XP-G patient cells [55] . Yet , in accordance with findings by Harada and coworkers [63] , we found that cells from our Xpg−/− mice do not display increased vulnerability to inducers of oxidative DNA lesions such as KBrO3 ( Figure S1B ) . This is despite clear indications of endogenous damage in , for example , the retina of these mice , as has been previously observed for Csb/Xpa mice that are sensitive to oxidative damage [23] . It is currently unclear whether this is a peculiarity of these particular mouse cells in culture , as we have reported that cells from CSB-deficient mice are sensitive to IR and paraquat [117] , [118] . It has been argued that cultured cells may build up defense responses that mask the increased vulnerability of these cells in vivo [119] . To what extent do Xpg−/− mice reproduce the nervous system abnormalities of patients carrying XPG mutations ? Roughly , the progressive widespread neurodegenerative changes of Xpg−/− mice are reminiscent of neuropathological changes of patients with ‘XP-type neurological degeneration’ [39] , [40] , [97] , [120] . In well documented cases these patients , carrying XPA mutations resulting in complete NER deficiency , develop a wide array of neurological symptoms that show early juvenile onset , over time become more severe , and ultimately cause premature death in mid-adult life [39] , [40] , [97] . However , the limited documented cases indicate that XP-G patients either develop no neurological symptoms , or reproduce mild to severe neurological and neuropathological features of CS [39] , [40] , [57] , [94] , [120] . In CS and XP/CS patients , neuronal degeneration generally is less prominent . Instead , these patients , including documented XP-G/CS cases , show prominent white matter degeneration , vascular pathology , calcium depositions , and , in severe cases , developmental abnormalities [39] , [94] , [97]–[99] . We have recently noted that CSA- and CSB-deficient CS mouse models , in addition to mild neurodegenerative changes , develop subtle white matter abnormalities and glial pathology reminiscent of the glia and white matter degenerative changes of CS patients , albeit milder [62] , [91] . The higher levels of neuronal degeneration in Xpg−/− mice hamper detection of primary glial and white matter pathology , due to secondary glial pathology caused by neuronal degeneration . However , the severe white matter pathology in the corpus callosum and fimbria-fornix in our dorsal forebrain specific XPG-deficient mice strongly indicates that XPG deficiency triggers CS-like white matter pathology in mice . In this study , we show that Xpg−/− mice from young age onwards develop a multisystem degenerative phenotype and die before the age of 20 weeks . This phenotype strongly resembles the progeroid features of CS and XP/CS patients . In addition , the Xpg−/− mouse model shows a number of similarities to other NER-based mouse models of progeria such as Xpa/Csb and Ercc1 mutants [3] , [62] , pointing to the importance of NER in multiple tissues . In particular , a detailed analysis of commonalities and differences between Xpg−/− and Ercc1 mutant mice may aid in our understanding of the contribution of different types of DNA damage and DNA repair defects in the accelerated aging process , since both endonucleases have a joint role in the damage excision step of NER but have divergent additional non-NER roles . Together our findings further stress the relationship between compromised DNA repair and acceleration of specific aging features , as well as progressive neurodegeneration . Finally , the neurodegenerative phenotype indicates that Xpg−/− mice may serve as a model to test intervention strategies aimed at reducing the formation of detrimental DNA lesions in neurons .
The Xpg targeting construct was generated using multiple elements . First , a cassette consisting of a Neomycin ( NEO ) resistance marker , flanked by Frt sites , and followed by a single LoxP site was cloned into a modified pBlueScript SK+ vector containing a PGK-DTA negative selection marker , making use of a klenow blunted ApaI ( insert ) /XbaI ( vector ) and a NotI restriction site . Second , Xpg homologous arms were PCR amplified from C57BL6 genomic DNA ( originating from BAC clone RP24-343K18 ) and cloned into the same plasmid . The following primers ( non-homologous regions indicated in italics; the LoxP sequence is underlined ) were used for amplification of the 5′ and 3′ arm , respectively: LAF2 ( 5′-CGCACCCGGGTGTGATCCTGTGGTCCTGTAGT-3′ ) and LAR2 ( 5′-CCATCGATATCCTCAGAAAGGTATCTCTTAAGCA-3′ ) , yielding a 3 . 2-kb XmaI-ClaI fragment; RAF1 ( 5′-CCCTGCTAGCGGGATGAGGAATCGTGACTAAGGAG-3′ ) and RAR1 ( 5′-CCGCAGCGGCCGCAAACAAGGGACCCAAATGTAGGCT-3′ ) , yielding a 2 . 0-kb NheI-NotI fragment , where the restriction sites were introduced in the PCR primers . Last , the third exon of Xpg followed by a PCR-introduced LoxP site was amplified using the primers Ex3LoxF2 ( 5′-GGGAACCGGTTTGAGTGTCCTTGGTGACAGG-3′ ) and Ex3LoxR2 ( 5′-CCCTGCTAGCATAACTTCGTATAGCATACATTATACGAAGTT ATCC-3′ ) , yielding a 350-bp AgeI-NheI fragment , which was inserted between the neomycin cassette and the 5′ homology arm . Next , a total of 10 µg of NotI-linearized targeting vector was electroporated to Ola129 ES cells , and the targeted clones were selected through the use of the Neomycin selection marker ( G418 200 µg/ml ) . Clones resistant for G418 were initially screened by PCR , using a forward primer in exon 3 ( F3 5′-GAGACAGGCTCTGAAAACTGCTT-3′ ) and a reverse primer outside the 3′ homologous region ( R3 5′-CACTGAACAAACAAGGGACCCAAA-3′ ) . ES clones showing a 2 . 2-kb fragment in addition to the wild type 2 . 3-kb fragment after NheI digestion of the PCR product were further screened by Southern blot . ES genomic DNA was digested with EcoRI and hybridized with a 0 . 9-kb DpnI restriction fragment from BAC RP24-343K18 , spanning the 2nd exon of Xpg . The probe hybridizes to a 7 . 4-kb fragment in wild type DNA and to an additional 4 . 1-kb fragment in targeted DNA . ES cells from two independent targeted clones were micro-injected into C57BL6 blastocysts . Heterozygous mutant mice were generated by crossing the male chimeras with C57BL6 females and verified by coat color and PCR genotyping . The Neomycin ( NEO ) resistance gene was flanked by Frt sites to allow specific elimination of this dominant selectable marker by an Flp recombinase to avoid potential undesired influence of the Neo gene on Xpg transcription or mRNA processing . The NEO cassette was removed by crossing mice carrying the targeted allele with Cag-Flp recombinase FVB/N transgenic animals [121] . These mice carry the floxed allele , and are referred to as Xpgf throughout this paper . Thereafter , the F3 offspring was crossed with a Cag-Cre C57BL6 transgenic [73] , resulting in Cre-mediated recombination and excision of the third exon . Xpg+/− animals were backcrossed to C57BL6 and FVB/N in parallel , at least ten times , and interbred to obtain C57BL6 , FVB/N and C57BL6/FVB F1 hybrid Xpg−/− mice . To achieve liver specific Xpg gene inactivation , a transgenic line with Cre recombinase under the control of the albumin promoter ( hereafter referred to as Alb-Cre ) was used [95] . Female Alb-Cre+ mice were crossed with male Xpg+/− mice ( both in a C57BL6 background ) . Female Xpg+/− Alb-Cre+ mice , obtained from these breedings , were crossed with male Xpgf/f FVB/N mice to yield hybrid Xpgf/− Alb-Cre+ mice . Xpgf/− Alb-Cre+ mice ( in a C57BL6/FVB F1 hybrid background ) are heterozygous for Xpg in their entire body , except for the hepatocytes in the liver , which are homozygous for Xpg after Cre excision of the floxed allele . All littermates , with and without Cre-recombinase expression were used as controls ( referred to as wt ) . FVB/N Xpgf/f animals were similarly bred to the female offspring from C57BL6 Xpg+/− and C57BL6 Emx1-Cre mice [100] to obtain forebrain specific XPG knock-out animals ( referred to as Emx1-Xpg ) . All animals used in the studies described in this paper were of the same C57BL6/FVB F1 hybrid background ( unless otherwise stated ) and had ad libitum access to water and standard mouse food ( CRM pellets , SDS BP Nutrition Ltd . ; gross energy content 4 . 39 kcal/g dry mass , digestible energy 3 . 2 kcal/g or AIN93G synthetic pellets , Research Diet Services B . V . ; gross energy content 4 . 9 kcal/g dry mass , digestible energy 3 . 97 kcal/g ) . Since the Xpg knockout animals were smaller , food was administered within the cages and water bottles with long nozzles were used from around two weeks of age . Experiments were performed in accordance with the Principles of Laboratory Animal Care ( National Institutes of Health publication no . 86-23 ) and with the guidelines approved by the Erasmus University Animal Care Committee . For PCR genotyping the following primers were used: F1 forward primer ( 5′-TCTGTTTAGGTGGTGCCCATTT-3′ ) annealing 5′ of the third exon of Xpg; NeoF forward primer ( 5′-GCTTCCTCGTGCTTTACGGTAT-3′ ) located in the Neomycin resistance marker; R1 reverse primer ( 5′-CGACAGCACTTCTTTCTCCTTAGT-3′ ) annealing 3′ of the third exon of Xpg . A 711-bp fragment was generated from the targeted allele using primers NeoF and R1 , whereas a 495-bp and 227-bp fragment are amplified from the wild type and knockout allele , respectively , using the F1/R1 primerset . Cycling conditions were 95°C for 45 sec , 58°C for 45 sec , 72°C for 1 min ( 35 cycles ) , followed by an extension at 72°C for 5 min . Total RNA was extracted from wild type and Xpg−/− liver using TRIzol reagent and reverse transcribed with SuperScript II Reverse Transcriptase ( Life technologies ) , according to the manufacturer's instructions , to generate cDNA . A 0 . 4-kb PCR fragment ranging from the 2nd to 6th of Xpg was produced using the following primers: Ex2F ( 5′-GCTCATCTTCTCACATTATTCC-3′ ) and Ex6R ( 5′- GGTAAACTCTTTCATGTCAGTC-3′ ) and analyzed by Sanger sequencing . The mice were weighed and visually inspected weekly , and were scored for the onset of various phenotypical parameters . Clasping was measured by suspending mice from their tails for 20 seconds . A clasping event was scored when retraction of both hind limbs towards the body was observed for at least 5 seconds . Whole body tremor was scored if mice were trembling for a combined total of at least 10 seconds , when put on a flat surface for 20 seconds . Mice showing an abnormal curvature of the spine were scored as having kyphosis . Primary MDFs were isolated from the tail of 12–14 week old Xpg−/− animals and wild type littermates . Minced tail skin was immersed in F10/DMEM culture medium supplemented with 20% fetal calf serum , 50 µg/ml penicillin/streptomycin , and 1 . 6 mg/ml type II collagenase ( Gibco , Life Technologies ) . After incubation at 37°C for 24 hours , MDFs were filtered through a 40 µm cell strainer , centrifuged for removal of the collagenase , and cultured at 37°C , 5% CO2 , and 3% O2 . Whole cell extracts were prepared from cultured MDFs using cells isolated from four wild type mice and four Xpg−/− littermates . Proteins from 30 µl of each extract were separated by electrophoresis on 7% SDS-PAGE gels and transferred overnight onto a nitrocellulose membrane . As previously described [31] , XPG protein was detected with a rabbit polyclonal antibody designated R2 ( 97727 ) that was raised against a conserved peptide from the spacer region ( R-domain ) of XPG corresponding to residues 267–281 of the human protein , which are identical with the same residues in mouse XPG except for amino acid 267 , which is E in the human protein but Q in mouse . Whole cell extract from human embryonic kidney 293 cells was used as a positive control for XPG protein . As a loading control , tubulin was detected using a commercial antibody . MDFs were seeded in triplicates at equal densities and treated 24 h after seeding as indicated . After 48 h recovery cell survival was determined by cell count on Beckman Coulter , Z2 Coulter particle count and size analyzer . UDS was performed using the Click-iT EdU imaging kit ( Life technologies ) . MDFs were seeded on coverslips and 24 h later washed with PBS and irradiated with 16 J/m2 UV-C ( Philips ) or mock treated . Cells were directly labelled for 3 h in thymidine-free Ham's F10 medium supplemented with 10% dialyzed serum , 50 µg/ml penicillin/streptomycin , 20 µM Ethynyl-deoxyuridine and 1 µM Fluoro-desoxyuridine . After a PBS wash , the cells were chased with 10 µM non-labelled thymidine in normal medium for 15 minutes and fixed with 3 . 7% formaldehyde . Slides were washed 3× with PBS/3%BSA and permeabilized with PBS/0 . 5%Triton-X100 for 20 min . The Click-iT reaction , linking azide-conjugated Alexa dye to ethynyl groups was performed for 30 min in a dark , humid environment . After 3× PBS/3%BSA and 2× PBS wash , the slides were mounted with vectashield containing DAPI ( Vector Laboratories ) to stain nuclei . RRS was performed using the Click-iT EU imaging kit ( Life Technologies ) . MDFs were seeded on coverslips and 8 h later washed with PBS and irradiated with 16 J/m2 UV-C ( Philips ) or mock treated . After 14 h recovery , the cells were labelled for 2 h in 0 . 1 mM EU-containing medium ( Ham's F10 medium supplemented with 10% dialyzed serum , 50 µg/ml penicillin/streptomycin and 20 mM HEPES ) . After a PBS wash , the cells were fixed and processes as described for UDS . Force-deflection curves from the left femora of 14-week-old and 16-week-old mice were acquired in a three-point bending assay using a Chatillon TCD series mechanical test frame ( Technex BV , The Netherlands ) , equipped with 3 mm hemi-cylindrical supports with a 8 . 5 mm total span . Width between the supports was adjusted according to the anatomical landmarks of the femur , i . e . lesser trochanter and condyles . The femora were aligned such that the femoral head was in the horizontal plane and the posterior aspect of the condyles was facing down . All samples were preconditioned for five cycles to 2 Newton ( N ) at a rate of 0 . 6 mm/min before testing to failure at a rate of 0 . 1 mm/min . The obtained force-deflection curves were analyzed for bone strength ( N/mm ) , which was represented by the Δ force/deflection of the linear part of the curve . Xpg−/− and wild type mice were sacrificed by cervical dislocation at scheduled ages ( 7 , 14 , 16 , 18 weeks ) , femora were excised and non-osseous tissue was removed . Left femora were placed in PBS and stored at −20°C until further use for mechanical testing and the right femora were fixated ( 4% formalin ) . Two days post-fixation , the right femora were scanned using the Skyscan 1076 in vivo X-Ray computed tomography ( Skyscan , Kontich , Belgium ) with a voxel size of 8 . 88 µm . Osseous tissue was distinguished from non-osseous tissue by segmenting the reconstructed grayscale images with an automated algorithm using local thresholds [122] . The region of interest ( ROI ) , i . e . distal metaphysis of the femora , was selected by using 3D data analysis software . To compensate for bone length differences between the Xpg−/− and wild type mice , the length of each ROI was determined relative to the largest specimen femur of the cohort . Cortex and trabeculae of the metaphysis were separated using in-house developed automated software . Thickness of the trabeculae and cortices were assessed using 3D analysis software as described [123] using the CT analyzer software package ( Skyscan ) . A bone specimen with known bone morphometrics was included within each scan as a quantitative control . TRIzol reagent ( Life Technologies ) was used to isolate total RNA from mouse tissue specimens . 4 µg RNA was reverse transcribed using SuperScript II ( Life Technologies ) . Real-time PCR was performed on a Bio-Rad CFX96 thermocycler using SYBR Green ( Sigma-Aldrich ) and Platinum Taq polymerase ( Life Technologies ) . Generation of specific PCR products was confirmed by melting-curve analysis and gel electrophoresis . For data analysis , the induction of target cDNA was calculated by the method described by [124] . p-values were determined using two-tailed t-tests . The used gene specific real-time PCR primers are listed in Supplementary Table S1 . Mice were euthanized by CO2 asphyxiation and blood was immediately collected from the heart . Glucose levels were measured using a Freestyle mini blood glucose meter . Albumin levels were measured in blood plasma using a mouse albumin ELISA kit ( Immunology Consultants Laboratory , Inc . ) . Primary antibodies ( supplier; dilutions ) used in this study were as follows: rabbit anti-Calbindin ( Swant; 1∶10 , 000 ) ; goat anti-ChAT ( Millipore; 1∶500 ) ; rabbit anti-GFAP ( DAKO; 1∶8 , 000 ) ; mouse anti-GM130 ( BD Transduction; 1∶100 ) ; rabbit anti-Hsp25 ( Enzo; 1∶8 , 000 ) ; rabbit anti-Iba-1 ( Wako; 1∶5 , 000 ) ; rat anti-Ki67 ( DAKO; 1∶200 ) ; rat anti-Mac2 ( Cedarlane;1∶2 , 000 ) ; mouse anti-NeuN ( Millipore; 1∶1 , 000 ) ; rabbit anti-p53 ( Leica; 1∶1 , 000 ) . For avidin– biotin–peroxidase immunocytochemistry biotinylated secondary antibodies from Vector Laboratories , diluted 1∶200 were used . FITC- , Cy3- , and Cy5-conjugated secondary antibodies raised in donkey ( Jackson ImmunoResearch ) diluted at 1∶200 were used for confocal immunofluorescence . TUNEL staining on liver , brain and retina: To quantify apoptotic cells , specimens were fixed overnight in 10% buffered formalin , paraffin-embedded , sectioned at 5 µm , and mounted on Superfrost Plus slides . Paraffin sections were employed for TdT-mediated dUTP Nick-End Labeling ( TUNEL ) assay using a commercial kit ( Apoptag Plus Peroxidase in situ apoptosis detection kit , Millipore ) . Sections were deparaffinized and incubated as described by the manufacturer . HE staining on liver , skin and small intestine: Liver , skin and intestine specimens were processed using the same fixation and sectioning methods as described for the TUNEL staining . Paraffin sections were deparaffinized , rehydrated in decreasing concentrations of ethanol and stained with haematoxilin and eosin . For immunohistochemistry , paraffin sections of liver and intestine specimens were deparaffinized , rehydrated in decreasing concentrations of ethanol , treated for 10 minutes with 3% H2O2 to quench endogenous peroxidase activity and heated to 100°C for 1 h in 10 mM sodium citrate buffer , pH 6 , for antigen retrieval . The amount of damaged ( p53 ) and proliferating ( Ki67 ) cells were subsequently detected using the avidin–biotin–immunoperoxidase complex method ( ABC , Vector Laboratories , USA ) with diaminobenzidine ( 0 . 05% ) as chromogen . Gelatin sections of brain and spinal cord: Mice were anaesthetized with pentobarbital and perfused transcardially with 4% paraformaldehyde . Brain and spinal cord were dissected out , post-fixed for 1 h in 4% paraformaldehyde , cryoprotected , embedded in 12% gelatin , rapidly frozen , and sectioned at 40 µm using a freezing microtome or stored at −80°C until use . Frozen sections were processed free floating using the ABC method ( ABC , Vector Laboratories , USA ) or single- , double- , and triple-labelling immunofluorescence . Immunoperoxidase-stained sections were analyzed and photographed using an Olympus BX40 microscope . Immunofluorescence sections were analyzed using a Zeiss LSM700 confocal . Average time spent on an accelerating rotarod ( Ugo Basile ) . Xpg−/− mice and wild type controls were given four consecutive trials of maximally 5 minutes with inter-trial intervals of 1 hour . Emx1-Xpg mice were given two trials per day with a 1 hour inter-trial interval for four consecutive days . Grip strength was determined by placing mice with forelimbs or all limbs on a grid attached to a force gauge , and steadily pulling the mice by their tail . Grip strength is defined as the maximum strength produced by the mouse before releasing the grid . For each value the test is performed in triplicate . 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 protocol was approved by the Committee on the Ethics of Animal Experiments of the Erasmus MC ( Permit Numbers: 139-03-08 , 139-09-03 , 139-12-18 ) . | Accumulation of DNA damage has been implicated in aging . Many premature aging syndromes are due to defective DNA repair systems . The endonuclease XPG is involved in repair of helix-distorting DNA lesions , and XPG defects cause the cancer-prone condition xeroderma pigmentosum ( XP ) alone or combined with the severe neurodevelopmental progeroid disorder Cockayne syndrome ( CS ) . Here , we present a novel ( conditional ) Xpg−/− mouse model which -in a C57BL6/FVB F1 hybrid background- displays many progressive progeroid features , including early cessation of growth , cachexia , kyphosis , osteoporosis , neurodegeneration , liver aging , retinal degeneration , and reduced lifespan . In a constitutive mutant with a complex phenotype it is difficult to dissect cause and consequence . We have therefore generated liver- and forebrain-specific Xpg mutants and demonstrate that they exhibit progressive anisokaryosis and neurodegeneration , respectively , indicating that a cell-intrinsic repair defect in neurons can account for neuronal degeneration . These findings strengthen the link between DNA damage and the complex process of aging . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"neuroscience",
"animal",
"models",
"physiological",
"processes",
"developmental",
"biology",
"model",
"organisms",
"organism",
"development",
"dna",
"research",
"and",
"analysis",
"methods",
"mouse",
"models",
"aging",
"biochemistry",
"cell",
"biology",
"nucleic",
"acids",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"dna",
"repair",
"molecular",
"cell",
"biology",
"genetics",
"of",
"disease"
]
| 2014 | Cell-Autonomous Progeroid Changes in Conditional Mouse Models for Repair Endonuclease XPG Deficiency |
Voluntary motor commands produce two kinds of consequences . Initially , a sensory consequence is observed in terms of activity in our primary sensory organs ( e . g . , vision , proprioception ) . Subsequently , the brain evaluates the sensory feedback and produces a subjective measure of utility or usefulness of the motor commands ( e . g . , reward ) . As a result , comparisons between predicted and observed consequences of motor commands produce two forms of prediction error . How do these errors contribute to changes in motor commands ? Here , we considered a reach adaptation protocol and found that when high quality sensory feedback was available , adaptation of motor commands was driven almost exclusively by sensory prediction errors . This form of learning had a distinct signature: as motor commands adapted , the subjects altered their predictions regarding sensory consequences of motor commands , and generalized this learning broadly to neighboring motor commands . In contrast , as the quality of the sensory feedback degraded , adaptation of motor commands became more dependent on reward prediction errors . Reward prediction errors produced comparable changes in the motor commands , but produced no change in the predicted sensory consequences of motor commands , and generalized only locally . Because we found that there was a within subject correlation between generalization patterns and sensory remapping , it is plausible that during adaptation an individual's relative reliance on sensory vs . reward prediction errors could be inferred . We suggest that while motor commands change because of sensory and reward prediction errors , only sensory prediction errors produce a change in the neural system that predicts sensory consequences of motor commands .
Our motor commands generally produce two kinds of consequences: a sensory consequence in terms of activity in our primary sensory organs ( e . g . , vision , proprioception ) , and a rewarding consequence in terms of forming a subjective measure of utility or usefulness of these sensations ( e . g . , release of dopamine ) . For example , while dancing , the motor commands that move our body produce proprioceptive feedback , while internal evaluation of that feedback indicates a pleasurable experience . These two consequences of the motor command form the basis for two kinds of prediction error: a sensory prediction error , and a reward prediction error . In principle , learning from sensory prediction error should alter an internal model that predicts the sensory consequences of motor commands , i . e . , a forward model [1] , [2] . In contrast , learning from reward prediction error should alter the valuation of the sensory states that are the consequence of those motor commands , i . e . , a value function . Motor adaptation studies often focus on learning from sensory prediction error [1] , [2] , [3] , [4] , [5] , [6] , [7] , despite the fact that people are also rewarded for each movement . Similarly , studies that focus on learning from reward prediction error ( e . g . , decision making tasks ) often do not consider potential sensory prediction errors [8] , [9] , [10] . It seems rational that most learning would rely on both kinds of error . Here , we focus on a simple motor adaptation task and consider a mathematical framework in which both reward and sensory prediction errors could contribute to the trial-to-trial change in the motor commands . We attempt to ask whether learning from these two distinct signals can be behaviorally dissociated . Our idea is that while motor commands might change because of sensory or reward prediction errors , only in the former case would there also be a change in the map that predicts the sensory consequences of the motor command . We focus on a well studied motor adaptation protocol: reaching in the context of visuomotor perturbations . While there have been numerous models of motor adaptation [4] , [5] , [11] , [12] , [13] , [14] , to our knowledge all current models assume that the process of motor adaptation is driven by sensory prediction errors . Our objective is to test the hypothesis that during motor adaptation , learning from sensory prediction errors leaves a behavioral signature that is distinct from learning from reward prediction errors .
Consider a typical adaptation task in which the learner experiences a perturbation . The limb is covered by a screen to prevent direct observation of the hand , and a cursor that represents hand position undergoes a kinematic rotation so that when the hand moves straight ahead , the cursor moves slightly to the left ( Fig . 1A ) . Reward is provided if the cursor passes through the target area . In this reach adaptation task there are two kinds of error: the difference between the expected and observed visual feedback of the hand ( i . e . visual cursor ) , and the difference between the expected and observed success of the reach . Our hypothesis is that learning mechanisms engaged by the two types of error may be behaviorally dissociable . To examine this hypothesis , we recruited two groups of subjects in Experiment 1 . One group ( RWD ) was provided only with information regarding whether they succeeded or failed at each trial ( reward r = 1 or 0 ) , indicated by explosion of the target , and received no other visual feedback regarding their movement ( Fig . 1B ) . Another group was provided with full visual feedback of the cursor as well as the reward so that they were able to use both potential error signals ( ERR ) . We asked two questions: 1 ) In the ERR paradigm in which sensory consequences of motor commands were available , would adaptation of the motor commands accompany a change in the motor-sensory map ( i . e . , a change in the perceived sensory consequences of motor commands ) , and 2 ) in the RWD paradigm in which sensory consequences of motor commands were unavailable , would adaptation of the motor commands take place but without a change in the motor-sensory map . Fig . 1C shows data from representative subjects in the ERR and RWD paradigms . In this figure , the yellow line in the ERR group is the ideal reach angle ( shifts gradually up to 8° ) . The gray area indicates the region that provided reward , which shifts with the same schedule in both groups . The subjects were provided with different kinds of error feedback , but updated their motor commands by roughly the same amount ( group data , mean change in reach direction , 7 . 49° for RWD and 7 . 63° for ERR , not significantly different from each other p>0 . 8 , t-test ) . The total amount of adaptation of the two groups was comparable . However , the variability of reach angles was larger for the RWD subject ( Fig . 1C ) , and this was consistent across the entire group ( Fig . 1D ) . Before and after this adaptation task ( PRE and POST adaptation ) , we measured how subjects predicted the sensory consequences of their motor commands . In this localization part of the task , after subjects completed a reach with their right hand , their hand was returned to the center location , and they were then asked to estimate the location of their right hand in the previous trial by pointing with their left hand over the screen ( Fig . 1A ) . During the localization neither the cursor nor the target was projected . The localization data for representative subjects are shown in Fig . 1E . As a consequence of adaptation , the subject in the ERR group had a sensory remapping in which she estimated her hand to be to the left of its actual position . In contrast , the subject in the RWD group had little or no sensory remapping , suggesting that the changes in the motor commands did not accompany a change in the motor-sensory map . Fig . 2A shows the group data for the localization task . We compared the change in the estimate of hand position from the PRE to the POST adaptation condition and found that the subjects in the ERR group estimated their hand position to have changed by 8 . 8°+/−0 . 6° to the left of actual position . In contrast , in the POST condition of the RWD group , the subjects had no significant change in their sensory estimates ( there was a significant difference between PRE and POST in the ERR group p<0 . 0001 , whereas the difference in the RWD group was not significant p = 0 . 8 ) . If the sensory and reward prediction errors engage learning in distinct neural structures , then adaptation might result in distinct generalization patterns [15] , [16] , [17] , [18] . To test this idea , we recruited subjects for Experiment 2 and quantified the patterns of generalization that accompanied adaptation . In the adaptation session , the target was projected at 0° ( straight ahead ) . In the pre and post adaptation periods the target appeared randomly at various angular displacements ( −30 to 30 deg ) . For these generalization targets , we provided neither the cursor nor reward information . Fig . 2B plots the average reach angle across subjects for each target direction . We found that the RWD group had a narrower generalization function than the ERR group ( ANOVA , F ( 1 , 126 ) = 9 . 632 , p = 0 . 005 ) . In summary , in the RWD condition the learning that produced changes in the motor commands accompanied a narrow generalization function and no change in the map that predicted the sensory consequences of motor commands . In contrast , in the ERR paradigm the learning that produced changes in the motor commands accompanied a broad generalization function and a significant change in the perceived sensory consequences of motor commands . In the RWD paradigm the binary feedback signal carried much less information than the continuous sensory error signal available in the ERR paradigm . This may have forced the subjects to adopt a completely new strategy , making the learning that we see in the RWD paradigm irrelevant for a typical adaptation paradigm . In Experiment 3 we considered a paradigm ( EPE ) in which the visual cursor was available only at the endpoint of the movement and was otherwise invisible during the reach . In this new experiment we measured the localization change ( as in Exp . 1 ) and the generalization ( as in Exp . 2 ) , attempting to test the results of experiments 1 and 2 in the same population . Fig . 1C shows the reach angles of a representative subject in the EPE group . The adaptation in the EPE group was comparable with the ERR group ( mean change in reach direction , t-test , p = 0 . 64 ) , i . e . , the motor commands in the three groups adapted by approximately the same amount . Interestingly , in the localization task the subject in the EPE group had a sensory illusion that was in between the ERR and RWD groups ( Fig . 1E ) . In the group data in the POST adaptation condition , the strength of the localization illusion in the EPE group was weaker than in the ERR group ( t-test , p<0 . 007 ) , but stronger than the RWD group ( t-test , p<0 . 006 ) ( Fig . 2A ) . The generalization of the EPE group appeared to be in between ERR and RWD ( we did not see a significant difference from either ERR or RWD , Fig . 2B ) . In Experiment 2 we had found that learning from reward produced a narrow generalization , while in Experiment 1 we had found that learning from error produced a motor-sensory remapping . In Experiment 3 we had the means to test a crucial prediction: across subjects , individuals who relied more on reward ( narrow generalization ) should show a smaller motor-sensory remapping . Indeed , we found a significant correlation between the amount of generalization and the localization illusion across subjects ( Fig . 2C ) . That is , it appeared that when a subject had a larger sensory illusion ( suggesting that learning was driven more by sensory prediction errors ) , they also had a wider generalization . To explore the mechanism behind these findings , we considered a model of adaptation that relied on both sensory and reward prediction errors ( Fig . 3A ) . Suppose that the brain generates a motor command u , resulting in a change in the state of the hand h , which also depends on a perturbation p . The nervous system senses the resulting motion of the limb y as well as whether that motion was rewarded r . Here , we considered a learner who updates motor command u to maximize reward . In theory , producing the motor commands that maximize probability of reward may rely on two kinds of learning: forming an optimal action selector , and forming an optimal state predictor ( Fig . 3B ) . On trial k , action selector outputs motor commands . This depends on the estimated perturbation ( which depends on sensory prediction error ) , as well as the reward prediction error . Therefore , in theory the trial-to-trial change in the motor commands is driven by two different error signals: the state estimator updated by the sensory prediction error , and the action selector updated by the reward prediction error . An important prediction from this model is that reliance on the sensory prediction error is modulated by the Kalman gain , which is the ratio of estimation uncertainty to observation uncertainty . Therefore , if the uncertainty of visual feedback is large , the credit on the sensory prediction error becomes small , which makes the credit on the reward prediction error larger . Fig . 3C shows results of simulations for different uncertainty levels of visual feedback . When the learner is provided with high quality visual feedback ( analogous to ERR condition , Fig . 3C left column ) , it updates its estimate of perturbation , resulting in a motor-sensory remapping . As a result , the estimated hand position is near the location of the cursor and different from actual hand position . In contrast , when the learner is provided with uncertain visual feedback ( analogous to EPE condition , middle column in Fig . 3C ) , the learner alters the motor commands using both the sensory prediction error and the reward prediction error . In this case , the adaptation produces a partial sensory remapping ( is not very different from in the middle column of Fig . 3C ) . Finally , when the learner is provided with extremely poor visual feedback ( analogous to RWD condition , right column of Fig . 3C ) , all that is available to the learner is success or failure ( = 0 or 1 ) . The learner still alters the motor commands to compensate for the perturbation , but the adaptation does not produce a sensory remapping ( is not different from in the right column of Fig . 3C ) . These three different patterns of sensory remapping generated by the model help explain the reason why we observed different patterns of sensory remapping in the three different paradigms . In the ERR condition in which high quality sensory feedback was available , adaptation produced large change in the state predictor , producing the sensory remapping . In RWD condition in which the visual feedback of the cursor was not available , adaptation focused on the action selector , which was updated by reward prediction error . Because this process did not involve a sensory remapping , we did not observe a change in the localization behavior of the subjects . In the EPE condition in which partial visual feedback was provided , learning depended on both an updating of the state predictor and the action selector . As a result , we observed the partial sensory remapping . To validate our model , we used it to estimate how much of the change in the motor commands that we observed in our subjects was due to each type of error . We imagined that the motor commands were generated by the sum of two states with a search noise , , where represents the estimate of the perturbation as updated by sensory prediction error and the is updated by reward prediction error . Using a nonlinear optimization algorithm , we fit the model to the trial-to-trial behavior of each subject ( reach direction on each trial ) , and the state of reward on that trial . In the RWD paradigm , the only feedback available was reward prediction error , i . e . , . The results of our model fit are shown in Fig . 4 via the average of estimated parameters and , and their sum . These estimated values were superimposed on the average of subjects' trial-to-trial reach angle ( black line ) with SEM across subjects . In the ERR condition , by the end of adaptation the contributions of these two states were [ , ] = [7 . 82 , 0 . 26]+/−[0 . 18 , 0 . 31] . Despite the fact that we used the exact the same model to fit the data for ERR and EPE , the best fit estimates of these two states in EPE were [ , ] = [4 . 53 , 3 . 33]+/−[0 . 59 , 0 . 69] , which were significantly different from those of ERR ( ANOVA , F ( 1 , 18 ) = 18 . 93 , p<0 . 001 ) . By fitting the model to the data , we were able to estimate the search noise . We found that the variance of the search noise in ERR was , which was significantly smaller than that of EPE ( , p<0 . 001 ) , and RWD ( , t-test , p<0 . 01 ) . Our estimate of a significantly smaller search noise in the ERR condition is consistent with our inference that with high quality sensory feedback , the change in the motor commands is driven almost entirely by sensory prediction errors . This is also consistent with the fact that in the ERR condition , there was a scarcity of reward prediction error: In ERR , more than 95% of trials were rewarded , whereas the probability of reward in EPE was 83% and that in RWD was 76% . Therefore , our analysis suggests that in the ERR paradigm the change in the motor commands was due primarily to adaptation of the state estimator ( accounting for the sensory remapping ) , whereas in the RWD paradigm the change was due to adaptation of the action selector ( accounting for the lack of sensory remapping ) . In the EPE paradigm the change was due to both the state estimator and the action selector .
Our goal was to determine whether during motor adaptation one could dissociate between learning from reward prediction errors vs . learning from sensory prediction errors . We considered a reaching task in which visual feedback regarding cursor position was altered . The quality of this feedback was manipulated so that in one group the sensory feedback was of high quality ( available throughout the reach , ERR group ) , in another group the sensory feedback was of low quality ( available only at the end of the reach , EPE group ) , and in a third group the sensory feedback was unavailable ( RWD group ) . All groups had access to reward ( success or failure ) at the end of their movement . We found that after a long period of training , all three groups adapted their motor commands . In the ERR group this adaptation accompanied a wide pattern of generalization and a significant change in the perceived sensory consequences of motor commands . In contrast , in the RWD group the adaptation accompanied a narrow pattern of generalization and no change in the perceived sensory consequences of motor commands . In the EPE group , generalization and sensory remapping were intermediate . Interestingly , in the EPE group individuals who demonstrated a larger sensory remap also had a wider generalization function . Increasing the uncertainty in the sensory prediction error altered both the width of generalization function and the amount of sensory remapping , while it did not affect the level of adaptation . While previous models of motor adaptation have relied exclusively on sensory prediction errors to form an estimate of the perturbation [4] , [5] , [19] , [20] , the comparable levels of motor adaptation in our groups ( ERR , RWD , and EPE ) suggest that the brain relied on another source of error , the reward prediction error , when the sensory prediction error was not informative . In fact , it has been shown that the reward may modulate motor planning [21] , [22] . Thus , it seems more rational that the purpose of learning is not merely to estimate the magnitude of a perturbation , but to produce motor commands that maximize reward [23] . We formulated this adaptation as a reward maximization process by assuming an “optimal learner” . The optimization relied on two update equations: one was the optimal estimator that inferred the state of the body , and the other was the optimal policy that selected the action as a function of the estimated state [24] , [25] , [26] . Based on this theory , our model of the optimal learner was composed of two components: reinforcement learning for action selection , and state estimation for identifying the sensory consequences of motor commands [27] . In this model , the objective of state estimation was to estimate the perturbation in the environment and the hand position as a consequence of the motor command , while the objective of the reinforcement learning was to update how to select the action to maximize reward probability [28] . The simulation showed that the learner relied mostly on the sensory prediction error in ERR paradigm . As a result , the learner updated the parameter associated with the sensory consequence of the motor command , which predicted the illusion that we observed in Experiment 1 . In contrast with the ERR paradigm , the RWD paradigm did not provide the sensory prediction error . Thus , the simulation with the RWD paradigm showed that the reward-prediction error updated the action but did not change the estimate of hand position . Thus , high quality sensory feedback produced learning that depended primarily on sensory prediction errors . While our model was not designed to account for the distinct generalization patterns in the ERR and the RWD paradigms , previous studies have speculated that generalization patterns are a reflection of the neural encoding of information during learning [16] , [29] . For example , generalization patterns during reach adaptation in force fields appear consistent with an encoding in which the neurons have activity fields that resemble those in the primary motor cortex [13] , [30] . In contrast , generalization patterns in visuomotor rotations appear more consistent with an encoding similar to cells in the posterior parietal cortex [31] . In this framework , the two different generalization patterns seen in RWD and ERR paradigms suggest engagement of two different neural mechanisms that each learn from reward and sensory prediction error . Another possibility , however , is that the two forms of prediction error converge on a single neural structure that guides motor learning . By presenting the optimal learner model that includes two forms of prediction error , we built a connection between two disparate areas of research that has focused on different parts of the brain . Motor adaptation has focused on tasks that typically depend on the integrity of the cerebellum [32] , [33] . Habit learning [34] , visuomotor sequence learning [35] , or action selection [36] , [37] have focused on tasks that depend on the integrity of the basal ganglia [38] , [39] . In fact , goal directed action in habitual learning is mediated by two representations: a representation of the instrumental contingency between the action and the outcome , and a representation of the outcome as a goal for the agent [40] . Because motor adaptation is also a goal directed action , the two learning mechanisms observed in this paper might be the general systems involved in a broad category of procedural learning . For example , these two distinct memories might be mediated by parallel cortico-basal ganglia mechanisms with different sensory domains [35] . Patients with basal ganglia disorders show little or no deficits in motor adaptation paradigms like force fields [33] or visuomotor perturbations [41] , [42] ( although patients with PD appear to show a deficit in consolidation of the memory [43] ) . Why is this ? Our theory provides a potential answer: in the typical force field or visuomotor tasks , high quality sensory feedback is available , making it likely that sensory prediction errors play a dominant role . Because learning from sensory prediction errors likely depends on the integrity of the cerebellum [2] , [32] , [44] , the implication is that the ability of basal ganglia patients to adapt to visuomotor and force field perturbations is not evidence for normal motor adaptation , but rather evidence for the idea that changes in motor output in these tasks are primarily driven by sensory prediction errors . The other implication of the theory is that the inability to adapt the sensory consequences of motor commands did not prevent adaptation of the motor commands in response to reward prediction errors . Indeed , when we altered the adaptation paradigm and made it so that changes in the motor output were driven by reward prediction errors , we found that in response to the reward prediction error subjects altered their motor commands . This theory predicts that by providing rewards appropriately during a motor adaptation task , the cerebellar patients may be able to update their motor commands without sensory recalibration . Another implication of the theory is that the active search noise to explore the motor commands plays an important role in updating the action selector . Indeed , we found that trial-to-trial variability was modulated depending on types of error with significantly larger variability in the RWD than in ERR and in EPE . In previous studies , movement variability is generally thought to be due to signal dependent noise in the neuronal structures that generate motor commands [45] , [46] , [47] . However , noise is present even in the planning stage of movements [48] . Here , we found that during adaptation variability in movements was not due to meaningless noise , but an inherent part of a search that the brain engaged in to find motor commands that acquired a more rewarding state . In summary , changes that take place in motor commands during adaptation are likely to be driven by both sensory and reward prediction errors . Learning from sensory prediction error alters the predicted sensory consequences of motor commands , leaving behind a sensory remapping . During motor adaptation , the reliance on reward prediction errors can be increased by degrading the quality of the sensory feedback . Learning from reward prediction error does not accompany a sensory remapping . It is likely that the neural basis of learning from sensory and reward prediction errors are distinct because they produce different generalization patterns .
Subjects sat in front of a robotic arm and held its handle [25] . A video projector painted the screen that covered the manipulandum and the subject's arm . A trial began by the robot positioning the subject's hand in a start box , at which point a target of 6° width appeared at 10 cm . Subjects were instructed to perform a ‘shooting’ motion so that their hand crossed within the target area , at which point the target was animated to show an explosion , and a score was increased by one point . In the error-based learning ( ERR ) paradigm , the cursor position was displayed during the movement toward the target . In the reward-based learning ( RWD ) paradigm , the cursor position was not displayed . For both groups , target explosion indicated success of the trial . The cursor was not displayed during the return of the hand to the start position . Protocols were approved by the local IRB and all subjects signed a consent form . | It is thought that motor adaptation relies on sensory prediction errors to form an estimate of the perturbation . Here , we present evidence that motor adaptation can be driven by both sensory and reward prediction errors . We found that learning from sensory prediction error altered the predicted consequences of motor commands , leaving behind a sensory remapping , whereas learning from reward prediction error produced comparable change in motor commands , but did not produce a sensory remapping . It is possible that the neural basis of learning from sensory and reward prediction errors are distinct because they produce different generalization patterns . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"biology",
"neuroscience"
]
| 2011 | Learning from Sensory and Reward Prediction Errors during Motor Adaptation |
miR-122 , a liver-specific microRNA , is one of the determinants for liver tropism of hepatitis C virus ( HCV ) infection . Although miR-122 is required for efficient propagation of HCV , we have previously shown that HCV replicates at a low rate in miR-122-deficient cells , suggesting that HCV-RNA is capable of propagating in an miR-122-independent manner . We herein investigated the roles of miR-122 in both the replication of HCV-RNA and the production of infectious particles by using miR-122-knockout Huh7 ( Huh7-122KO ) cells . A slight increase of intracellular HCV-RNA levels and infectious titers in the culture supernatants was observed in Huh7-122KO cells upon infection with HCV . Moreover , after serial passages of HCV in miR-122-knockout Huh7 . 5 . 1 cells , we obtained an adaptive mutant , HCV122KO , possessing G28A substitution in the 5’UTR of the HCV genotype 2a JFH1 genome , and this mutant may help to enhance replication complex formation , a possibility supported by polysome analysis . We also found the introduction of adaptive mutation around miR-122 binding site in the genotype 1b/2a chimeric virus , which originally had an adenine at the nucleotide position 29 . HCV122KO exhibited efficient RNA replication in miR-122-knockout cells and non-hepatic cells without exogenous expression of miR-122 . Competition assay revealed that the G28A mutant was dominant in the absence of miR-122 , but its effects were equivalent to those of the wild type in the presence of miR-122 , suggesting that the G28A mutation does not confer an advantage for propagation in miR-122-rich hepatocytes . These observations may explain the clinical finding that the positive rate of G28A mutation was higher in miR-122-deficient PBMCs than in the patient serum , which mainly included the hepatocyte-derived virus from HCV-genotype-2a patients . These results suggest that the emergence of HCV mutants that can propagate in non-hepatic cells in an miR-122-independent manner may participate in the induction of extrahepatic manifestations in chronic hepatitis C patients .
Hepatitis C virus ( HCV ) infects over 170 million people worldwide and is a major cause of chronic hepatitis , liver cirrhosis , and a high rate of hepatocellular carcinoma [1] . Chronic infection with HCV is often associated with extrahepatic manifestations ( EHMs ) such as mixed cryoglobulinemia , non-Hodgkin lymphoma , thyroiditis and diabetes mellitus [2] . A low level of HCV-RNA replication has been detected in both peripheral blood mononuclear cells ( PBMCs ) and neuronal tissues [3 , 4] , suggesting that EHMs might be induced by the extrahepatic propagation of HCV . If so , however , the detailed mechanisms remain unknown . Although there is no effective vaccine against HCV due to its genetic variability ( quasispecies ) , the recent development of direct-acting antivirals ( DAA ) targeting viral proteins , such as NS3-4A protease , NS5A or NS5B polymerase [5] , has improved the outcome . However , it has been reported that during DAA treatment , adaptive mutations in the HCV genome can be readily induced due to the low fidelity of NS5B RNA polymerase [6] , resulting in the emergence of drug-resistant viruses [7–9] . A liver-specific microRNA ( miRNA ) , miR-122 , has been shown to be one of the most important host factors for HCV replication [10] . In general , miRNA negatively regulates the translation of mRNA through an interaction with the 3’UTR in a sequence-specific manner . miR-122 regulates gene expressions important for functions involved in the maintenance of liver homeostasis , including the functions of lipid metabolism , iron metabolism , and carcinogenesis [11 , 12] . On the other hand , the HCV 5’UTR has two binding sites for miR-122 , which are highly conserved beyond the HCV genotype [13] . Upon HCV infection , miR-122 has been shown to stabilize HCV-RNA [14] and enhance internal ribosome entry site ( IRES ) -mediated translation [10 , 15 , 16] and replication [17] of HCV-RNA through direct interaction with the HCV 5’UTR [18 , 19] . However , the detailed function of miR-122 in the HCV lifecycle has not been understood well . In addition , this interaction results in the sequestration of miR-122 from host mRNA targets , which may be responsible for the long-term oncogenic potential of HCV [20] . Recent studies have revealed the importance of Ago2 , a catalytic component of the RNA-induced silencing complex ( RISC ) , and of Xrn1 and Xrn2 , 5’ to 3’exonucleases involved in mRNA decay , on the stabilization of HCV-RNA; however , the detailed mechanisms are largely unclear . Ago2 participates in the stabilization and enhancement of translation and replication of HCV-RNA via direct interaction with miR-122 and the HCV 5’UTR [21 , 22] , and Xrn1 and Xrn2 are involved in degradation of HCV-RNA [23–25] , while miR-122 might protect the HCV genome from such degradation . A locked nucleic acid ( LNA ) complementary to miR-122 has advanced to phase II clinical trial , and subcutaneous injection of LNA into chronic hepatitis C patients was shown to suppress the propagation of HCV without the appearance of any drug-resistant virus or adaptive mutation in the 5’UTR of the HCV genome [26] . Recently , however , the G28A mutation adjacent to the miR-122 binding site in the 5’UTR of the genotype 2a Jc1 genome introduced during infection in the presence of an miR-122 decoy has been shown to possess more potent accessibility and affinity to miR-122 [27] . In addition , we and another group have previously shown that a low level of replication of HCV occurred in several miR-122-deficient cells including non-hepatic Hec1B and hepatoma Hep3B cells [28 , 29] , suggesting that HCV-RNA is capable of propagating in an miR-122-independent manner . In this study , we investigated the roles of miR-122 not only on replication of HCV-RNA but also on production of infectious particles by using miR-122-knockout Huh7 cells . Adaptive HCV mutants obtained after serial passages in miR-122-knockout cells exhibited efficient RNA replication in an miR-122-independent manner by introduction of the G28A mutation for miR-122-independence and several mutations for higher infectivity . The emergence of an HCV mutant that can propagate in an miR-122-independent manner may participate in the induction of extrahepatic manifestations in chronic hepatitis C patients .
To clarify the biological significance of miR-122 in the life cycle of HCV in more detail , we established miR-122-knockout Huh7 ( Huh7-122KO ) cells by using artificial endonucleases ( transcription activator-like effector nucleases: TALEN ) . Two clones ( clone #1 and #2 ) of Huh7-122KO cells were obtained and mutations in the miR-122-coding region were assessed by Surveyor nuclease assay and sequencing analysis ( S1A and S1B Fig ) . Lack of expression of miR-122 in Huh7-122KO cells was confirmed by Northern blotting and qRT-PCR ( S1C Fig ) . In addition , we established Huh7-122KO cells in which the expression of miR-122 was recovered by infecting Huh7-122KO#1 and Huh7-122KO#2 cells with a lentiviral vector expressing miR-122 and obtained Huh7-122KOR#1 and Huh7-122KOR#2 cells , respectively . Next , to confirm that the loss of miR-122 activity suppressed translation of the target gene in Huh7-122KO cells , a pmirGLO vector carrying the complementary sequence of miR-122 under the luciferase gene was transfected into Huh7-122KO cells . Suppression of luciferase activity was observed not only in parental Huh7 cells , but also in Huh7-122KOR#1 and Huh7-122KOR#2 cells , while no suppression was observed in Huh7-122KO#1 and Huh7-122KO#2 cells ( S1D Fig ) . Furthermore , we confirmed that cell growth and viability of Huh7-122KO cells were comparable to those of restored cells ( S2 Fig ) . These results suggest that Huh7-122KO cells are functionally deficient in miR-122 activity and exogenous expression of miR-122 restored their biological function . First , we compared the expression levels of host mRNAs between Huh7-122KO and Huh7-122KOR cells by cDNA microarray . From the results of pathway prediction in Huh7-122 KO cells by Ingenuity Pathway Analysis ( IPA ) , the expression levels of genes involved in lipid metabolism were changed in Huh7-122KO cells ( S1 Table ) , as in previous reports on miR-122-knockout mice [11 , 12] . Previous reports have shown that lipid metabolism participates in the entry or assembly of HCV . Therefore , to examine the effect of miR-122-knockout-induced changes in the lipid composition on the entry of HCV , pseudotyped VSV bearing HCV envelope proteins , HCVpv , was inoculated into the cell lines . No difference in luciferase activity was observed irrespective of miR-122 expression upon infection with HCVpv ( S3 Fig ) , suggesting that the knockout of miR-122 has no significant effect on the entry of HCV . Next , to investigate the function of miR-122 on the replication of viral RNA , we examined the effects of stable and transient expression of miR-122 in the miR-122-knockout cells on colony formation . HCV subgenomic replicon ( SGR ) RNA of the JFH1 strain was electroporated into Huh7-122KO and Huh7-122KOR cells . The numbers of colonies were significantly higher in Huh7-122KOR cells than in Huh7-122KO cells ( Fig 1A , left panels ) , while the expression of NS5A and formation of membranous webs were similarly observed in both Huh7-122KO-SGR and Huh7-122KOR-SGR cells ( S4A and S4B Fig ) , respectively . In addition , the inhibitory effect by treatment with IFNα and BILN was comparable with Huh7-122KO-SGR cells ( S5 Fig ) . Although higher levels of miR-122 expression in Huh7-122KO cells transduced with the mature form of miR-122 ( miR-122-mimic ) were detected at 72 h post-electroporation ( Fig 1B ) , no miR-122 was detected at the time of colony harvest ( at 21 days post-electroporation ) due to the instability of miR-122-mimic ( Fig 1C , lower panel ) . However , enhancement of colony formation was observed upon electroporation of SGR RNA together with miR-122-mimic but not of control-mimic in Huh7-122KO cells ( Fig 1A , right panels ) , suggesting that not continuous but transient expression of miR-122 is sufficient for the enhancement of colony formation . Next , we confirmed that HCV SGR RNA in miR-122 KO cells could replicate independently from miR-122 . High levels of HCV-RNA replication were detected in Huh7-122KO-SGR cells ( Fig 1C , upper panel ) , which were resistant to the treatment with miR-122 inhibitor , in contrast to Huh7-122KOR-SGR cells ( Fig 1D ) . In addition , more abundant HCV-RNA was detected in the Ago2 complex immunoprecipitated with a specific antibody in Huh7-122KOR-SGR cells than in Huh7-122KO-SGR cells ( Fig 1E ) . These results suggest that miR-122 is required in the early phase for the efficient translation of viral RNA crucial for formation of the replication complex , and transient expression of miR-122 is able to rescue viral replication in Huh7-122KO cells . However , once the replication complex is formed , the HCV genome can replicate in an miR-122-independent manner . Next , we examined the effect of the knockout of miR-122 in Huh7 cells on the propagation of HCV . Although Huh7-122KOR cells exhibited efficient HCV replication , a slight increase of intracellular HCV-RNA levels and detectable infectious titers in the culture supernatants was observed in Huh7-122KO cells upon infection with HCV at an MOI of 3 ( Fig 2A ) . To rule out the possibility of incorporation of miR-122 into viral particles , in vitro transcribed full-genomic HCV-RNA was electroporated into Huh7-122KO cells together with either control- or miR-122-mimic , and infectious titers in the culture supernatants were determined ( S6 Fig ) . A slight but significant increase of infectious titer was detected in the culture supernatants of Huh7-122KO cells co-electroporated with viral RNA and control-mimic . To further confirm the miR-122 independent propagation of HCV , we examined the effect of inhibitors for HCV , including IFNα , DAA and anti-CD81 antibody , on the propagation of HCV in Huh7-122KO cells . The level of HCV-RNA in Huh7-122KO cells upon infection with HCV was decreased by the treatment with the reagents ( Fig 2B ) . However , as we expected , the treatment with LNA-miR-122 exhibited no effect on the replication of HCV in Huh7-122KO ( Fig 2C ) . Previous reports showed that a specific membrane alteration that is involved in viral replication , i . e . , the formation of a membranous web , was observed in HCV-replicating cells [30] . Correlative fluorescence microscopy-electron microscopy ( FM-EM ) also revealed the localization of NS5A on convoluted structures such as the membranous web in Huh7-122KO cells ( S7 Fig ) . Moreover , HCV core proteins and lipid droplets were co-localized in Huh7-122KO cells ( S8 Fig ) , suggesting that HCV can propagate in an miR-122-independent manner . Next , to examine the role of miR-122 on the spread of HCV infection , we compared the focus formation in Huh7-122KO and Huh7-122KOR cells at 72 h post-infection . The size and number of foci formed in Huh7-122KO cells were smaller and fewer than those in Huh7-122KOR cells ( Fig 3A ) . Focus-formation efficiency was 1000-fold lower in Huh7-122KO cells than in Huh7-122KOR cells upon infection with HCV at an MOI of 1 ( Fig 3B ) . In addition , the average numbers of infected cells constituting one focus in Huh7-122KO cells ( approximately 3 cells/focus ) were significantly smaller than those in Huh7-122KOR cells ( approximately 10 cells/focus ) ( Fig 3C ) . To investigate the propagation of HCV in Huh7-122KO cells in more detail , the expression of dsRNA and NS5A protein in cells upon infection with HCV was assessed by immunofluorescence analysis . Co-localization of dsRNA and NS5A in Huh7-122KOR cells was observed at 24 h post-infection , but was detected at 48 h post-infection in Huh7-122KO cells ( Fig 3D ) , supporting the hypothesis that miR-122 is required for the efficient formation of the replication complex as shown by replicon assay . Because expression of lipoprotein-associated ApoB , ApoE or MTTP was decreased in Huh7-122KO cells ( S9 Fig ) , we examined the involvement of miR-122 in HCV particle formation . However , no significant difference was observed in particle formation relative to intracellular copy number at 72 h post-infection between Huh7-122KO and Huh7-122KOR cells ( S10 Fig ) , suggesting that miR-122 does not participate in infectious particle formation . Collectively , these results suggest that the low levels of focus formation of HCV in Huh7-122KO cells are attributable to inefficient translation and replication of the HCV genome , but not to the particle formation process . Generally , it has been reported that cured cells established by the elimination of viral RNA from HCV replicon cells exhibited more potent propagation of HCV than parental cells . Previous reports have shown that the high susceptibility of cured cells to HCV infection is due to the high level of miR-122 expression [31–33] . On the other hand , Huh7 . 5 . 1 cells , which are also derived from HCV replicon cells , carry mutations in RIG-I , a key innate immune sensor for viral RNA [34] and express a low level of CREB3L1/OASIS , which specifically inhibits the proliferation of cells infected with HCV [35] . To determine the difference in propagation of HCV in the absence of miR-122 between Huh7 and Huh7 . 5 . 1 cells , we established 3 clones of miR-122-knockout Huh7 . 5 . 1 ( 751-122KO ) cells ( #1- #3 ) and the miR-122 deficiency in each clone was confirmed by sequencing analysis and qRT-PCR ( S11A and S11B Fig ) . The results showed that RNA replication and infectious particle formation in the supernatants of 751-122KO cells were higher than those in Huh7-122KO cells or cured cells upon infection with HCV at an MOI of 1 ( Fig 4A ) . To further confirm the difference of infectivity between parental and cured cells , we also established cured cells from Huh7-122KO-SGR cells ( #3 and #5 ) by treatment with IFNα and BILN ( S12 Fig ) . However , the infectivity of cured cells was comparable to that of parental Huh7-122KO cells , suggesting that the high susceptibility of Huh7 . 5 . 1 cells to HCV infection depends not only on a high level expression of miR-122 but also on other factors . Because 751-122KO cells exhibit higher susceptibility to HCV propagation than Huh7-122KO cells , we tried to generate an adapted mutant that could propagate in an miR-122-independent manner in 751-122KO cells . Infectious titers in the supernatants of 751-122KO cells upon infection with high titers of HCV were increased by cell passaging and reached 106 FFU/ml ( Fig 4B ) . The adapted HCV mutant capable of propagating in miR-122-knockout cells was designated HCV122KO . Although HCV propagated well in Huh7 . 5 . 1 cells , the intracellular HCV-RNA levels and infectious titers in the supernatants were impaired in 751-122KO cells . In contrast , HCV122KO exhibited an efficient and comparable propagation in both cell lines irrespective of the expression of miR-122 ( Fig 4C ) . Next , to confirm miR-122-independent propagation of HCV122KO , replication of HCV was assessed by an indicator system [36] , which monitored HCV propagation by the nuclear localization of the IPS-1-GFP fusion protein through cleavage by NS3-4A protease upon infection with HCV . Although nuclear localization of GFP was similarly detected from 24 h post-infection and completed at 72 h post-infection in Huh7 . 5 . 1 cells infected with either HCV or HCV122KO , it was severely impaired in 751-122KO cells infected with HCV , and only small numbers of GFP molecules were detected in the nucleus at 48 h post-infection . In contrast , nuclear localization of GFP comparable to that in Huh7 . 5 . 1 cells was observed in 751-122KO cells upon infection with HCV122KO ( Fig 4D ) , suggesting that HCV122KO is capable of propagating in an miR-122-independent manner . To verify the ability of HCV122KO to propagate efficiently in an miR-122 independent manner , several miR-122-deficient cells , including non-hepatic cells , were infected with HCV122KO . miR-122-independent intracellular RNA replication and infectious particle formation of HCV122KO was observed in Hep3B cells deficient in miR-122 expression [29] ( Fig 5A ) . Moreover , compared to HCV , HCV122KO exhibited more potent replication comparable to that by the overexpression of miR-122 , as previously reported [28] in uterus-derived Hec1B cells ( Fig 5B ) and other non-hepatic cell lines , including A549 ( lung ) , MC-IXC ( neuron ) , Caki-2 ( kidney ) and RERF-LC-AI ( lung ) ( S13 Fig ) and primary HMSC with overexpression of CLDN1 and OCLN ( S14 Fig ) . However , no infectious particles were detected in the culture supernatants of the non-hepatic cell lines infected with HCV122KO . It has been reported that ApoE plays crucial roles in the efficient production of infectious particles in 293T cells expressing Claudin1 ( 293T-CLDN1 ) [37] . Although co-expression of miR-122 and ApoE is required for the production of infectious particles in 293T-CLDN1 cells upon infection with HCV , expression of ApoE alone permits particle formation in cells infected with HCV122KO ( Fig 5C and 5D ) . In addition , the replications of HCV and HCV122KO in non-hepatic Hec1B and 293T-CLDN1 cells were also inhibited by the NS3-4A inhibitor BILN as observed in Huh7-122KO cells ( S15 Fig ) , suggesting that HCV122KO can propagate in non-hepatic cells in the absence of miR-122 . To clarify the molecular mechanisms of the miR-122-independent propagation of HCV122KO , we first confirmed the introduction of adaptive mutations into the viral genome of three independently isolated HCV122KO by direct sequencing analyses . The mutation of G28A in the 5’UTR of HCV was identified in all independently isolated HCV propagated in 751-122KO cell clones ( 751-122KO-#1~#3 ) ( Fig 6A ) . To analyze this result in more detail , deep sequencing analysis of a mixture of six independently isolated HCV122KO was performed . Deep sequencing by Pac-bio enables long-reads ( ~3 kbp ) and detects single nucleotide variants ( SNVs ) including accidental sequencing errors . Compared with HCV at five serial passages in Huh7 . 5 . 1 cells ( JFH-P5 ) , only the dominant mutation G28A was observed from a mixture of six independently isolated HCV122KO ( Fig 6B and S16A–S16C Fig ) . We also identified a few synonymous substitutions at nucleotide position ( nt ) 4579 in NS3 , nt7658 in NS5A and nt7888 or nt7951 in NS5B at a low rate; however , they were from different fragments of cDNA of HCV-RNA . These results indicated that , except in the case of G28A , these mutations arise irregularly , and are not specific to an miR-122-deficient condition . In addition , G28A mutation also emerged by the transfection of a plasmid encoding a full-genomic HCV cDNA , pHH-JFH1-E2p7NS2mt , into 751-122KO cells ( S17A and S17B Fig ) , while there was no adaptive mutation in HCV SGR RNA from replicon cells ( S17C Fig ) . Therefore , we examined the difference between the G28A virus and HCV122KO infectivity in Huh7-122KO and Huh7-122KOR cells ( S18 Fig ) . Compared to the wild type HCV , both the G28A and HCV122KO virus showed miR-122-independence , while HCV122KO had higher infectivity than the G28A virus both in Huh7-122KO and Huh7-122KOR cells . Therefore , we concluded that not only the G28A mutation responsible for miR-122-independence but also other adaptive mutations involved in higher replicative fitness are required for efficient propagation in an miR-122-independent manner . Because genotype 1b originally has adenine at nt29 , which corresponds to the position of G28A in genotype 2 , we tried to identify adaptive mutations that could facilitate efficient propagation in an miR-122-independent manner by using the chimeric virus of genotype 1b Con1 and 2a JFH1 strain ( Con1C3/JFH ) at the C3 loop in NS2 , which includes the 5’UTR of Con1 . Because serial passages of Con1C3/JFH only in 751-122KO cells failed due to the low efficiency of viral replication , both Huh7 . 5 . 1 and 751-122KO cells were used to obtain the adaptive virus . Infectious titers in the supernatants of 751-122KO cells were gradually increased during alternative passages and reached 105 FFU/ml , and the resulting virus was designated Con1C3/JFH122KO ( Fig 7A ) . In addition to HCV122KO , Con1C3/JFH122KO showed efficient propagation both in Huh7 . 5 . 1 and 751-122KO cells ( Fig 7B ) , while the HCV-RNA replication level of Con1C3/JFH122KO was significantly higher than that of Con1C3/JFH both in Huh7 . 5 . 1 cells ( Fig 7C ) and in non-hepatic 293T-CLDN1 cells ( Fig 7D ) . Therefore , we compared Con1C3/JFH and Con1C3/JFH122KO by sequence analysis . Fifteen adaptive mutations were introduced throughout Con1C3/JFH RNA during serial passages ( S19A Fig ) . Especially , G1A in 5’UTR has known to emerge during passages of gt1b replicon [38] or in gt1b patients [39] . Compared with Con1C3/JFH , Con1C3/JFH122KO has additional 3 non-synonymous mutations located at the structural protein-coding region , 3 synonymous mutations at the non-structural protein-coding region and a single nucleotide mutation C30U in 5’UTR ( S19B and S19C Fig ) . These results suggest that Con1C3/JFH122KO also requires adaptive mutation that was directly involved in the role of miR-122 as with G28A mutation in 5’-UTR of JFH1 RNA to acquire miR-122-independence . Previous reports suggest that miR-122 binds to HCV 5’UTR and protects HCV-RNA from degradation by exonucleases Xrn1 or Xrn2 [23 , 24] and Xrn1 plays a dominant role in this process [25] . To determine the roles of exonucleases in the introduction of G28A mutation under an miR-122-deficient condition , we generated three clones each of Xrn1 or both Xrn1 and Xrn2 stable knockdown cells based on miR-122 knockout Huh7 . 5 . 1 cells ( 751-122KO-shXrn1 or -shXrn1/Xrn2 cells ) ( S20A Fig ) . Infectious titers in the supernatants not only of 751-122KO control cells but also of 751-122KO-shXrn1 and 751-122KO-shXrn1/Xrn2 cells were gradually increased during serial passages ( Fig 8A ) with the emergence of G28A mutation ( S20B Fig ) , suggesting that the emergence of G28A mutation in miR-122-deficient cells is independent of RNA decay by exonucleases . Next , to clarify the effect of mutation of G28A in the 5’UTR of HCV122KO on the replication of HCV-RNA , wild type or G28A-mutated JFH-SGR RNA was electroporated into Huh7-122KO and Huh7-122KOR cells . Although colony formation in Huh7-122KO cells transduced with G28A-mutated JFH-SGR RNA was higher than that in the cells transduced with the wild type RNA , there was no significant difference in colony formation between the Huh7-122KOR cells transduced with G28A and those transduced with wild type RNA ( Fig 8B ) . Collectively , these results indicated that G28A mutation conferred a viral growth advantage in the absence of miR-122 by enhancing formation of the replication complex . To examine the effect of G28A mutation on the propagation of HCV , wild type and G28A-mutated JFH1 RNAs were co-electroporated into Huh7 . 5 . 1 and 751-122KO cells ( Fig 8C ) . Although infectious titers in the supernatants of Huh7 . 5 . 1 cells did not change appreciably with the number of passages , those of 751-122KO were increased by cell passaging . The G28A mutant exhibited higher replicative fitness for survival in 751-122KO cells , while replication of the G28A mutant was comparable to that of the wild type in Huh7 . 5 . 1 cells ( Fig 8D ) . In addition , no revertant virus emerged during the eight serial passages of HCV122KO in Huh7 . 5 . 1 cells ( S21 Fig ) . These results suggest that the G28A mutant has a major advantage for propagation under an miR-122-deficient condition , but comparable infectivity to the wild type under an miR-122-rich condition . Several studies have reported that HCV-RNA replication is detected in PBMCs , and malignant lymphoma sometimes occurs in chronic hepatitis C patients [3 , 40–42] , indicating the presence of persistent infection in the non-hepatic tissues of patients . Therefore , to examine whether or not G28A mutation is introduced into HCV-RNA replicating in non-hepatic cells in vivo , we analyzed the 5’UTR sequence of the HCV-RNA in PBMCs and hepatocyte-derived virus in the serum from genotype 2a patients ( n = 36 ) . Among these samples , 15 from PBMCs and 9 from serum showed emergence of the G28A mutation; notably , 7 patients had the G28A mutation in both PBMCs and serum ( Fig 9A–9C ) , suggesting that the G28A mutation is induced dominantly in PBMCs and replication of the G28A virus is also maintained in hepatocytes , as shown by the competition assay ( Fig 8C ) . We also found mutations from G to T ( 1 from serum ) at nt28 , though there were no mutations in the other regions . There were some patients with EHMs such as hypothyroidism ( n = 2; #29 and #31 in Fig 9C ) and malignant lymphoma ( n = 1; #16 in Fig 9C ) . Interestingly , all of them had mutations at G28; the former two patients had the G28A and the latter patient the G28U mutation . Because the HCV122KO genome still contains two miR-122-binding sites in the 5’UTR , we further examined the effect of the miR-122 inhibitor on the propagation of HCV122KO in Huh7-122KO and Huh7-122KOR cells ( Fig 10A ) . Although intracellular viral RNA replication and infectious particle formation in the culture supernatants of Huh7-122KOR cells treated with the LNA control upon infection with HCV and HCV122KO are comparable , those treated with LNA-miR-122 exhibited more potent suppression upon infection with HCV than HCV122KO . Interestingly , propagation of HCV122KO in Huh7-122KOR cells was still significantly suppressed by the treatment with LNA-miR-122 , suggesting that HCV122KO utilizes miR-122 for its propagation , even in the presence of an abundance of miR-122 . Next , to examine the effect of G28A mutation on the interaction with other miRNAs , Ago2 complexes in HCV infected Huh7 . 5 . 1 and 751-122KO cells were immunoprecipitated with anti-Ago2 antibody and the levels of Ago2 and HCV-RNA were determined by immunoblotting and qRT-PCR , respectively at 12 dpi . The interaction of Ago2 with viral genome was detected in Huh7 . 5 . 1 cells infected with HCV , but not in 751-122KO cells ( Fig 10B ) , suggesting that no other miRNA helps to compensate for the role of miR-122 on HCV replication in 751-122KO cells . We observed that replication complex formation was enhanced by G28A mutation , but propagation of HCV122KO was comparable with that of wild type HCV in parental Huh7 cells ( Fig 8D ) . Recently , it has been reported that miR-122 plays a role that rebalances RNA state from protein synthesis to replication by displacing PCBP2 from viral RNA [17] . To further examine the balance between translation and replication of HCV122KO , we analyzed the distribution of polysome in cells infected with either HCV or HCV122KO . Lysates of Huh7-122KO and Huh7-122KOR cells infected with either HCV or HCV122KO were fractionated by a sucrose gradient . The proportion of polysome-free HCV-RNA was higher in Huh7-122KOR cells infected with HCV122KO than in those infected with the wild type ( Fig 11; fractions 7–9 ) , suggesting that G28A mutation promotes replication rather than translation . Although miR-122 has been reported to enhance replication [17] , the proportion of polysome-free HCV-RNA was higher in Huh7-122KO cells than Huh7-122KOR cells upon infection with HCV122KO . These results suggest that G28A mutation facilitates HCV propagation by enhancing formation of the replication complex under an miR-122-deficient condition and by fine-tuning the balance between replication and translation under an miR-122-abundant condition .
HCV exists as a quasispecies due to the low fidelity of its RNA-dependent RNA polymerase , and escape mutants often appear during treatment with anti-HCV drugs . On the other hand , miR-122 interaction sites in the 5’UTR of HCV-RNA are highly conserved . A previous report has shown that the administration of an miR-122 inhibitor to chimpanzees chronically infected with HCV achieved a long-term suppression of viral load without the emergence of a resistant virus or any adverse effects [43] . In addition , a recent phase II clinical trial of an miR-122 inhibitor revealed that it suppressed HCV propagation in chronic hepatitis C patients [26] . However , several studies have reported that adaptive mutations can be introduced around miR-122 interaction sites [44] and substitution of G28A in the 5’UTR of HCV has been selected by serial passages in miR-122-knockdown cells [27] and knockout cells [45] . In this study , we could also obtain an adaptive mutant with G28A mutation in the 5’UTR by adaptation to miR-122 knockout cells , HCV122KO , which was capable of propagating efficiently in miR-122-deficient cells , and analyzed its biological significance . Most of the HCV genotypes , including genotype 1 , originally have adenine at the position corresponding to G28A of genotype 2a . Although Israelow et al . showed that A28 genotypes have lower sensitivity to an miR-122 inhibitor than G28 genotypes [27] , the emergence of adaptive mutation involved in the resistance to miR-122 inhibitor has been reported by the study about gt1b HCV . The C3U mutation adjacent to the miR-122 binding site I in the 5’UTR of the genotype 1b genome introduced during treatment with an miR-122 inhibitor in clinical practice has been shown to be resistant to the miR-122 inhibitor [38 , 39] . Interestingly , in the current study , adaptive mutation C30U was introduced in the 5’UTR of Con1C3/JFH122KO , which has 5’UTR of genotype 1b . Although there is a difference of genome structure between the full-length viral genome and chimeric virus Con1C3/JFH used in our study , adaptive mutations were introduced in miR-122-binding sites . Together with the requirement of several substitutions throughout the genomic RNA of Con1C3/JFH122KO to increase infectivity , these findings suggest that the adaptive mutations required not only for higher replication fitness but also for miR-122-independence such as G28A substitution of gt2a and C3U or C30U of gt1b should be introduced into the HCV genome to allow the virus to replicate efficiently under an miR-122-deficient condition . To examine the role of adaptive mutation C3U or C30U for miR-122-independence in gt1b HCV , further investigation is needed . Based on our previous and current experiments , chronic HCV infection might induce viral replication not only in the liver but also in other non-hepatic tissues [28 , 46] such as PBMCs and neuronal cells [3 , 4] . The current study shows that during persistent infection in non-hepatic PBMCs of HCV gt2a patients , G28A mutation can be introduced into HCV-RNA , in some cases resulting in EHMs such as hypothyroidism and malignant lymphoma . G28A mutation can spontaneously emerge in non-hepatic cells such as PBMCs upon establishment of persistent infection . Once G28A mutation is introduced into viral genome , mutant viruses can replicate efficiently through an enhancement of replication complex formation without interaction of Ago2/miR-122 with HCV-RNA in non-hepatic cells . Most recently , several studies about HCV and lymphoma were reported . Lymphotropic HCV strain ( HCV-SB ) was isolated from HCV gt2b-positive B cell lymphoma with adaptive mutations in 5’UTR and E1/E2 region and B7 . 2 ( CD86 ) was identified as a lymphotropic HCV receptor [47] . Moreover , after HCV treatment with DAA ( sofosbuvir ) and ribavirin to HCV gt3a patient , remission of lymphoma was induced [48] . These observations suggest that low levels of HCV-RNA replication in the non-hepatic tissues or liver with treatment of miR-122 inhibitors may persist in an miR-122-independent manner , resulting in the emergence of HCV mutants that might induce EHMs . Therefore , it is important to control the emergence of miR-122 inhibitors-resistant mutants , which confers a replicative advantage to HCV propagation in miR-122 deficient cells . It has been shown that P-body components consisting of Xrn1 , Lsm1 and PatL1 , which are involved in mRNA decay , participate in translation of the HCV genome [49] . Although previous siRNA-mediated knockdown studies showed that miR-122 contributes to stabilization of the HCV genome by protecting it from exonuclease Xrn1 or Xrn2 [23 , 24] , HCV-RNA replication was not affected by Xrn1 knockdown in Huh7 cells [23] , and the ability of miR-122 to promote HCV infectivity was shown to be independent of the protection of viral RNA from Xrn1 by a reconstituted in vitro system [50] . In this study , we observed no significant effect of Xrn1 and Xrn2 on the emergence of G28A mutation , suggesting that HCV-RNA decay by exonuclease was not involved in G28A-mediated miR-122-independence . On the other hand , Lsm1 was suggested to be involved in the miR-122-dependent enhancement of translation of HCV-RNA in Huh7 cells [51] . Bromo mosaic virus , a single-stranded positive-sense RNA virus , requires the Lsm complex for switching from primary translation to viral RNA replication via viral protein-mediated trafficking of viral RNA to the ER [52] . Similarly , a recent study showed the importance to the HCV life cycle of a switching system in which miR-122 stimulates RNA synthesis by altering the balance of replication and translation [17] . PCBP2 , which binds to the stem-loop I in the 5’UTR of HCV [53] , is predicted to be required for the circularization of viral RNA through direct binding to the 5’ and 3’ ends of viral RNA , but this interaction was inhibited by miR-122 [54] followed by a switch from translation to viral RNA replication [17] . These reports suggest that the inefficient replication of HCV in miR-122-deficient cells may be attributable to degradation of viral RNA and a delay in the switching from translation to replication . We have shown that HCV replicon RNA possessing G28A mutation exhibits efficient colony formation in the absence of miR-122 . In addition , a competition assay revealed that HCV122KO carrying a mutation of G28A in the 5’UTR exhibits a growth advantage compared to wild type HCV in the absence but not in the presence of miR-122 , suggesting that G28A mutation in the 5’UTR of HCV confers at least limited ability to propagate under an miR-122-deficient condition . Polysome analysis clearly showed the excessive replication status in the G28A mutant in miR-122-deficient cells , resulting in efficient propagation . Although a system for equilibrating PCBP2 and the miR-122-Ago2 complex exists , the role of miR-122 on the G28A mutant is to moderate excessive replication but enhance translation , which is the opposite of its role on the wild type virus . These observations indicate that miR-122 may fine-tune the balance between translation and replication not only in the wild type but also the G28A mutant and thereby lead to efficient propagation . In summary , we have shown that HCV can replicate in an miR-122-independent manner . This finding indicates that treatment of hepatitis C patients with an inhibitor for miR-122 would be effective to eliminate HCV in the liver but not in non-hepatic tissues , and may induce the emergence of adaptive mutants capable of propagating in miR-122-deficient cells , leading to the induction of EHMs .
cDNA clones of wild type miR-122 ( WT-miR-122 ) and AcGFP were inserted between the XhoI and XbaI sites of a lentiviral vector , pCSII-EF-RfA , which was provided by M . Hijikata , and the resulting plasmids were designated pCSII-EF-WT-miR-122 and pCSII-EF-AcGFP , respectively . pHH-JFH1 and pJFH1 encoding full-length cDNA and pSGR-JFH1 encoding subgenomic cDNA of the JFH1 strain [55 , 56] were kindly provided by T . Wakita . pHH-JFH1-E2p7NS2mt and pJFH1-E2p7NS2mt contain three adaptive mutations in pHH-JFH1 and pJFH1 [57] , respectively . pCon1/C3/JFH1 , which encodes Con1/JFH1 ( genotype 1b/2a ) chimeric virus genome , was fused at the first transmembrane domain of NS2 ( C3; described by Pietschmann et al . [58] ) . pmirGLO-1 ( pmirGLO-compl-miR-122 ) was described previously [28] . The complementary sequences of miR-122-5p were introduced into the multicloning site of the pmirGLO vector ( Promega ) , and the resulting plasmid was designated pmirGLO-2 . The plasmid pX330 ( Addgene plasmid 42230 ) designed for the CRISPR-Cas9 system [59 , 60] was provided by Addgene . The plasmids used in this study were confirmed by sequencing with an ABI PRSM 3130 genetic analyzer ( Life Technologies , Tokyo , Japan ) . All cell lines were cultured at 37°C under the conditions of a humidified atmosphere and 5% CO2 . Human hepatocellular carcinoma cell line Huh7 , human embryonic kidney cell line 293T and human endometrial adenocarcinoma cell line Hec1B were obtained from Japanese Collection of Research Bioresources ( JCRB ) Cell Bank ( JCRB0403 , JCRB9068 and JCRB1193 ) . Human lung squamous-cell carcinoma cell line RERF-LC-AI cells was provided by the RIKEN BRC through the National Bio-Resource Project of MEXT , Japan ( RCB0444 ) . The human lung adenocarcinoma epithelial cell line A549 , human renal cell carcinoma cell line Caki-2 and human neuroepithelioma cell line MC-IXC were obtained from the American Type Culture Collection ( ATCC CCL-185 , ATCC HTB-47 , and CRL-2270 ) . The Huh7-derived hepatocellular carcinoma cell line Huh7 . 5 . 1 was provided by F . Chisari . Human marrow stromal cells ( HMSCs ) were obtained from Cell Applications Inc . ( San Diego , CA ) . Except for HMSC , all cell lines were maintained in Dulbecco’s modified Eagle’s medium ( DMEM ) ( Sigma , St . Louis , MO ) supplemented with 100 U/ml penicillin , 100 μg/ml streptomycin , and 10% fetal bovine serum ( FBS ) . HMSCs were maintained in MF-medium with 1% FBS ( Toyobo , Japan ) . Replicon cell lines harboring HCV-RNA were maintained in DMEM containing 10% FBS and 1 mg/ml G418 ( Nakalai Tesque , Kyoto , Japan ) . pHH-JFH1-E2p7NS2mt was transfected into Huh7 . 5 . 1 cells , and the culture supernatants were collected after serial passages . Infectivity of HCV was determined by focus-forming assay and expressed in focus-forming units ( FFU ) [56] . Unless otherwise noted , cells were infected with HCV at an MOI of 1 . The lentiviral vectors and ViraPower Lentiviral Packaging Mix ( Life Technologies , San Diego , CA ) were co-transfected into 293T cells and the supernatants recovered at 48 h post-transfection were centrifuged at 1000 x g for 5 min and cleared through a 0 . 45 μm filter . The infectious titer of lentivirus was determined by a Lenti-X qRT-PCR Titration Kit ( Clontech , Mountain View , CA ) . The vesicular stomatitis virus ( VSV ) variant NCP12 . 1 , derived from the Indiana strain , was provided by M . Whitt . The pseudotype VSVs bearing the HCV E1 and E2 glycoproteins ( HCVpv ) and VSV G protein ( VSVpv ) were prepared as described previously [61] . Infectivity of the pseudotype viruses was assessed by the expression of luciferase as determined by a Bright-Glo Luciferase assay system ( Promega ) , following a protocol provided by the manufacturer and expressed in relative light units ( RLU ) . Thirty-six HCV gt2a patients being treated at Hiroshima University Hospital were enrolled in this study . PBMCs were isolated using Ficoll-Hypaque density gradient centrifugation from a patient . The study was approved by the Ethical Committee of the Research Institute for Microbial Diseases , Osaka University and Hiroshima University of Medicine . Written informed consent was obtained from all enrolled patients . Mouse monoclonal antibodies to HCV NS5A and β-actin were purchased from Austral Biologicals ( San Ramon , CA ) and Sigma-Aldrich , respectively . Rabbit anti-HCV core antibody was prepared as described previously [62] . Mouse monoclonal antibodies to HCV core ( C7-50 ) were purchased from Thermo Fisher Scientific ( Waltham , MA ) . Anti-human CD81 ( hCD81 ) monoclonal antibody ( JS-81 ) was purchased from BD Biosciences ( Franklin Lakes , NJ ) . Mouse anti-double-stranded RNA ( dsRNA ) IgG2a ( J1 and K2 ) antibodies were from English and Scientific Consulting Kft . ( Szirak , Hungary ) . Mouse anti-IgG was purchased from Jackson ImmunoResearch ( West Grove , PA ) . Mouse monoclonal anti-Ago2 was purchased from Abcam ( Cambridge , UK ) . Rabbit polyclonal anti-Xrn1 and anti-scavenger receptor class B type1 ( SR-BI ) antibodies were purchased from Novus Biologicals ( Littleton , CO ) . Rabbit polyclonal anti-Xrn2 was purchased from Proteintech ( Rosemont , IL ) . Rabbit anti-CLDN1 and anti-occludin ( OCLN ) antibodies , Alexa Fluor ( AF ) 488-conjugated anti-rabbit IgG , AF594-conjugated anti-mouse IgG and rabbit anti-CLDN1 antibodies were from Life Technologies . Mouse anti-apolipoprotein E ( ApoE ) antibody was purchased from Santa Cruz ( Santa Cruz , CA ) . Human recombinant IFNα was purchased from PBL Biomedical Laboratories ( Piscataway , NJ ) . The HCV NS3-4A protease inhibitor BILN was purchased from Acme Bioscience ( Salt Lake City , UT ) . The NS5A inhibitor BMS-790052 and NS5B inhibitor PSI-7977 were purchased from Shanghai Haoyuan Chemexpress ( Shanghai , China ) . Anti-CD81 antibody ( 5 ng/ml ) was pre-treated before HCV infection and IFNα ( 100U/ml ) , BILN ( 0 . 5 μM ) , BMS-790052 ( 0 . 5 nM ) and PSI-7977 ( 100 nM ) were treated at 2 h post-transfection . BODIPY 558/568 lipid probe and DAPI ( 4’ , 6-diamidino-2-phenylindole ) were purchased from Life Technologies and Vector Laboratories Inc . ( Burlingame , CA ) , respectively . Locked nucleic acids ( LNA ) complementary to miR-122 ( 5′-CcAttGTcaCaCtCC-3′; LNA-miR-122 ) and its negative control ( 5’-CcAttCTgaCcCtAC-3’; LNA-Cont ) ( LNA in capitals , DNA in lowercase; sulfur atoms in oligonucleotide phosphorothioates are substituted for non-bridging oxygen atoms; capital C indicates LNA methylcytosine ) [63] were purchased from Gene Design ( Osaka , Japan ) and transfected into cells using Lipofectamine RNAi MAX ( Life Technologies ) according to the manufacturer’s protocol ( reverse transfection protocol ) . The miScript miRNA mimic hsa-miR-122 and its negative control were purchased from Qiagen ( Valencia , CA ) . Total RNA was prepared from cells by using an RNeasy mini kit ( Qiagen ) . For quantitation of HCV-RNA , quantitative RT-PCR was performed by using TaqMan EZ RT-PCR Core Reagents and a ViiA7 system ( Life Technologies ) according to the manufacturer’s protocol . For quantitation of gene expression , the synthesis of the first-stranded cDNA was performed by using a PrimeScript RT Reagent Kit ( Perfect Real Time ) ( Takara Bio ) and quantitative RT-PCR was performed by using Platinum SYBR Green qRT-PCR SuperMix UDG ( Life Technologies ) according to the manufacturer’s protocol . ApoB , ApoE and MTTP were amplified using the primer pairs described previously [28] . For quantitation of miRNA , total RNA was prepared from cells by using an miRNeasy mini kit ( Qiagen ) and miR-122 was determined by using miR-122-specific RT primers and amplified by using specific primers provided in the Taqman MicroRNA Assays ( Life Technologies ) according to the manufacturer’s protocol . U6 small nuclear RNA ( snRNA ) was used as an internal control . Fluorescent signals were analyzed by using a ViiA7 system ( Life Technologies ) . Cells were transfected with the plasmids by using Trans IT LT-1 transfection reagent ( Mirus , Madison , WI ) according to the manufacturer’s protocols . The cells were lysed on ice in Triton lysis buffer ( 20 mM Tris-HCl ( pH 7 . 4 ) , 135 mM NaCl , 1% Triton-X 100 , 10% glycerol ) supplemented with a protease inhibitor mix ( Roche ) . The samples were boiled in loading buffer and subjected to 5–20% gradient sodium dodecyl sulfate-polyacrylamide gel electrophoresis ( SDS-PAGE ) . The proteins were transferred to polyvinylidene difluoride membranes ( Millipore , Bedford , MA ) , and reacted with primary antibody and then secondary horseradish peroxidase-conjugated antibody . The immunocomplexes were visualized with Super Signal West Femto substrate ( Pierce , Rockford , IL ) and detected by using an LAS-3000 image analyzer ( Fujifilm , Tokyo , Japan ) . Left and Right Custom TALEN ( Cellectis , Paris , France ) targeting miR-122 seed sequences were designed by TALEN Hit Search ( TTCCTTAGCAGAGCTGTGGAGTGTGACAATGGTGTTTGTGTCTAAACTA; the TALEN targeting sequence is underlined and the left or right TAL effector DNA-binding domains are on either side of the target ) , and the corresponding expression plasmids were purchased from Takara Bio ( Shiga , Japan ) . Two TALEN targeting miR-122 RNAs were synthesized by using an mMESSAGE mMACHINE T7 Ultra kit ( Life Technologies ) according to the manufacturer’s protocol . Huh7 cells and Huh7 . 5 . 1 cells were seeded onto 6-well plates at the concentration of 2x105 cells/well and transfected with 2 μg of each of the Custom TALEN RNAs using Lipofectamine 2000 ( Life Technologies ) . The culture medium was replaced with fresh DMEM containing 10% FBS at 4 h post-transfection , incubated at 30°C for 48 h and then shifted to 37°C . When the cells became confluent , single cells were seeded onto 96-well plates and total DNA was isolated from each clone after it had grown to maturity . PCR products amplified by using a primer set ( miR-122-F: 5’-CAAGATGCTTGTACCCGTGA-3’; miR-122-R: 5’-GTGCCTGGTCTGCAATCTTT-3’ ) were denatured , annealed , and then treated with SURVEYOR ( Transgenomic ) nuclease or sequenced after cloning into a pGEM-Teasy vector ( Promega ) . A lentiviral vector that expresses a short hairpin RNA ( shRNA ) was generated as previously reported [64] by using pFTRE3G_pGK_GFP [65] . Briefly , the U6 promoter and pGK_Hygro cassette were inserted between PacI and AscI sites of pFTRE3G_pGK_GFP and the resulting plasmid was designated FU6_pGKhygro . To knockdown the Xrn1 and Xrn2 genes , oligonucleotides ( shXrn1-s: 5’-GATCCGAGGTGTTGTTTCGCATTATTCAAGAGATAATGCGAAACAACACCTCTTTTTTG-3’; shXrn1-as: 5’-AATTCAAAAAAGAGGTGTTGTTTCGCATTATCTCTTGAATAATGCGAAACAACACCTCG-3’ , shXrn2-s: 5’-GATCCGAACCGAACTTTACCATTATTCAAGAGATAATGGTAAAGTTCGGTTCTTTTTTG-3’; shXrn2-as: 5’-AATTCAAAAAAGAACCGAACTTTACCATTATCTCTTGAATAATGGTAAAGTTCGGTTCG-3’ ) were annealed and inserted into the EcoRI and BamHI sites of the FU6_pGKHygro vector , respectively , and used for generating lentivirus . 751-122KO#1 cells were infected with lentivirus and then cultured in DMEM with 200 μg/ml Hygromycin . Each cell clone was used for Western blotting . Cells cultured on glass slides were fixed with 4% paraformaldehyde in phosphate buffered saline ( PBS ) at room temperature for 30 min . After washing three times with PBS , the cells were permeabilized for 20 min at room temperature with PBS containing 0 . 2% Triton-X and blocked with phosphate buffer containing 2% bovine serum albumin ( BSA ) for 1 h at room temperature . The cells were incubated with blocking buffer containing rabbit anti-NS5A or rabbit anti-core at room temperature for 1 h , washed three times with PBS and incubated with blocking buffer containing appropriate AF488-conjugated and AF594-conjugated secondary antibodies at room temperature for 1 h . Finally , the cells were washed three times with PBS and observed with a FluoView FV1000 laser scanning confocal microscope ( Olympus , Tokyo , Japan ) . Cell growth was determined by the Cell Titer-Glo Luminescent Cell Viability Assay ( Promega ) according to the manufacturer’s protocol and expressed in RLU at 24 , 48 and 72 h post-seeding . Total RNAs extracted from cells were purified by using an miRNeasy Kit ( Qiagen , Valencia , CA ) according to the manufacturer’s protocol . Eluted RNAs were quantified using a Nanodrop ND-1000 v3 . 5 . 2 spectrophotometer ( Thermo Scientific , Wartham , MA ) . RNA integrity was evaluated using the RNA 6000 LabChip and Bioanalyzer ( Agilent Technologies , Santa Clara , CA ) . Each RNA that had an RNA integrity number ( RIN ) greater than 9 . 0 was used for the microarray experiments . Expression profiling was generated using the 4 x 44 K whole human genome oligo-microarray ver . 2 . 0 G4845A ( Agilent Technologies ) . Each microarray uses 44 , 495 probes to interrogate 27 , 958 Entrez gene RNAs . One-hundred nanograms of total RNA were reverse-transcribed into double-strand cDNAs by AffinityScript multiple-temperature reverse-transcriptase and amplified for 2 h at 40°C . The resulting cDNAs were subsequently used for in vitro transcription by the T7 RNA polymerase and labeled with cyanine-3-labeled cytosine triphosphate ( Perkin Elmer , Waltham , MA ) for 2 h at 40°C using a Low Input Quick-Amp Labeling Kit ( Agilent Technologies ) according to the manufacturer’s protocol . After labeling , the rates of dye incorporation and quantification were measured with a Nanodrop ND-1000 v3 . 5 . 2 spectrophotometer ( Thermo Scientific ) and then were fragmented for 30 min at 60°C in the dark . Differentially labeled samples of 1650 ng of cRNA were hybridized on Agilent 4 x 44K whole genome arrays ver . 2 . 0 ( Agilent Design #026652 ) at 65°C for 17 h with rotation in the dark . Hybridization was performed using a Gene Expression Hybridization Kit ( Agilent Technologies ) following the manufacturer’s instructions . After washing in GE washing buffer , each slide was scanned with an Agilent Microarray Scanner G2505C . Feature extraction software ( Version 10 . 5 . 1 . 1 ) employing defaults for all parameters was used to convert the images into gene expression data . Raw data were imported into Subio platform ver . 1 . 12 ( Subio ) for database management and quality control . Raw intensity data were normalized against GAPDH expression levels for further analysis . These raw data have been accepted by GEO ( a public repository for microarray data , aimed at storing MIAME [Minimum Information About Microarray Experiment] ) . Access to data for this study may be found under GEO experiment accession number GSE32886 . Cells were cultured on a Cell Desk polystyrene coverslip ( Sumitomo Bakelite , Osaka , Japan ) , and were fixed with 2% formaldehyde and 2 . 5% glutaraldehyde in 0 . 1 M cacodylate buffer ( pH 7 . 4 ) containing 7% sucrose . Cells were post-fixed for 1 h with 1% osmium tetroxide and 0 . 5% potassium ferrocyanide in 0 . 1 M cacodylate buffer ( pH 7 . 4 ) , dehydrated in a graded series of ethanol and embedded in Epon812 ( TAAB ) . Ultrathin ( 80 nm ) sections were stained with saturated uranyl acetate and lead citrate solution . Electron micrographs were obtained with a JEM-1011 transmission electron microscope ( JEOL , Tokyo , Japan ) . Correlative FM-EM allows individual cells to be examined both in an overview with FM and in a detailed subcellular-structure view with EM . The NS5A was stained and observed in the HCV-infected cells by the correlative FM-EM method as described previously [66] . The plasmids pSGR-JFH1 , pJFH1-E2p7NS2mt and pCon1/C3/JFH1 were linearized with XbaI and then treated with Mung bean exonuclease . The linearized DNAs were transcribed in vitro by using the MEGAscript T7 kit ( Life Technologies ) according to the manufacturer’s protocol . The in vitro transcribed RNA ( 5 μg ) was electroporated into each cell at 5x106 cells/0 . 4 ml under conditions of 190V and 950 μF using a Gene Pulser ( Bio-Rad , Hercules , CA ) and plated on DMEM containing 10% FBS . For colony formation , the medium was replaced with fresh DMEM containing 10% FBS and 1 mg/ml G418 at 24 h post-electroporation . The remaining colonies were cloned by using a cloning ring ( Asahi Glass , Tokyo , Japan ) or fixed with 4% paraformaldehyde and stained with crystal violet at 21 days post-electroporation . Cells were suspended with IP buffer ( 25 mM Tris-HCl ( pH 7 . 4 ) , 150 mM KCl , 5 mM EDTA , 5 mM DDT , RNase inhibitor ( 100 U/ml ) , protease inhibitor ( Roche ) ) and placed on ice for 15 min , followed by sonication for 2 min . Cell lysates were collected and centrifuged for 10 min at 2 , 000 rpm at 4°C . Supernatants were preincubated with Protein G Sepharose ( GE Healthcare ) for 1 h and anti-Ago2 or control IgG at 4°C for 2 h followed by addition of 30 μl of Protein G Sepharose ( GE Healthcare ) for 1 h . The Sepharose beads were washed three times in PBS and RNAs were extracted using the Qiazol reagent ( Qiagen ) . HCV-RNA associated with Ago2 protein was detected by qRT-PCR as described as above . For detection of miRNA , total RNA was prepared from cells by using an miRNeasy mini kit ( Qiagen ) and samples were subjected to 15% TBE-Urea polyacrylamide gel electrophoresis after boiling in loading buffer . The total RNAs were transferred to nylon membranes ( Roche , Mannheim , Germany ) , and miR-122 was detected by using an miR-122-specific 5’- digoxigenin ( DIG ) -labeled miRCURY LNA detection probe ( Exiqon , Vedbaek , Denmark ) and visualized with a DIG luminescence detection kit ( Roche ) according to the manufacturer’s protocol . RNA was extracted from 100 μl of virus-containing culture supernatants after 5 passages in Huh7 . 5 . 1 cells ( JFH-P5 ) or a mixture of those after 5 and 4 passages in 751-122KO#1 and 751-122KO#2 cells , respectively ( 6 x 122KO ) . The first-stranded cDNA was synthesized by using a PrimeScript 1st strand cDNA Synthesis Kit ( Takara Bio ) and three fragments of the HCV genome region were amplified . PacBio DNA libraries were prepared from three pooled fragments ( each 100 ng ) of the respective HCV genomes using a DNA Template Prep Kit 2 . 0 ( 3–10 kbp ) ( Pacific Biosciences ) according to the manufacturer’s instructions . Sequencing was performed by the PacBio RS II system with a 240 min movie using the DNA Sequencing Kit 4 . 0 ( Pacific Biosciences ) with P6 polymerase . Circular consensus sequences ( CCS ) constructed from more than four full-pass subreads were produced through PacBio SMRT analysis . For a rapid identification of 5’UTR sequence of HCV , RNA was extracted from 100 μl of virus-containing supernatants or PBMCs . The first-stranded cDNA was synthesized by using a PrimeScript RT reagent Kit ( Perfect Real Time ) ( Takara Bio ) and the 5’UTR of HCV was amplified . 5’RACE was performed by using a 5'RACE System for Rapid Amplification of cDNA Ends , Version 2 . 0 ( Life Technologies ) as described by Li et al . [44] with modification ( see S2 Table ) . Polysome analysis was performed as described by Masaki et al . [17] with some modifications . In brief , three 10-cm dishes containing 5x105 of Huh7-122KO or Huh7-122KOR cells were infected with HCV or HCV122KO . At 3 dpi , the cells were incubated with CHX ( 100 μg/ml ) for 10 min . After washing with PBS with CHX ( 100 μg/ml ) , the cells were harvested from three dishes ( total number of cells: ~2x107 cells/dish ) by using a cell scraper and centrifuged for 5 min at 1 , 400 rpm at 4°C . Cell pellets were suspended in 500 μl of polysome lysis buffer ( PLB; 140 mM KCl , 5 mM MgCl2 , 20 mM Tris-HCl ( pH 7 . 4 ) , 0 . 01% Triton X-100 , 10 mM DTT , 100 μg/ml CHX ) with RNase inhibitor ( 100 U/ml ) and passaged 5 times with a 27-gauge needle on ice . Cell pellets were removed by two 5-min rounds of centrifugation at 13 , 000 rpm and 4°C . The supernatant was layered on the top of a linear 10%-50% sucrose gradient in PLB and centrifuged in an SW41Ti rotor ( Beckman Coulter , CA , USA ) for 2 h at 32 , 000 rpm at 4°C ( no brake ) . The absorbances at OD254 of the 76 fractions collected from the top by using a Piston Gradient Fractionator ( BioComp , NB , Canada ) were determined and divided into 19 fractions for quantification of HCV-RNA and β-actin mRNA by qRT-PCR as described above . The data for statistical analyses are the average of three independent experiments . Results were expressed as the means ± standard deviation . The significance of differences in the means was determined by Student’s t-test . | A liver-specific microRNA , miR-122 , is one of the key determinants of hepatitis C virus ( HCV ) hepatotropism and is required for efficient propagation of HCV . On the other hand , chronic infection with HCV is often associated with extrahepatic manifestations ( EHMs ) , and a low level of HCV-RNA replication has been detected in some non-hepatic cells . Nonetheless , the detailed mechanisms underlying these phenomena remain unknown . Here , we show that miR-122 is dispensable for low-level replication or infectious particle formation , and a mutant virus adapted to miR-122-knockout cells exhibited efficient but miR-122-independent propagation . The adaptive virus of HCV genotype 2a possessed a G28A substitution in the 5’UTR and facilitated efficient replication complex formation under an miR-122-deficient condition , while it propagated at a level comparable to the wild type HCV in the presence of miR-122 . Moreover , various adaptive mutations including C30U were introduced into genotype 1b , which originally had an adenine at the nucleotide position 29 . These observations suggest that substitutions that yield miR-122-independent propagation are not induced during propagation in hepatocytes; however , treatment with an miR-122 inhibitor or persistent infection of HCV in non-hepatic cells may induce the emergence of mutant viruses , as evidenced by clinical samples . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"sequencing",
"techniques",
"medicine",
"and",
"health",
"sciences",
"microbial",
"mutation",
"pathology",
"and",
"laboratory",
"medicine",
"hepacivirus",
"pathogens",
"rna",
"extraction",
"microbiology",
"cloning",
"viruses",
"mutation",
"substitution",
"mutation",
"rna",
"viruses",
"molecular",
"biology",
"techniques",
"rna",
"sequencing",
"extraction",
"techniques",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"rna",
"sequence",
"analysis",
"bioinformatics",
"medical",
"microbiology",
"microbial",
"pathogens",
"hepatitis",
"c",
"virus",
"hepatitis",
"viruses",
"viral",
"replication",
"molecular",
"biology",
"flaviviruses",
"virology",
"viral",
"pathogens",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"organisms"
]
| 2017 | Characterization of miR-122-independent propagation of HCV |
Clostridium difficile infection affects a significant number of hospitalized patients in the United States . Two homologous exotoxins , TcdA and TcdB , are the major virulence factors in C . difficile pathogenesis . The toxins are glucosyltransferases that inactivate Rho family-GTPases to disrupt host cellular function and cause fluid secretion , inflammation , and cell death . Toxicity depends on receptor binding and subsequent endocytosis . TcdB has been shown to enter cells by clathrin-dependent endocytosis , but the mechanism of TcdA uptake is still unclear . Here , we utilize a combination of RNAi-based knockdown , pharmacological inhibition , and cell imaging approaches to investigate the endocytic mechanism ( s ) that contribute to TcdA uptake and subsequent cytopathic and cytotoxic effects . We show that TcdA uptake and cellular intoxication is dynamin-dependent but does not involve clathrin- or caveolae-mediated endocytosis . Confocal microscopy using fluorescently labeled TcdA shows significant colocalization of the toxin with PACSIN2-positive structures in cells during entry . Disruption of PACSIN2 function by RNAi-based knockdown approaches inhibits TcdA uptake and toxin-induced downstream effects in cells indicating that TcdA entry is PACSIN2-dependent . We conclude that TcdA and TcdB utilize distinct endocytic mechanisms to intoxicate host cells .
Clostridium difficile , a gram-positive , spore-forming anaerobe , is the most common cause of healthcare-associated infections and gastroenteritis-associated death in the United States [1–3] . The pathogenesis of C . difficile is mediated by two large homologous exotoxins , TcdA and TcdB ( 308 kDa and 270 kDa , respectively ) , capable of causing epithelial cell death , fluid secretion and inflammation [4] . Recent studies , using isogenic single and double toxin knockout strains , have shown that either TcdA or TcdB alone can cause disease in animal models , with TcdB linked to severe disease phenotypes [5–7] . Most pathogenic isolates produce TcdA and TcdB emphasizing the need to consider both toxins when developing C . difficile therapeutics [8 , 9] . TcdA and TcdB are broadly classified as AB toxins , wherein a B subunit is involved in the delivery of an enzymatic A subunit into the cytosol of a target cell . For C . difficile toxins , the A subunit is an N-terminal glucosyltransferase domain ( GTD ) that inactivates small GTPases , such as RhoA , Rac1 and Cdc42 [10 , 11] . The B subunit is composed of the combined repetitive oligopeptides ( CROPs ) domain , delivery/pore-forming and autoprotease domains . The CROPs has been proposed to function as the receptor-binding domain because it can bind cell surface carbohydrates [12–14] , and antibodies against the CROPs region of TcdA and TcdB can neutralize toxicity [15–17] . However , recent studies reveal that toxins lacking the CROPs domain can still bind , enter and perturb host cellular function , highlighting the presence of alternative or additional receptor binding regions within the toxins [18–21] . Upon binding to cells , toxins are taken up by endocytosis and transported to acidified endosomal compartments [4] . Acidification is thought to trigger a conformational change in the delivery domain , allowing it to insert into the membrane of the endosome and form a pore through which the enzymatic domains can be translocated [18 , 22 , 23] . Once inside the cytosol , host inositol hexakisphosphate binds the autoprotease domain to induce cleavage and release of the GTD [24] . The GTD transfers a glucose from UDP-glucose onto the switch I region of Rho family GTPases . This inactivation results in perturbation of the actin cytoskeleton and cell rounding ( cytopathic effect ) as well as apoptotic cell death ( cytotoxic effect ) [25–28] . At higher concentrations , TcdB is also capable of inducing aberrant production of reactive oxygen species , resulting in cell death by necrosis [29 , 30] . Despite their homology , TcdA and TcdB appear to engage different receptors on the cell surface . Multiple receptors have been proposed for TcdA , including Galα1-3Galβ1-4GlcNac , rabbit sucrase-isomaltase and glycoprotein 96 [31–33] . Three recent studies have shown that poliovirus receptor-like protein 3 , chondroitin sulfate proteoglycan 4 , and frizzled proteins can function as TcdB receptors on epithelial cells [19 , 34 , 35] . Receptor binding by TcdB is followed by internalization via clathrin-dependent endocytosis [36] , but the mechanism ( s ) by which TcdA enters cells has been less clear [36 , 37] . In this study , we investigated TcdA cellular uptake by systematically perturbing the function of key host factors involved in various endocytic pathways using RNAi-based knockdown approaches and small molecule inhibitors , and by analyzing the toxin colocalization with markers of endocytic pathways by confocal microscopy . Our results indicate that cellular uptake of TcdA is mediated by a PACSIN2- and dynamin-dependent pathway and does not involve clathrin- or caveolae-mediated endocytosis .
We first examined whether TcdA-induced cytotoxicity in colonic epithelial cells requires clathrin-mediated endocytosis ( CME ) . Human colorectal adenocarcinoma ( Caco-2 ) cells were transduced with non-targeting shRNA ( ctrl shRNA ) and shRNAs ( sh489 and sh887 ) targeting two different sequences in the clathrin heavy chain ( CHC ) . Expression of sh489 resulted in greater than 90% reduction in CHC protein levels , whereas sh887 did not alter CHC levels in cells ( Fig 1A; inset ) . We challenged these shRNA-expressing cells with TcdA and TcdB concentrations ranging from 100 pM to 100 nM and assayed for cellular viability using CellTiterGlo . As expected , cells expressing sh489 ( that were depleted of CHC ) showed increased survival relative to cells expressing ctrl shRNA or sh887 when challenged with TcdB ( Fig 1A; top panel ) . However , depletion of CHC did not affect TcdA-induced cell death across the range of concentrations tested ( Fig 1A; bottom panel ) . A similar observation was made using a transient knockdown of CHC with small interfering RNA ( siRNA ) ( S1 Fig ) . Taken together , these data show that clathrin heavy chain is dispensable for TcdA-induced toxicity in Caco-2 epithelial cells . Our cytotoxicity data suggest that CME may not be required or involved in TcdA entry . To test this , we checked for colocalization of fluorescently labeled TcdA ( TcdA-546 ) with labeled TcdB ( TcdB-647 ) and CHC , markers of clathrin-mediated endocytosis . We verified that fluorescent labeling with Alexa dyes did not affect TcdA function prior to using TcdA-546 in our imaging assays ( S2 Fig ) . For the confocal assays , we intoxicated cells with 50 nM of TcdA-546 as it provided sufficient signal and dynamic range needed for image analyses ( S3 Fig ) . Confocal microscopy revealed minimal to no detectable colocalization of TcdA-546 with TcdB-647 during cell entry ( S4 Fig ) . However , our labeling efficiency and signal intensity for TcdB was poor and less than desirable for colocalization and other imaging-based analyses . The technical challenges associated with obtaining higher labeling efficiencies while maintaining toxin function and internalization prevented us from using TcdB as a control in future immunofluorescence assays . As a result , in our subsequent experiment , transferrin ( Tf-647 ) was used as a positive control for colocalization with CHC . As expected , Tf-647 exhibited significant colocalization with clathrin-positive vesicles ( Fig 1C and 1D ) . However , in similar experiments , TcdA-546 did not colocalize with clathrin-positive structures during cell entry ( Fig 1B and 1D ) . Taken together , our findings support a clathrin-independent mechanism of entry for TcdA and indicate that TcdA and TcdB utilize distinct endocytic mechanisms to intoxicate epithelial cells . Clathrin-independent endocytic ( CIE ) pathways can be dynamin-dependent or -independent [38 , 39] . Dynamin is a large GTPase that facilitates scission and release of newly formed endocytic vesicles from the plasma membrane . To determine if the clathrin-independent uptake of TcdA requires dynamin function , we perturbed dynamin activity in cells by siRNA depletion or pharmacological inhibition and studied the effect on toxin-induced Rac1 glucosylation and cell death . Caco-2 cells were transfected with siRNAs targeting dynamin-1 or luciferase ( non-targeting control ) and subsequently challenged with TcdA . We found that depletion of dynamin-1 improved survival of cells treated with TcdA by at least three-fold compared to the luciferase control ( Fig 2A ) . We verified that dynamin is important for the TcdA cytotoxic mechanism by using dynasore , a potent inhibitor of dynamin GTPases [40] . Dynasore treatment prevented Rac1 glucosylation by TcdB in Caco-2 cells , consistent with the known role of dynamin in CME ( Fig 2B and 2C ) . Furthermore , pretreatment of cells with dynasore completely inhibited Rac1 glucosylation by TcdA ( Fig 2B and 2C ) , supporting our earlier observation that TcdA intoxication is dynamin-dependent . To determine what step in the toxin pathway the inhibitor was affecting , we performed time-of-addition assays . Dynasore was added prior to intoxication ( pretreatment ) , at the same time as toxin ( 0 min ) , or at various times post-intoxication , and Rac1 glucosylation in cells by TcdB ( S5A and S5B Fig ) and TcdA ( S5C and S5D Fig ) was measured . Results from these experiments show that inhibition of toxin-induced glucosylation can be bypassed by adding dynasore 5 to 10 min post-intoxication , suggesting that the inhibitor is acting at the stage of toxin entry . It is important to note that it takes several minutes for dynasore to appreciably inhibit dynamin-dependent pathways [40] , which might explain the glucosylation occurring when the inhibitor and toxin are added together . In summary , our data indicate that TcdA entry and intoxication in epithelial cells require functional dynamin . The above findings indicated that TcdA uptake occurs through a clathrin-independent and dynamin-dependent endocytic mechanism . Dynamin has been implicated or shown to be involved in several CIE pathways such as caveolar endocytosis , the RhoA-dependent pathway , flotillin-dependent endocytosis and endophilinA2-mediated endocytosis ( FEME: fast endophilin-mediated endocytosis ) [38 , 39 , 41 , 42] . To rapidly assess which of these pathway ( s ) , if any , contribute to TcdA uptake , we performed siRNA-mediated depletion of a panel of host factors involved in these uptake mechanisms and examined their impact on TcdA-induced cell death . Caveolae-mediated endocytosis is a commonly studied clathrin-independent and dynamin-dependent pathway [38 , 43] . However , there are conflicting reports regarding the expression of caveolin1 ( Cav1 ) in Caco-2 cells [44–46] . Cav1 is typically expressed as two isoforms , Cav1α and Cav1β , and both isoforms can be found in caveolae [47] . Western blotting and RT-PCR analyses show that Caco-2 cells preferentially express the beta isoform of Cav1 ( S6 Fig ) . Since Caco-2 cells express Cav1 , we decided to include host factors from the caveolar pathway , namely Cav1 , cavin1 and PACSIN2 ( protein kinase C and casein kinase substrate in neurons 2 ) , in our siRNA panel . Results from our siRNA screen show that depletion of flotillin1 or 2 ( flotillin-dependent pathway ) , RhoA ( RhoA-dependent pathway ) or endophilinA2 ( endoA2; FEME pathway ) does not affect TcdA-induced cytotoxicity in Caco-2 cells . However , knockdown of Cav1 , cavin1 or PACSIN2 protected cells from TcdA challenge ( Fig 3 , S7 and S8 Figs ) . This protective effect was not observed for cells treated with TcdB , which enters via CME and was used as a negative control . Cav1 , cavin1 and PACSIN2 are host proteins involved in caveolae-mediated endocytosis [48–52] . Results from our siRNA screen , therefore , lead to the hypothesis that TcdA uptake in Caco-2 cells is mediated by caveolae-dependent endocytosis . However , Vogel et al . had previously shown that Caco-2 cells contain extremely few , if any , caveolae [46] . Consistent with their finding , we observed that the cytoplasmic staining of cavin1 in Caco-2 cells was diffuse and atypical of caveolae-associated pools ( S9A Fig ) . Caco-2 cells appear to lack caveolae despite expressing Cav1β and cavin1 . Furthermore , we did not observe appreciable colocalization of TcdA-546 with Cav1 or cavin1 in Caco-2 cells ( S9A Fig ) . To better understand the contribution of caveolae-mediated endocytosis to TcdA uptake , we decided to investigate TcdA entry in mouse embryonic fibroblast ( MEF ) cells , which are sensitive to TcdA and contain caveolae . Confocal microscopy revealed no detectable colocalization of TcdA-546 with Cav1 or cavin1 in MEFs ( S9B Fig ) . Furthermore , investigation of TcdA-induced cell rounding in wildtype and Cav1-/- MEF cells shows that Cav1 is not required for the TcdA cytopathic mechanism ( S9C and S9D Fig; S1 and S2 Videos ) . A similar observation was made using a transient knockdown of Cav1 in wildtype MEF cells ( S10A and S10B Fig ) . Taken together , our data suggest that TcdA uptake in MEFs is caveolae-independent . To test this directly , we measured toxin uptake in MEFs depleted of Cav1 . MEF cells were transfected with control ( luciferase ) or Cav1 siRNA and allowed to internalize TcdA-546 . Cav1 and TcdA-546 fluorescence intensities in cells were measured and compared between the two conditions to determine knockdown efficiency and extent of toxin uptake . We found that MEFs transfected with Cav1 siRNA showed a 67% decrease in Cav1 fluorescence staining compared to controls ( S10C and S10D Fig ) . The toxin levels in cells remained unaffected by Cav1 depletion , however , supporting the idea of a caveolin-independent uptake mechanism for TcdA in MEFs ( S10C and S10E Fig ) . In summary , our data from both Caco-2 and MEF cells show that caveolae-mediated endocytosis does not contribute to cellular uptake of TcdA . Interestingly , despite the lack of significant colocalization with Cav1-positive vesicles , TcdA-546 colocalized with PACSIN2 in wildtype MEF cells ( Fig 4 ) . PACSIN2/Syndapin-II is a BAR ( Bin/amphiphysin/rvs ) -domain-containing protein that has been shown to interact with dynamin and regulators of actin to induce membrane curvature and the formation of vesicular-tubular invaginations that can promote receptor-mediated endocytosis [53–55] . Our cell imaging studies show that there is a pool of PACSIN2 that colocalizes with Cav1-positive vesicles , consistent with the known role of PACSIN2 in caveolar endocytosis ( Fig 4B and 4C ) . However , the PACSIN2 that colocalizes with TcdA-546 in MEF cells is not Cav1-associated ( Fig 4B and 4C ) . We also found that PACSIN2 depletion , unlike that of Cav1 , protects wildtype MEF cells from TcdA-induced cytopathic effects ( Fig 5A and 5B ) . To test whether PACSIN2 is involved in TcdA entry , we depleted PACSIN2 in wildtype MEFs by using siRNAs and examined the impact on TcdA binding and uptake . We observe that a 61% reduction in PACSIN2 staining has no impact on the overall toxin binding to cells ( S11 Fig ) . However , a 48% reduction in PACSIN2 fluorescence correlates with a 36% decrease in TcdA uptake ( Fig 5C , 5D and 5E ) . Similar uptake assays performed with transferrin , a clathrin-dependent cargo , show that transferrin uptake is not affected by PACSIN2 depletion , and indicates a specific role for PACSIN2 in TcdA entry ( S12 Fig ) . We next checked if PACSIN2 is involved in TcdA uptake in Caco-2 cells , which lack the caveolar pathway . Similar to our findings in MEF cells , we observed significant colocalization of TcdA-546 with PACSIN2-positive structures in Caco-2 cells by confocal microscopy ( Fig 6 ) . Unlike MEF cells , where TcdA colocalizes with PACSIN2 at 3 min , colocalization in Caco-2 cells was strongest at 10 min post-switch to 37°C ( Fig 6B and 6C ) . PACSIN2 has been shown to associate with Rac1 on early endosomes [56] . It is possible that the TcdA-containing PACSIN2 structures in Caco-2 cells are early endosomes and that we are capturing colocalization at the stage of toxin translocation , where TcdA can access Rac1 . To address this , we performed colocalization studies of TcdA-546 with PACSIN2 and early endosomal antigen 1 ( EEA1 ) in Caco-2 cells . As expected , a fraction of PACSIN2 colocalized with EEA1 ( S13 Fig ) . However , these PACSIN2-positive early endosomes were distinct from the toxin-containing PACSIN2 structures observed at 0 , 5 , 10 and 15 min post-switch to 37°C ( S13 Fig ) . Taken together , our data show that TcdA-546 colocalizes with a Cav1- and endosome-independent pool of PACSIN2 in Caco-2 cells . Strong colocalization between TcdA-546 and PACSIN2 and inhibition of TcdA-induced cell death upon PACSIN2 depletion ( Fig 3 and S14 Fig ) suggest that PACSIN2 is required for TcdA entry in Caco-2 cells . To test this , we depleted PACSIN2 and examined the effect on TcdA binding and uptake . Caco-2 cells were transduced with a non-targeting shRNA ( ctrl shRNA ) and shRNA 982 ( sh982 ) targeting PACSIN2 . Expression of sh982 resulted in 94 . 7 ± 2 . 1% reduction in PACSIN2 protein levels by western blotting ( S15 Fig ) . Since TcdA binding to cells might be temperature-sensitive [57] , we performed binding assays at two different conditions ( 10°C and 37°C ) . Irrespective of the temperature , we found that PACSIN2 depletion does not affect TcdA binding to cells ( S15 Fig ) . We then investigated the effect of PACSIN2 depletion on TcdA uptake by imaging-based approaches . Caco-2 cells stably expressing shRNAs were allowed to bind and internalize TcdA-546 and were then stained for PACSIN2 ( Fig 7A ) . Expression of sh982 resulted in a 66% decrease in PACSIN2 fluorescence in cells ( Fig 7C ) . In contrast to control cells , TcdA-546 signal in sh982-expressing cells was typically restricted to the cell periphery suggesting that toxin internalization is inhibited in these cells ( Fig 7A ) . Consistent with that , cells expressing sh982 showed a 44% reduction in TcdA-546 fluorescence compared to cells expressing ctrl shRNA ( Fig 7D ) . Despite similar cell surface binding , the overall levels of cell-associated toxin in PACSIN2-depleted cells were lower than that of controls . PACSIN2 depletion inhibited TcdA entry , and the toxin that is stuck on the outside and unable to enter the cells was likely lost in the subsequent wash steps . We also observed a strong correlation ( R2 = 0 . 8 ) between TcdA-546 and PACSIN2 fluorescence in cells ( Fig 7B ) . Linear regression analyses show that an increase in PACSIN2 fluorescence correlated with a corresponding increase in toxin fluorescence ( and vice versa ) , supporting the conclusion that TcdA uptake is PACSIN2-dependent . Lastly , to evaluate the specificity of our observations with PACSIN2 , we performed colocalization and uptake assays in Caco-2 cells with transferrin ( a clathrin-dependent cargo ) . Confocal assays reveal minimal to no colocalization between transferrin and PACSIN2 at 1 min post-switch to 37°C ( S16B and S16C Fig ) . However , the degree of colocalization increased with time . Transferrin has been previously shown to be transported to PACSIN2-positive perinuclear vesicles upon entry [56] . Consistent with that , we found a significant portion of transferrin in PACSIN2-positive endosomes at 3 min post-entry ( S16A Fig ) . This is in contrast to TcdA , which colocalizes with an endosome-independent pool of PACSIN2 in Caco-2 cells ( S13 Fig ) . It is important to note that while transferrin colocalizes with PACSIN2-positive endosomes , the uptake of this cargo in Caco-2 cells does not require PACSIN2 ( S17 Fig ) . In sum , our findings emphasize a specific requirement for PACSIN2 in the TcdA uptake mechanism .
TcdA and TcdB are the key virulence factors that mediate the pathology associated with C . difficile infection [5 , 6] . Cellular intoxication by TcdA and TcdB depends on endocytosis and transport to acidified endosomal compartments within cells [18 , 22 , 23 , 58] . Since these toxins represent excellent targets for therapeutic intervention , understanding the mechanism of toxin entry is a significant priority . TcdB has been shown to require clathrin-mediated endocytosis ( CME ) to induce Rac1-inactivation and cell rounding [36] . However , the endocytic mechanisms utilized by TcdA for entry and intoxication have not been clearly defined . In this study , we have combined several independent approaches to obtain a detailed understanding of TcdA entry into cells . We perturbed the function of several key endocytic factors using RNAi-mediated depletion or pharmacological inhibition and determined the subsequent effect on TcdA uptake and toxin-induced downstream effects such as Rac1 glucosylation , cell rounding or cell death . The use of TcdB and transferrin as a controls in these assays allowed for comparative analyses of the endocytic factors ( or pathways ) relevant for the cytotoxic mechanism of both toxins and for evaluation of the specificity of the perturbations made . Additionally , we validated our findings from the perturbation studies by examining the colocalization of fluorescently labeled TcdA with markers of specific endocytic pathways . Using these complementary approaches , we find that TcdA uptake in Caco-2 cells is independent of CME . First , both transient and stable depletion of CHC had no effect on TcdA-induced cytotoxicity . Second , TcdA did not colocalize with markers of the clathrin-mediated endocytic pathway . Our results contradict previous reports that propose a role for CME in TcdA uptake [36 , 37] . While the first study reported that TcdA uptake occurs via CME [36] , the second study has implicated both clathrin-dependent and–independent pathways in TcdA uptake [37] . Both studies relied primarily on pharmacological inhibition of CME by chlorpromazine to investigate the contribution of CME to TcdA entry . Pretreatment of HeLa or HT-29 cells with chlorpromazine was shown to reduce Rac1 glucosylation and cell rounding by TcdA [36 , 37] . Consistent with these studies , we observed that chlorpromazine treatment also reduced Rac1 glucosylation by TcdA in Caco-2 cells . While chlorpromazine has been widely used to disrupt clathrin-coated pits , several studies demonstrate that the drug can also interfere with clathrin-independent endocytic mechanisms [41 , 59–61] . It is therefore important to corroborate the data obtained by pharmacological inhibition with other more specific approaches . Interestingly , in contrast to their chlorpromazine experiments , Gerhard et al . did not observe a significant decrease in Rac1 glucosylation by TcdA in HT-29 cells depleted of CHC [37] . This would argue that TcdA intoxication in HT-29 cells is independent of CME , consistent with our findings in Caco-2 cells . Clathrin-independent endocytic ( CIE ) pathways can be subdivided based on whether or not they use a dynamin GTPase for vesicle scission [62 , 63] . Macropinocytosis , clathrin-independent carriers ( CLIC ) and Arf6-regulated pathways are dynamin-independent , whereas dynamin has been implicated or shown to be involved in caveolae- , RhoA- , flotillin- and endophilinA2-mediated endocytosis [38 , 39 , 41 , 62 , 63] . Results from our perturbation studies indicate that TcdA internalization is dynamin-dependent . Using a siRNA-based screen of endocytic factors from dynamin-dependent pathways , we were able to identify Cav1 , cavin1 and PACSIN2 to be important for TcdA-mediated cell death . Cav1 , cavin1 and PACSIN2 are key proteins involved in caveolae formation and endocytosis . Cav1 is a major structural component of caveolae membrane coats [51]; disruption of Cav1 leads to loss of caveolae [48] , and ectopic expression of Cav1 in cells lacking caveolae results in de novo formation of caveolae [64] . Cavin1 or PTRF ( polymerase I transcript release factor ) is a caveolae-associated protein that is required for the formation of caveolae via sequestration of caveolins into caveolae [50] . PACSIN2/syndapin-II is a Fer-CIP4 homology-BAR ( F-BAR ) domain-containing protein that is involved in the membrane sculpting of caveolae and recruitment of dynamin for caveolae fission [49 , 52] . We found no evidence for a direct involvement of caveolae-mediated endocytosis in TcdA uptake . Imaging studies in Caco-2 cells showed no detectable colocalization between TcdA and Cav1 or cavin1 . However , we made two observations worth noting . First , Caco-2 cells do not express the α isoform of Cav1 . Fujimoto et al . reported that expression of the Cav1 α isoform , but not β isoform , resulted in the formation of caveolar invaginations in cells that lack endogenous caveolae , suggesting that the α isoform is required for functional caveolae formation [47] . Second , we observed a strong nuclear but diffuse cytoplasmic staining for cavin1 in Caco-2 cells . These observations are in line with a previous report by Vogel et al . [46] , which showed that Caco-2 cells lack functional caveolae . Lack of Cav1α and functional caveolae likely affects the localization and function of cavin1 in these cells resulting in the atypical staining pattern . We also did not observe colocalization between TcdA and Cav1 or cavin1 in mouse embryonic fibroblast cells that do contain functional caveolae . Furthermore , depletion of Cav1 does not affect TcdA uptake or TcdA-induced cytopathic effects in MEFs indicating that TcdA uptake can occur independent of caveolae-mediated endocytosis . We speculate that Cav1 and cavin1 promote TcdA-induced toxicity in Caco-2 cells through indirect mechanisms . One possibility is that these proteins regulate the expression or function of endocytic factors involved in the TcdA uptake mechanism . There is emerging evidence for such crosstalk between caveolar proteins and other CIE pathways [65 , 66] . Additionally , Cav1 and cavin1 are involved in cholesterol trafficking and homeostasis [67–69] . TcdA requires cholesterol for pore-formation and toxicity [70] . Depletion of caveolar proteins may modulate the lipid composition of cell membranes leading to indirect effects on TcdA toxicity . However , Cav1 depletion does not affect TcdA-induced cytopathic effects in MEF cells , making this unlikely to be the mechanism involved . Finally , caveolar proteins could be involved in signaling mechanisms or recycling of receptors that promote TcdA-induced toxicity . We currently do not know the receptor ( s ) or the exact uptake mechanism for TcdA in Caco-2 and MEF cells . Therefore , it is difficult to determine which of these indirect mechanisms , if any , contribute to the effects observed in our siRNA-viability assay . Interestingly , we find that TcdA uptake , while caveolae-independent , is dependent on PACSIN2 . In MEF cells , which contain caveolae , TcdA colocalizes with PACSIN2 , and depletion of PACSIN2 inhibits TcdA entry and toxin-induced downstream effects . We made similar observations in Caco-2 cells , which lack caveolae . Proteins in the PACSIN/syndapin family have an N-terminal F-BAR domain that mediates F-actin binding and membrane bending and a C-terminal Src homology 3 ( SH3 ) domain that can interact with dynamin , synaptojanin and Neuronal Wiskott-Aldrich Syndrome Protein ( N-WASP ) , a component of the actin polymerization machinery [53 , 71–75] . PACSIN2 is ubiquitously expressed , whereas PACSIN1 is mainly expressed in brain , and PACSIN3 is expressed predominantly in skeletal muscles , lung and heart [73 , 76 , 77] . PACSINs form homo- and hetero- oligomers that allow them to function as adapter or scaffolding proteins that can link the actin cytoskeleton with the endocytic machinery [54 , 55 , 78] . Previously , PACSIN2 has been shown to be involved in epidermal growth factor receptor ( EGFR ) internalization and cholera toxin B ( CTxB ) entry [52 , 56 , 79] . Similar to our observations with TcdA , CTxB has been shown to colocalize with PACSIN2 in HeLa cells [56] , and depletion of PACSIN2 results in a significant decrease in CTxB incorporation into HeLa cells [52] . It is not clear from these studies however , whether PACSIN2 functions independently of caveolae-mediated endocytosis in promoting CTxB entry . While PACSIN2 is required for TcdA uptake in both Caco-2 and wildtype MEF cells , the molecular and mechanistic details of this PACSIN2-dependent uptake may differ between these cells . In wildtype MEF cells , our images show TcdA in PACSIN2-positive structures that have a vesicular appearance , whereas in Caco-2 cells , we observe TcdA in PACSIN2 structures with different curvature . PACSIN2 is a BAR-domain protein that bends membranes . In addition to being associated with vesicular structures such as caveolae , PACSIN2 is also known to create tubular membrane invaginations [53] . We speculate that in Caco-2 cells , TcdA is internalized into small tubules or tubular constrictions induced by PACSIN2 , resulting in the extended structures we observe in our images . We do not know why toxin is internalized into PACSIN2 structures with different curvatures in Caco-2 vs MEF cells , and this is an area for future investigation . Overall , our data indicate that in Caco-2 and MEF cells TcdA uptake and intoxication occurs by a clathrin- and caveolae-independent endocytic mechanism that requires PACSIN2 . While our work supports an important role for PACSIN2 and dynamin in TcdA uptake and cytotoxicity , we cannot conclude that TcdA entry occurs solely by this mechanism . Alternate routes of entry may exist for TcdA , but based on our perturbation studies , we anticipate that their contribution will be minor . Our work also shows that TcdA and TcdB utilize distinct endocytic pathways to intoxicate epithelial cells . TcdA and TcdB bind different cell surface proteins and sugars [19 , 31–33 , 35] , which likely explains their internalization by distinct endocytic pathways . Importantly , the differences in entry between TcdA and TcdB can have implications regarding their cytotoxic mechanisms . TcdB is more than potent than TcdA in cell culture and animal models [5 , 80 , 81] . TcdB causes necrosis and extensive damage to the colonic epithelium by inducing the production of reactive oxygen species ( ROS ) . ROS generation by TcdB requires internalization of TcdB-receptor complexes and the activated NADPH oxidase complex via CME and the subsequent formation of a redox active endosome [30] . TcdA , however , is unable to induce ROS [30] . We speculate that the clathrin-independent , PACSIN2-dependent entry mechanism utilized by TcdA prevents the assembly of the redox active endosomes , resulting in reduced toxicity compared to TcdB . Many aspects of this PACSIN2- and dynamin-dependent endocytic mechanism remain to be elucidated , including how PACSIN2 mediates vesicle formation . Recently , Boucrot et al . and Renard et al . described a new endocytic route ( FEME pathway ) mediated by endophilinA2 , which is a BAR domain-containing protein similar to PACSIN2 [41 , 42] . The FEME pathway is a clathrin-independent and dynamin-dependent pathway that mediates internalization of various clathrin-independent cargoes including Shiga and cholera toxins [41 , 42] . Binding to cargo receptor and recruitment of dynamin is mediated by the SH3 domain of endophilin , membrane curvature is induced by the BAR domain , and membrane scission is achieved by the cooperative actions of endophilin , actin and dynamin [41 , 42 , 82] . We speculate that PACSIN2 can mediate vesicle formation and release in a manner similar to that of endophilin . We also do not know how TcdA is able to gain entry by this pathway and what other host proteins in addition to PACSIN2 and dynamin are required for this process . For future studies , it will be important to identify the TcdA receptor and characterize the toxin-receptor interactions that are necessary for entry by this pathway . We hope to use TcdA as a tool to screen for host proteins that play a role in this pathway . In conclusion , our study identifies an important route of entry for TcdA in cells that could be targeted for therapeutic purposes , and expands our understanding of PACSIN2’s role in endocytosis . In the future , it will be important to investigate how this PACSIN2 pathway is regulated and if this is a generalized mechanism that TcdA can utilize in cell types other than Caco-2 and MEF cells .
Caco-2 cells ( ATCC HTB-37 ) were maintained in Minimum Essential Medium ( MEM ) supplemented with 10% fetal bovine serum ( FBS; Atlanta Biologicals ) , 1% MEM non-essential amino acids ( M7145; Sigma ) , 1% Hepes buffer ( 15630080; Gibco ) and 1% sodium pyruvate ( S8636; Sigma ) . HeLa cells ( ATCC CCL-2 ) , HEK 293T cells ( ATCC CRL-11268 ) , wildtype ( ATCC CRL-2752 ) and caveolin1-/- mouse embryonic fibroblast ( MEF ) cells ( ATCC CRL-2753 ) were grown in Dulbecco’s Modified Eagle’s Medium ( DMEM ) supplemented with 10% FBS . Dynasore ( D7693; Sigma ) was dissolved in DMSO to obtain a 25 mM stock and was used at a final concentration of 80 μM . Dynasore experiments were performed under serum-free media conditions as the inhibitor binds to serum proteins and loses activity [83] . Plasmids encoding wildtype TcdA and TcdB were transformed into Bacillus megaterium according to the manufacturer’s instructions ( MoBiTec ) . Recombinant toxins were expressed and purified as described previously with some modifications [84] . B . megaterium expression strains were grown in LB containing 10 mg/L tetracycline and 35 mL overnight culture was used to inoculate 1 L of media . Bacteria were grown at 37°C with shaking at 220 rpm . Toxin expression was induced with 5 g D-xylose once the culture reached OD600 = 0 . 5 . Cells were harvested after 4 h and resuspended in 200 mL of binding buffer ( 20 mM Tris [pH 8 . 0] , 100 mM NaCl for TcdA and 20 mM Tris [pH 8 . 0] , 500 mM NaCl for TcdB ) supplemented with DNase , 400 μL of lysozyme ( 10 mg/mL ) and protease inhibitors ( P8849; Sigma ) . Cells were lysed using an Emulsiflex homogenizer , and lysates were centrifuged at 48 , 000 g for 30 min . The proteins were purified from the supernatant by Ni-affinity , anion exchange and size exclusion chromatography . Toxins were eluted and stored in 20 mM HEPES ( pH 7 . 0 ) , 50 mM NaCl . Caco-2 cells were seeded at a density of 1 , 000 cells per well in a 384-well plate and incubated at 37°C for 48 h . Cells were then challenged with serial dilutions of TcdA ( unlabeled or alexa546 labeled ) or TcdB in triplicate . ATP levels of TcdB- and TcdA-treated cells were quantified 24 h and 48 h post intoxication , respectively , by addition of CellTiter-Glo ( G7571; Promega ) and used as a measure of cellular viability . Luminescence was read using a BioTek Synergy 4 plate reader . Relative cell survival was determined by normalizing the ATP levels of toxin-treated cells to untreated controls . For viability assays , Caco-2 cells ( 1000 cells/well ) were reverse-transfected with 10 nM siRNA against luciferase ( non-targeting; negative control ) or various targets ( Thermo Fisher Scientific ) using RNAiMax transfection reagent ( 13778075; Thermo Fisher Scientific ) as described previously [30] . Transfections were performed in a 384-well plate format , with 8 wells per target and toxin treatment . Three wells received mock treatment and three wells received 50 nM of TcdA for 48 h or 50 nM TcdB for 24 h and viability was assayed using CellTiter-Glo . Cells from the remaining two wells were collected and used for RNA isolation and RT-PCR analyses . Relative cell survival was determined by normalizing the ATP levels of toxin-treated cells to untreated controls ( which is at a value of 1 . 0 ) . In some instances , cytotoxicity data are represented as fold change of survival , which was obtained by normalizing the relative viability for each target to that of luciferase control . For immunofluorescence assays , wildtype MEF cells ( 14 , 000 cells/well ) were reverse-transfected with 20 nM siRNA against luciferase , Cav1 or PACSIN2 ( Thermo Fisher Scientific ) using lipofectamine RNAiMax ( 1 . 5 μL per well ) as described by the manufacturer . Cells were seeded on 12 mm glass coverslips ( # 1 . 5; Fisherbrand ) in 24-well plates and incubated at 37°C for 48 h to achieve sufficient knockdown . Transfected cells were subsequently used for toxin binding and uptake analyses . Non-targeting control shRNA ( RHS4346 ) and shRNAs targeting sequences in clathrin heavy chain ( V2LHS_67887 and V3LHS_359489 ) were purchased from GE Healthcare Dharmacon , Inc . PACSIN2 shRNA ( TRCN0000037982; Sigma ) was a gift from Matt Tyska ( Vanderbilt University , Nashville , TN ) . The packaging plasmids ΔR8 . 91 and pCMVG were a kind donation from Chris Aiken ( Vanderbilt University Medical Center , Nashville , TN ) . For stable knockdowns , shRNA plasmids were packaged into lentiviral particles for transduction . Briefly , HEK293T cells were plated in a 10 cm dish and transfected with 1 mL of serum free media containing 30 μg total DNA ( 15 μg shRNA plasmid + 11 . 25 μg ΔR8 . 91 + 3 . 75 μg pCMVG ) preincubated with 90 μL of 1 mg/mL of PEI . After 48 h of transfection , a total of 10 mL media containing virus particles was collected and passed through a 0 . 45 μm filter and stored in 1 mL aliquots at– 80°C . Caco-2 cells were plated in 75- cm2 flasks such that they were 60% confluent on the day of transduction . Cells were incubated with 1 . 5 mL of virus supernatant diluted in 3 mL of conditioned media containing 4 μg/mL of polybrene ( 107689; Sigma ) at 37°C for 4 h , then supplemented with 5 mL of conditioned media and incubated overnight . Infected cells were passaged and allowed to recover for 2 days . Transduced Caco-2 cells were then selected by culturing in media containing 10 μg/mL puromycin ( P8833; Sigma ) for 96 h . To confirm knockdown , whole cell lysates were probed with antibodies against the target protein and GAPDH ( loading control ) . Total RNA was extracted using the RNeasy Mini Kit ( 74104; Qiagen ) . Target mRNAs were amplified from 10 ng of template RNA using a OneStep RT-PCR kit ( 210212; Qiagen ) . Primers used are listed in S1 Table . GAPDH mRNA was amplified as a loading control . The RT-PCR products were resolved on a 1–1 . 5% agarose gel and imaged using the KODAK EDAS 290 digital camera system . Caco-2 cells expressing ctrl shRNA and PACSIN2 sh982 were seeded at a density of 400 , 000 cells/well in a 6-well plate format and incubated at 37°C for 48 h . For the binding assay , cells were switched to 10°C for 1 h and then intoxicated with 30 nM TcdA . Toxins were allowed to bind at 10°C for 1 h . Media containing unbound toxin inoculum were then removed and cells were washed twice with ice cold PBS . Cells were dislodged by using a cell scraper , collected and pelleted at 1000 g for 5 min . Cell pellets were homogenized to obtain lysates for SDS PAGE and Western blot . For assays performed at 37°C , cells were initially incubated with toxin suspension for 30 min at 10°C and then switched to 37°C for 4 min . The brief incubation warms the cells to 37°C but does not allow for appreciable toxin internalization . Cells were then washed twice with PBS prewarmed to 37°C to remove unbound toxins and collected for lysis and western blotting . The blot was probed with antibodies against TcdA , PACSIN2 , unglucosylated Rac1 , total Rac1 and GAPDH . Additionally , cells that did not receive any toxin were used as control . To prepare samples for western blotting , cell pellets were suspended in 60 μL of lysis buffer ( 10mM Tris-Cl , pH 7 . 4 , 250 mM sucrose , 3 mM Imidazole ) supplemented with protease inhibitor cocktail ( 1:100 , P8340; Sigma ) and homogenized by passing 20-times through a 27G needle fitted to a sterile 1 mL syringe . Nuclei and debri were pelleted and removed by spinning at 1500 g for 15 min . Samples were then diluted with Laemmli sample buffer containing 2-mercaptoethanol and heated at 95°C for 5 min . Equal volumes were loaded on a 4–20% Mini-Protean gradient gels ( Bio-Rad ) . Proteins were transferred in Tris-Glycine buffer to PVDF membranes at 100 V for 1 h and blocked with 5% milk in PBS containing 0 . 1% Tween-20 ( PBST ) overnight . Primary antibodies against clathrin heavy chain ( 1:2000 , ab21679; Abcam ) , GAPDH ( 1:3000 , sc-25778; Santa Cruz ) , unglucosylated Rac1 ( 1:1000 , 610650; BD Biosciences ) , total Rac1 ( 1:1000 , 05–389; Millipore ) , caveolin1α ( 1:2000 , sc-894; Santa Cruz ) , caveolin1α/β ( 1:1000 , 610060; BD Biosciences ) , PACSIN2 ( 1:2000 , AP8088b; Abgent ) , TcdA ( 1:1000 , NB600-1066; Novus Biologicals ) and tubulin ( 1:5000 , 3873S; Cell Signaling ) were diluted in 5% milk-PBST and incubated with the membranes for 2 h at room temperature . Membranes were washed four times with PBST and then incubated with anti-mouse ( 7076S; Cell Signaling ) or anti-rabbit ( 7074S; Cell signaling ) HRP-linked secondary antibodies for 1 h at room temperature ( 1:2000 for TcdA , PACSIN2 , Rac1 , caveolin1 and CHC; 1:5000 for GAPDH and tubulin ) . Membranes were washed four times with PBST and HRP was detected using ECL Western Blotting Substrate ( 32106; Pierce ) . Images of film scans were converted to grayscale and cropped using Photoshop ( Adobe Systems ) . Purified toxins in 20 mM HEPES ( pH 7 . 0 ) , 50 mM NaCl were incubated with a five-fold molar excess ( over the cysteines ) of thiol-reactive Alexa Fluor dyes ( A10258 and A20347; Thermo Fisher Scientific ) at room temperature for 2 h in the dark . Excess dye was removed by dialysis overnight at 4°C using slide-A-lyzer dialysis cassettes ( 66380; Thermo Fisher Scientific ) . Toxin concentration and degree of labeling were determined according to manufacturer’s instructions and stored at– 80°C for future use . 1x104 cells were seeded on 12 mm glass coverslips ( # 1 . 5; Fisherbrand ) in 24-well plates and incubated at 37°C for 48 h . For internalization assays , cells were chilled at 10°C for 45 min and then incubated with media containing 50 nM labeled toxin or buffer ( no toxin control ) at 10°C for 45 min . Unbound toxins were removed , and cells were switched to 37°C for various time intervals to allow internalization of toxin . Caco-2 cells lose their actin cytoskeleton and begin to round after 30 min of toxin treatment . Therefore , for imaging experiments 30 min was chosen as the last time interval for staining and image analyses . At each time interval , cells were washed once with pre-warmed PBS and fixed with 4% paraformaldehyde in PBS at 37°C for 15 min . Following fixation , cells were quenched with 0 . 1 M glycine in PBS , washed three times with PBS and permeabilized with 0 . 2% Triton X-100/PBS for 3 min at room temperature ( RT ) . Cells were washed three times with PBS and blocked overnight at 4°C in PBS containing 4% BSA , 5% normal goat serum ( Life technologies ) , 0 . 1% Tween-20 . The following day , cells were washed once in BSA-PBST ( 1% BSA , 0 . 1% Tween-20 in PBS ) . Primary antibodies anti-CHC ( 1:1000 , ab21679; Abcam ) , rabbit anti-caveolin1 ( 1:50 for Caco-2 cells and 1:250 for wildtype MEFs , 610060; BD Biosciences ) , mouse anti-caveolin1 ( 1:25 for wildtype MEFs ( Fig 4 ) , 610493; BD Biosciences ) , anti-cavin1 ( 1:100 , ab48824; Abcam ) , anti-EEA1 ( 1:250 , 610457; BD Biosciences ) and anti-PACSIN2 ( 1:100 , AP8088b; Abgent ) were diluted in BSA-PBST and incubated with cells for 2 h at RT . Primary antibody was removed and cells were washed four times with BSA-PBST , and incubated with goat anti-rabbit Alexa Fluor 647 ( A-21245; Life Technologies ) , goat anti-mouse Alexa Flour 488 ( A-11029; life Technologies ) or goat anti-rabbit Alexa Fluor 546 ( A-11035; Life Technologies ) in BSA-PBST for 1 h at RT ( 1:1000 for CHC; 1:500 for caveolin1 , cavin1 , EEA1 and PACSIN2 ) . Cells were washed twice in BSA-PBST and twice in PBS and then mounted using Prolong Gold Antifade Mountant ( P36931; Life Technologies ) . For actin staining , cells were incubated with 1:100 dilution of Phalloidin-647 ( A2287; Thermo Fisher Scientific ) for 30 min at RT . The phalloidin incubation step was performed after secondary antibody staining and washes . After 30 min incubation , cells were washed twice in BSA-PBST and twice in PBS and then mounted using Prolong Gold Antifade Mountant . Slides were imaged using a 63x/ 1 . 40 numerical aperture ( NA ) Plan-Apochromat oil immersion objective on a LSM 710 Meta Inverted laser-scanning confocal microscope ( Zeiss ) located in the Vanderbilt Cell Imaging Shared Resource ( CISR ) Core . Alexa488 was excited using the 488 nm line of an Argon laser . Alexa546 was excited at 561 nm and Alexa647 was excited at 633 nm using a HeNe laser . Fluorescence emission was detected using filters provided by the manufacturer . Pinhole size was identical for the fluors used . Single sections of 0 . 49 μm thickness from a Z-stack are presented . For the purpose of presentation , raw images were exported in tiff format and brightness and contrast were adjusted to the same extent using Fiji [85] . For measurements of colocalization , individual cells were demarcated and Pearson’s correlation coefficient and Mander’s overlap coefficients were determined using the Colocalization plugin in Fiji . The Mander’s coefficient determines the fraction of channel1 that overlaps with channel2 and vice versa , with 100% overlap resulting in a value of 1 . Nuclei were excluded from the colocalization analyses . Qualitative analyses of colocalization were performed using the plot profile feature in Fiji . The pixel intensities along the line were obtained from the original 12-bit image for each channel and plotted as relative intensities over distance using GraphPad Prism . PACSIN2 , Cav1 , transferrin and TcdA signals in siRNA- or shRNA-expressing cells were determined by demarcating individual cells and measuring the mean fluorescence intensities for each channel using Fiji . The mean fluorescence intensities of PACSIN2 and TcdA ( or transferrin ) for each cell were converted to relative intensity values and plotted against each other to generate the scatter plot . To determine the correlation between PACSIN2 and TcdA ( or transferrin ) in cells , a linear regression analysis was performed on the entire data set and R2 value for best fit is represented . Cells were chosen at random for colocalization and intensity analyses . Wildtype and caveolin1-/- MEF cells were seeded , in triplicate , at 200 , 000 cells/well in 6-well plates . The following day , cells were chilled at 10°C for 45 min and allowed to bind 10 nM TcdA for 45 min . Plates were then moved to a Cytation 5 Cell Imaging Multi-Mode Reader ( BioTek ) , and bright field images of cell morphology were captured at 15 min intervals for a 5 h time period , using a 20x/ 0 . 45 NA objective ( 1220517; BioTek ) . A total of 4 frames per well were captured at each time point . The chamber was maintained at 37°C with 5% CO2 for the duration of this experiment . Real-time videos of cells rounding in response to toxin were generated using Fiji . For quantification , a total of 36 frames from three independent experiments were analyzed for each cell type and results were expressed as the percentage of rounded cells . Cell rounding assays with siRNA transfected cells were performed as described above with a few modifications . Wildtype MEF cells ( 200 , 000 cells/well ) were reverse-transfected with 20 nM siRNA against luciferase ( non-targeting; negative control ) , Cav1 or PACSIN2 ( Thermo Fisher Scientific ) using RNAiMax transfection reagent ( 7 μL per well ) as described by the manufacturer . After 48 h , cells were intoxicated with 5 nM TcdA and images were captured every 12 min for a total of 2 h . For quantification , a total of 12 frames ( > 100 cells/frame ) from three independent experiments were analyzed for each time point and siRNA condition , and results were expressed as the percentage of rounded cells . Statistical analyses are described in figure legends . A p-value of ≤ 0 . 05 was considered significant . | Clostridium difficile is a bacterial pathogen that causes nearly half a million infections each year in the United States . It infects the human colon and causes diarrhea , colitis and , in some cases , death . C . difficile infection is mediated by the action of two large homologous toxins , TcdA and TcdB . Disruption of host cell function by these toxins requires entry into cells . There are multiple ways for pathogens and virulence factors such as viruses and toxins to enter host cells . The entry mechanism is often directed by a cell surface receptor and can impact the trafficking and virulence properties of the pathogenic factor . Investigating the internalization strategy can provide critical insight into the mechanism of action for specific pathogens and virulence factors . In our current study , we sought to determine the strategy utilized by TcdA to enter host cells . We show that TcdA uptake occurs by a clathrin- and caveolae-independent endocytic mechanism that is mediated by PACSIN2 and dynamin . We also show that TcdA and TcdB can utilize different routes of entry , which may have implications regarding their cytotoxic mechanisms . In summary , our results provide new insights into the mechanism of cellular intoxication by TcdA and the role of PACSIN2 in endocytosis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"luciferase",
"toxins",
"pathology",
"and",
"laboratory",
"medicine",
"gene",
"regulation",
"enzymes",
"biological",
"cultures",
"cell",
"processes",
"enzymology",
"light",
"microscopy",
"toxicology",
"toxic",
"agents",
"microscopy",
"caco-2",
"cells",
"confocal",
"microscopy",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"small",
"interfering",
"rnas",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"proteins",
"coated",
"pits",
"gene",
"expression",
"oxidoreductases",
"cell",
"lines",
"cell",
"membranes",
"endocytosis",
"biochemistry",
"rna",
"cell",
"staining",
"cell",
"biology",
"nucleic",
"acids",
"secretory",
"pathway",
"genetics",
"biology",
"and",
"life",
"sciences",
"non-coding",
"rna"
]
| 2016 | Clostridium difficile Toxin A Undergoes Clathrin-Independent, PACSIN2-Dependent Endocytosis |
Ion channels catalyze ionic permeation across membranes via water-filled pores . To understand how changes in intracellular magnesium concentration regulate the influx of Mg2+ into cells , we examine early events in the relaxation of Mg2+ channel CorA toward its open state using massively-repeated molecular dynamics simulations conducted either with or without regulatory ions . The pore of CorA contains a 2-nm-long hydrophobic bottleneck which remained dehydrated in most simulations . However , rapid hydration or “wetting” events concurrent with small-amplitude fluctuations in pore diameter occurred spontaneously and reversibly . In the absence of regulatory ions , wetting transitions are more likely and include a wet state that is significantly more stable and more hydrated . The free energy profile for Mg2+ permeation presents a barrier whose magnitude is anticorrelated to pore diameter and the extent of hydrophobic hydration . These findings support an allosteric mechanism whereby wetting of a hydrophobic gate couples changes in intracellular magnesium concentration to the onset of ionic conduction .
Magnesium homeostasis is essential for life . In humans , the misregulation of magnesium is implicated in stroke [1] , heart disease [2] , and diabetes [3] . Magnesium transport is also crucial for bacteria [4] . The movement of magnesium through cell membranes , like that of other ions , is accomplished by integral membrane proteins that provide selective permeability across the dielectric barrier of the lipid bilayer [5] . In bacteria , magnesium uptake is mediated by the CorA protein [6–9] , which can substitute for its functional homologue in yeast mitochondria [10] . Electrophysiological data suggests that TmCorA is a channel , not a transporter [8] . Seven crystallographic structures exist for CorA , six of which are from Thermotoga maritima ( TmCorA ) [11–16] . These structures reveal a homopentamer in which 10-nm-long protomeric α-helices ( the “stalk” helices ) form a transmembrane ( TM ) pore through which magnesium is presumed to flow . This pore contains two hydrophobic constrictions: the “MM stretch” ( MM ) , a 1 . 9-nm-long constriction formed by pore-lining residues M291 , L294 , A298 , and M302; and the “lower leucine constriction” ( LC ) , a shorter steric bottleneck formed by the sidechain of L280 ( Fig 1 ) . Mutagenesis studies suggest that the MM , but not the LC , is involved in channel gating [17 , 18] . Hydrophobic gates are important for the function of many ion channels , including ligand-gated [19–22] , voltage-gated [23 , 24] , phosphorylation-gated [25] , and mechanosensitive channels [26 , 27] . In all crystal structures of CorA , both hydrophobic constrictions are too narrow to be hydrated , suggesting that the channel is in its closed state . Remarkably , the pore extends beyond the relatively small TM domain into a much larger funnel-shaped domain that protrudes by 6 nm into the cytosol ( Fig 1 ) . At the far rim of this funnel , divalent cations are bound between cytosolic protomer interfaces ( the divalent cation sensor or DCS; Fig 1 ) . Based on crystallographic structures , it was hypothesized that divalent cation occupancy of the DCS regulates magnesium transport by controlling the pore's diameter or electrostatic profile [11–13] . Recent studies suggest that divalent cation binding to the DCS locks TmCorA in a transport incompetent conformation and that loss of these cations leads to an open conformation of the channel [14 , 28] , which may be asymmetric [14] . To investigate the allosteric regulation of pore opening , we previously conducted a molecular dynamics ( MD ) study of TmCorA in a hydrated lipid bilayer , either with or without Mg2+ ions in the DCS [29] . The MM remained dehydrated throughout a 110-ns MD simulation in the presence of regulatory ions , but became dilated and hydrated in one of two trajectories generated after these ions were removed [29] . Wetting of the MM involved an iris-like mechanism initiated by the rearrangement of the cytosolic domain interfaces and transmitted to the MM by the long pore-lining stalk helices . These findings suggest a model of allosterically-regulated hydrophobic gating whereby decreasing cytosolic magnesium concentration reduces magnesium occupancy in the DCS , leading to sudden wetting of the pore . Here , we use massive sampling to examine the statistical significance and the kinetics of this apparent hydrophobic gating process , and to assess whether it results in the open state of the channel . Scaling up computational sampling by two orders of magnitude , we compare hundreds of 35-ns MD simulations in which the DCS are either fully occupied or empty . From a total of 54 microseconds of sampling , we observe many events in which completely connected columns of water condense and evaporate in the MM . We show that this hydrophobic gate is more likely to become hydrated in the absence of regulatory ions and we quantify the kinetics of wetting and dewetting transitions . Finally , we show that the extent of hydration of the MM is correlated to a reduction in the free energy barrier for the permeation of Mg2+ ions , supporting the hypothesis that the MM is an allosterically-regulated hydrophobic gate .
To determine the kinetics of allosterically-regulated wetting in the MM ( the presumed hydrophobic gate ) , we massively repeated MD simulations of TmCorA in the two limiting states of ionic regulation , namely , with all ten regulatory ( DCS ) binding sites either occupied or empty . The PDB:2HN2 crystallographic structure of TmCorA [13] was embedded in a hydrated lipid bilayer and seven-hundred 35-ns MD simulations were conducted for each regulatory state . Initially , water filled most of the pore but the MM and the LC were completely dehydrated . Although the MM remained dehydrated in most of the simulations , it was hydrated at least part of the time in 12 and 25% ( respectively 83 and 175 ) of the 700 simulations conducted respectively with and without regulatory ions ( Table 1 ) , allowing us to quantify the effect of regulatory ions on pore hydration . MM wetting is depicted as S1 Movie . In wetting transitions , water molecules entered the MM along the pore axis and , occasionally , through transient packing defects between pore-lining helices . The latter defects connected the middle of the pore directly to bulk water at the cytosolic membrane-water interface ( S1 Fig ) . To characterize wetting , we computed the maximum distance between consecutive water oxygen atoms along the pore axis , zgap . Wetting in the MM or LC is defined as end-to-end hydration by a completely connected water column ( zgap ≤ 0 . 38 nm ) . The fraction of simulations in which the MM was wetted , P ( zgap ≤ 0 . 38 nm ) , is shown as a function of simulation time in Fig 2A . In the presence of regulatory ions , this fraction appears to stabilize around 0 . 3% after 15 ns ( Fig 2A ) . In contrast , the MM was wetted five times more often in the absence of regulatory ions , and the wetting probability increased from 0 to 2% through the entire length of the simulations ( Fig 2A ) . Conversely , hydration of the LC increased from 0 to 50% independently of ionic occupancy of the DCS , although equilibration occurred more rapidly in the absence of regulatory ions ( Fig 2B ) . To quantify the extent of hydration of the hydrophobic constrictions , we used two metrics , the length of the longest dehydrated stretch , zgap , and the number of water molecules , Nwat . As expected , these two metrics are strongly anti-correlated ( S2A and S2E Fig ) . The normalized distributions of zgap ( S2B and S2F Fig ) and Nwat ( S2C and S2G Fig ) in the MM were similar for the two sets of simulations . Hydration defects were most common near M291 , at the cytosolic end of the MM ( S2D and S2H Fig ) . These findings are consistent with our previous simulations [29] and with the hypothesis of Lunin et al . that M291 and L294 form the main hydrophobic gate [11] . Finally , S3 Fig shows that hydration of the MM and LC pore constrictions are not correlated with each other . In both regulatory states , many of the wetting events were transient , half of them lasting less than 200 ps . For each simulation , Fig 3 shows the fraction of time that the MM was wetted as a function of the fraction of time that the MM contained >20 water molecules . This analysis leads us to identify eight simulations in which the MM was both stably and highly hydrated , all of which occurred in the absence of regulatory ions . These simulations are henceforth referred to as “stably superhydrated” ( SSH ) . In the most hydrated SSH simulation , the initially dehydrated MM abruptly and durably filled with ≥20 water molecules , at one point surging to twice this amount ( Fig 4A ) . This hydration was concurrent with an increase in mean MM pore diameter from 0 . 4 to 0 . 74 nm ( Fig 4A–4C ) and involved asymmetric reorientation of pore-lining sidechains away from the lumen , toward neighboring protomers ( Fig 4D and 4E ) . To assess whether hydration was coupled to reorganization of the pore , we computed the pore diameter along its long axis . On average , the volume available in the MM did not substantially depend on regulatory ion occupancy in the DCS ( Fig 5A ) . However , SSH simulations displayed an increase in MM diameter concurrent with wetting . The mean pore diameter in the MM occasionally fluctuated above 0 . 68 nm ( the diameter of a hexahydrated Mg2+ ion ) in the absence of regulatory ions , but never in their presence ( Fig 5A ) . Upon removal of these ions , there was , on average , a tightening of the pore at its extracellular end , near the proposed selectivity filter of the GMN motif [30] ( Fig 5A ) . Regardless of regulatory ion occupancy , the mean pore diameter in the MM was linearly correlated to the extent of hydration ( Fig 5B ) . It has been suggested that MM wetting could be triggered by the exposure of previously-hidden hydrophilic moieties to the pore lumen following axial rotation of the stalk helices [15 , 16] . To test this hypothesis , we analyzed protein-water hydrogen bond interactions in the MM ( S1 Text and S7 and S8 Figs ) . Taken together , our analysis indicates that pore wetting is not due to a decrease in the hydrophobic character of the MM but rather that its likelihood increases with the volume of the hydrophobic stretch . Accordingly , the increase in both wetting probability and hydration number with CorA pore size is consistent with hydrophobic wetting in simple nanoscopic systems [31] and in hydrophobic pore analogs such as carbon nanotubes [32] . The sharp dependence of hydrophobic wetting on dynamic fluctuations in pore diameter makes this process well suited to gating in a biological channel . Distributions of Nwat are shown in Fig 6A . Despite the presence of wetting transitions , these distributions are unimodal and values of Nwat greater than 10 are rare . The most likely value of Nwat in the MM is 2 for simulations both with and without regulatory ions ( Fig 6A ) . The distribution of Nwat is shifted to slightly larger values in the absence of regulatory ions , a shift that persists after the removal of the SSH simulations from this analysis ( Fig 6A ) . The distribution of Nwat computed exclusively from the SSH simulations is bimodal , with maxima at Nwat = 3 and 20 ( Fig 6A ) . Fig 6B shows the distribution of hydration number computed separately for the wetted state . In the SSH simulations , Nwat is distributed asymmetrically around 22 , while the remaining simulations , with and without regulatory ions , sampled values distributed normally around 15 ( σ = 3 ) . Thus , in the wet state , the dependence of the extent of hydration on regulatory ions was entirely due to the eight SSH simulations . To characterize the kinetics of wetting and dewetting of the hydrophobic MM , survival probabilities , S ( t ) , were computed separately for wet and dry states , either with or without regulatory ions ( Fig 7 ) . Although wet states are more stable in the absence of regulatory ions , the difference between survival probabilities obtained with and without regulatory ions is abrogated by removing the SSH simulations from this analysis ( Fig 7A ) . Thus , an essential difference between the two sets of simulations is that removal of the regulatory ions led to a hydration state not observed on the 35-ns time-scale in the presence of regulatory ions . The two wetted states differ both in the extent of hydration and in their kinetic properties . Conversely , the dewetted state of the MM was destabilized by the removal of regulatory ions , an effect that persists after removing the SSH simulations from this analysis ( Fig 7B ) . The survival probabilities of the wetted and dewetted MM states were fit to a double exponential decay function from which we computed the rate of channel wetting to be 8 . 2×106 ± 7×105 s-1 and 4 . 1×106 ± 3×105 s-1 in the absence and presence of regulatory ions , respectively ( S1 Table ) . The estimates of wetting and dewetting rates were used to compute free energy differences between hydrated and dehydrated states ( S2 Table ) , which we used to construct a model of hydrophobic gating in CorA ( see Discussion ) . To assess whether the dilation and hydration of the hydrophobic stretch led to the open state of the channel , we used umbrella sampling ( US ) [33 , 34] to compute the free energy associated with translocation of a divalent cation throughout the pore of TmCorA based on multiple starting conformations from each of four distinct states . These states correspond to hydrated and dehydrated MM with and without regulatory ions . Strictly speaking , our US simulations were conducted out of equilibrium since the channel conformation is slowly relaxing ( S9 Fig ) . As such , these profiles provide estimates of the work required for ion permeation given initial states of protein conformation , pore hydration , and regulatory ion occupancy . Multiple free energy profiles from each state are shown together as ensemble averages in Fig 8 . The general features of these profiles are similar: a small barrier at z = 2 . 6 nm , where the sidechain of N314 partially occludes the pore; a barrier of variable magnitude between −1<z<1 nm , which corresponds to the MM; a smaller barrier at z = −2 . 3 nm , which corresponds to the LC; and a minimum at z = −3 nm , which corresponds to the lower of two pore Mg2+ binding sites . The main difference between the four ensemble-averaged free energy profiles shown in Fig 8 is the height of the barrier in the MM . When the MM is fully dehydrated , it presents a very large barrier of 45±5 kcal/mol irrespective of regulatory binding site occupancy ( Fig 8 ) . The size of this barrier reduced to 25±5 kcal/mol when the MM was wetted , again irrespective of bound regulatory ions ( Fig 8 ) . Thus , the size of the free energy barrier in the MM correlates with its hydration and not with the presence of regulatory ions . This conclusion is supported and clarified by Fig 9 , in which the magnitude of the free energy barrier in the MM , ΔG‡ , is plotted against values of Nwat from initial US conformations , Nwat ( t0 ) , or Nwat ( US ) , the average value of Nwat from US simulations while the lumenal Mg2+ ion was in the MM . The magnitude of ΔG‡ is anticorrelated with both Nwat ( t0 ) and Nwat ( US ) ( Fig 9 ) . Linear fits to all data yield ΔG‡ = 46–0 . 82×Nwat ( t0 ) kcal/mol and ΔG‡ = 92–1 . 82×Nwat ( US ) kcal/mol , with r2 coefficients of 0 . 80 and 0 . 83 respectively . By extrapolating these linear relationships , we predict that reducing the magnitude of the barrier to ΔG‡ = 0 kcal/mol requires 56 or 50 water molecules based on the two above equations , respectively . A linear fit of the relation between average MM pore diameter , dpore¯ , and Nwat in the absence of regulatory ions ( Fig 5B ) , dpore¯ = 0 . 37+0 . 0067×Nwat nm , suggests that the aforementioned values of Nwat correspond to average pore diameters of 0 . 71 to 0 . 75 nm , slightly larger than the 0 . 68 nm diameter of hexahydrated Mg2+ . In turn , this result suggests that the barrier at the MM gate is essentially due to Mg2+ desolvation , including effects beyond its first hydration shell . Note that the predicted diameter at which the gating barrier vanishes is significantly smaller than the ~1 . 2 nm diameter of the putative open state proposed by Dalmas et al . [28] . Furthermore , note that the requirement ΔG‡ = 0 is excessive as the channel may conduct Mg2+ ions with moderate positive values of ΔG‡ . According to the model we present above , a pore diameter equal to that of hexahydrated Mg2+ corresponds to Nwat = 46 and ΔG‡ = 8 kcal/mol . Because pore hydration reduces the free energy barrier to Mg2+ flux , we assessed whether removing a ( pentameric ) bulky hydrophobic sidechain from the MM enhances its hydration . To this end , we conducted two-hundred 4-ns simulations for wild type and L294A TmCorA in three different wetted conformations extracted from our initial set of fourteen-hundred simulations . Consistent with the role of the MM as a hydrophobic gate , the L294A mutation enhances pore hydration both in the presence and in the absence of regulatory ions ( Fig 10 ) . In the presence of regulatory ions , L294A simulations initiated with Nwat = 25 stabilized at Nwat = 32 ( σ = 4 ) over the last 3 ns ( Fig 10C ) . Using the linear fit of Nwat ( t0 ) to ΔG‡ derived above , we predict that the pore remains closed to Mg2+ , with ΔG‡ = 20 kcal/mol . Therefore , even with the L294A mutation , further conformational rearrangement is required to attain an open state in the presence of regulatory ions . Similarly , wild type simulations initiated with Nwat = 25 or 42 remained stably hydrated with Nwat = 24 ( σ = 5 ) or spontaneously became less hydrated until Nwat = 32 ( σ = 7 ) at 4 ns , respectively ( Fig 10A and 10B ) . Conversely , L294A simulations initiated with Nwat = 25 or 42 approached Nwat values of 37 and 45 , respectively ( σ = 5; Fig 10A and 10B ) . Although these extents of hydration predict ΔG‡ values between 9 and 16 kcal/mol , the standard deviations of five water molecules among our two-hundred repeat simulations indicates that ~33 simulations initiated with Nwat = 42 continued to sample values of Nwat greater than 50 after 4 ns of simulation , and thus are predicted to present a barrier to Mg2+ flux of only 5 kcal/mol . Therefore , in the absence of regulatory ions , the conformational changes that we observe in TmCorA may be sufficient to permit pore opening in the L294A mutant . We previously proposed that the unbinding of regulatory ions induces changes in the arrangement of protomers in the cytosolic domain of TmCorA which increases the probability of hydration of the MM due to allosteric coupling via the long kinked and tilted helices that wrap around the pore in an iris-like fashion [29] . The present work confirms these conformational changes . Specifically , removing all ten regulatory Mg2+ ions results in increased separation between the centers of mass of adjacent Mg2+ binding domains ( S11A Fig ) and decreases the radial tilt angle of the intracellular region of helix 7 ( S11B Fig ) , which together define the width of the funnel-like cytosolic domain . In addition , both the pore radius and the lateral tilt of helix 7 in the MM region increase upon removing regulatory Mg2+ ions and further increase in the SSH runs ( S11C and S11D Fig ) . The first link in this allosteric mechanism is demonstrated by a clear correlation between fluctuations in Mg2+ binding domain arrangement and changes in the radial tilt of the pore helices; although this correlation exists regardless of regulatory Mg2+ occupancy , these structural fluctuations reach larger amplitudes when the regulatory sites are empty ( S12A and S12E and S12I Fig ) . At the other end of the allosteric transmission , pore diameter and hydration are strongly correlated in the SSH simulations ( S12J and S13B Figs ) , consistent with the above analysis ( Fig 5 ) . However , the allosteric mechanism by which we previously proposed radial tilt to cause pore dilation [29] was not strictly maintained throughout our new set of simulations . Specifically , the statistical linkage between average helix tilts and pore diameter is weak when all the SSH runs are considered together ( S12L and S12K Fig ) and only three out of eight SSH simulations show a significant correlation between average radial and lateral tilts of the pore helices ( as defined by a Pearson coefficient magnitude above 0 . 4; see labels d , e , and h in S13C Fig ) . Furthermore , our previous description of the iris mechanism [29] involved a positive correlation between average radial and lateral tilts , whereas this correlation is negative in two of the SSH simulations ( see labels d and e in S13C Fig ) , indicating that both increases and decreases in radial tilt of the cytosolic region of helix 7 can lead to pore dilation . A crystal structure of TmCorA recently obtained in the absence of divalent cations suggests that protomeric asymmetry involving bending of the intracellular domain with respect to the transmembrane region is a precondition for channel opening [14] . Accordingly , the analysis of our simulations shows that asymmetric stalk-helix bending is enhanced by the removal of regulatory ions ( S14 Fig ) . However , there is no correlation between stalk-helix bending and the hydration of the hydrophobic gate ( S15 and S16 Figs ) , suggesting that our 35-ns simulations sample early stages of pore wetting and dilation , and that longer timescales are required for full channel opening .
Like a previous simulation study of protein folding kinetics [35] , this study exploited the fact that rare events can be observed from massively-repeated short time trajectories provided that the simulation time exceeds the minimum passage time through the transition state . Our simulations were long enough for wetting and dewetting of the channel to occur , and the number of repeats was large enough to observe hundreds of such transitions , allowing us to derive quantitative estimates of wetting and dewetting kinetics . Significantly , the MM became hydrated more often when regulatory ions were absent from the simulations ( Fig 2 ) . Experimentally , Payandeh et al . have shown that a D253F mutation in one of the two pentameric regulatory binding sites enhances bacterial growth in limiting Mg2+ conditions [17] , presumably because it reduces the ability of these sites to bind Mg2+ , preventing closure of TmCorA . More recently , Dalmas et al . have shown that a D253K mutation abolishes current regulation by the DCS [28] . In addition , our finding that regulatory ions affect hydration of the MM , but not the LC ( Figs 2 and 7 ) , suggests that it is the former that is allosterically controlled by regulatory ions in early wetting . However , structural models of the open state generated by Dalmas et al . based on spectroscopic data predict an increased pore diameter throughout the entire channel lumen [28] , suggesting that longer-timescale pore opening events involve dilation of the LC . In this context , it remains unclear why a pore-widening mutation at the LC ( L280A ) compromises bacterial growth in limiting Mg2+ conditions [17] . The magnitude of the free energy barrier in the MM decreased as it became more hydrated ( Fig 9 ) . Consistent with this finding , mutations that shorten the sidechain of MM residue L294 enhance bacterial growth in limiting Mg2+ conditions [17 , 18] and L294A increases the rate of liposomal uptake of Mg2+ by 2 . 5-fold in a fluorescence-based flux assay [17] . Simulations of the L294A mutant confirm its increased propensity for MM wetting ( Fig 10 ) and suggest that the largest widening of the TmCorA pore observed in our simulations may be sufficient to permit Mg2+ flux in the L294A mutant . Our results suggest that these mutations enhance channel activity by increasing the water-accessible volume of the MM region , leading to increased hydration and thereby decreasing the barrier to Mg2+ flux . Hydrophobic gating has been discussed extensively in the literature [36–44] . The transient , reversible wetting of the MM revealed in our simulations ( Fig 7 ) is consistent with the concept that water molecules near hydrophobic surfaces are close to a vapor-to-liquid phase transition [45] . Hydrophobic surfaces alter the phase behavior of water [46] , especially when it is tightly confined [31] , and wetting equilibria can be dramatically altered by subtle changes in hydrophobicity [32] . Specifically , hydrophobic surfaces enhance water density fluctuations in their vicinity [45 , 47] , thereby increasing the probability of cavity formation [47] . When such a cavity spans the gap between hydrophobic surfaces , it can nucleate the formation of a vapor phase [31] via capillary drying [48] . Taken together , the above findings suggest that the hydrophobic gate of TmCorA is poised near the edge of a wetting/dewetting transition that can be pushed toward wetting by a single ( pentameric ) L→A or L→G mutation . Analogous roles have been ascribed to a single ( hexameric ) V→A mutation in the hydrophobic bottleneck of a calcium release-activated calcium channel [49] and a single I→G mutation that affects the wetting of hydrophobic surfaces during melittin dimerization [45] . In the present study , condensation of water in the MM was correlated to small-amplitude dilations of the pore-lining helices ( Figs 4 and 5 , S12B , S12F , and S12J Fig ) . Apolar surfaces only dry when they are sufficiently large [46] and capillary dehydration occurs more rapidly when the enclosing apolar surfaces are closer and longer [31] . At a modal length of 1 . 9 nm ( S2B and S2F Fig ) and lumenal diameter on the order of 0 . 4–0 . 6 nm ( Fig 5 ) , with an average helical spread of 1 . 54 nm ( S12C and S12G Fig ) , the hydrophobic MM appears to be well constructed to employ capillary dehydration as a gating mechanism . We observed the following three distinct hydration states in the MM: First , the majority of our simulations remained in a dry state . Second , a wetted state of low hydration , characterized by an average NWat¯ = 15 ( σ = 3 ) , was transiently populated and occurred more often when regulatory ions were absent ( Table 1 and Fig 7B and S2 Fig ) . Although the opening rate of TmCorA is unknown experimentally , the observed rate of wetting in our simulations , 8 . 2×106 ± 7×105 s-1 ( S1 Table ) , is 100–1 , 000 times faster than the opening rate of any known ion channel , including those involved in fast excitatory neurotransmission such as nicotinic acetylcholine receptors ( 1×104 to 8×104 s-1 ) [50 , 51] and AMPA glutamate receptors ( 7×104 to 8×104 s-1 ) [52 , 53] . Thus , together with the fact that this transient hydrated state is impermeable to Mg2+ ( Figs 8 and 9 ) , the high rate of wetting suggests that this state is an intermediate . Finally , the third state that we identified , the SSH state ( Fig 4 and S4 Fig ) , differed from transiently-hydrated states in terms of the extent and duration of hydration and in that it was not observed in the presence of regulatory ions ( Figs 3 , 6 and 7 and S6 Fig ) . The overall effect of removing regulatory ions was to increase the extent of hydration in the MM and to increase the stability of wetted states . We propose the following model of wetting and dewetting transitions in the MM , as depicted in Fig 11 . In the presence of regulatory ions , the dry state is favored over the transiently hydrated state by 4 . 2 ± 0 . 1 kcal/mol ( S2 Table ) . When regulatory ions are removed , the free energy of the transiently-hydrated state , and the barrier to hydration , are reduced by 0 . 4 ± 0 . 2 kcal/mol . Regulatory ions thus modulate the rate of wetting but not dewetting transitions respectively to and from the transiently-hydrated state . Furthermore , transitions to the SSH state only occurred in the absence of regulatory ions ( Figs 3 , 6 and 7 , and S6 Fig ) . This result , together with the linear increase in MM wetting with time ( Fig 2A ) , indicates that equilibrium has not been reached in the absence of regulatory ions and that longer simulations would lead to increasingly divergent fractions of MM wetting based on regulatory ion occupancy . The steepness of the drop in ΔG‡ vs . the extent of hydration , which was observed both with and without regulatory ions ( Figs 8 and 9 ) , indicates switch-like behavior . We hypothesize that the removal of regulatory ions stabilizes the SSH state much more dramatically than it stabilizes transiently hydrated states and that the SSH state plays an important role in TmCorA's gating mechanism . Extrapolation of the relationship between ΔG‡ and Nwat suggests that the channel may open to divalent cation flux when more than ~50 water molecules condense in the MM . Using large-scale sampling , we have characterized the mechanism and kinetics of wetting and dewetting transitions in a 1 . 9-nm-long hydrophobic constriction of the magnesium channel CorA . Massively-repeated simulations reveal essential differences in the dynamic fluctuations and relaxation of this hydrophobic gate in response to the removal of regulatory ions more than 6 nm away . The analysis of rate constants for wetting and dewetting transitions and of free energy profiles of ionic permittivity indicates that the hydrated states observed in the present study are intermediate states rather than the open state of the channel . These results lead to a model of functional gating of CorA in which regulatory ions control pore hydration , which in turn controls the onset of ionic conduction . A key event underpinning this allosteric mechanism , wetting constitutes a requisite but not sufficient step towards channel opening , as it is the extent of hydration of the pore , not the presence of water per se , that determines the barrier to ion permeation . Taken together , these findings demonstrate how capillary wetting transitions help mediate robust , non-linear switching of ionic conduction and underline the relevance of hydrophobic gating to ion channel structure and function .
The simulation systems consisted of TmCorA [13] , with and without 10 regulatory Mg2+ ions in the DCS , in a hydrated 1 , 2-dimyristoyl-sn-glycero-3-phosphatidylcholine ( DMPC ) bilayer . This system comprised 250 , 000 atoms . Simulations were performed with GROMACS [54] . The water model was TIP3P [55] . TmCorA was modeled by the OPLS-AA/L parameters [56 , 57] . DMPC was modeled by the Berger parameters [58] using the half-ε double-pairlist method [29] . The Mg2+ parameters were those of Åqvist [59] . For additional details of system setup and massively repeated simulation , and for a description of the algorithms that we used to quantify hydration , see S2 Text . To evaluate the free energy profile for the permeation of a divalent cation throughout the pore of TmCorA , we used umbrella sampling ( US ) MD simulations [33 , 34] . In these simulations , we inserted hexahydrated magnesium into the pore , instead of a naked ion , because interactions between magnesium and water molecules in its first hydration shell are very strong [60] , suggesting that magnesium is hexahydrated throughout a large part of the conduction pathway . We conducted 261 2-ns simulations in which the axial position of the magnesium ion relative to the center of mass of the hydrophobic gate , z , was harmonically restrained to a specified value , zi0 , for restraining potentials ( umbrellas ) , i , distributed every 0 . 05 nm in the range -8≤zi0≤5 nm . These 261 simulations constituted one set of US simulations and were used to generate one evaluation of the free energy profile . Altogether , we conducted 18 sets of US simulations , providing 18 independent evaluations of the free energy profile . In so doing , we computed free energy profiles for four different conformational basins of the TmCorA system , all drawn from our massively repeated sampling . The four basins were: those with the largest and smallest values of Nwat in the MM , obtained either in the presence or absence of regulatory ions . For each basin , we conducted 3 sets of US simulations , each using a different conformation extracted from a different simulation , except for the large-Nwat state in the absence of regulatory ions , for which 9 distinct conformations of TmCorA were used to compute 9 free energy profiles . For additional details , see S2 Text . | This study shows how rapid wetting/dewetting transitions in the pores of ion channels participate in the control of biological ion permeation . Ion channels catalyze ionic permeation across non-polar membranes via water-filled pores . However , non-polar stretches or hydrophobic bottlenecks are present in the pores of many ion channels . To clarify the relationship between channel regulation , pore hydration , and ion permeation , we examine how the slow relaxation of magnesium channel CorA from its closed state towards its open state modulates wetting of its hydrophobic bottleneck . Results provide a quantitative description of wetting and dewetting probabilities and kinetics and a quantitative relationship between the extent of pore hydration and the energetics of ion permeation , consistent with a mechanism of hydrophobic gating . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| []
| 2015 | Hydrophobic Gating of Ion Permeation in Magnesium Channel CorA |
Adequate predictions of mosquito-borne disease risk require an understanding of the relevant drivers governing mosquito populations . Since previous studies have focused mainly on the role of temperature , here we assessed the effects of other important ecological variables ( predation , nutrient availability , presence of conspecifics ) in conjunction with the role of temperature on mosquito life history parameters . We carried out two mesocosm experiments with the common brown house mosquito , Culex pipiens , a confirmed vector for West Nile Virus , Usutu and Sindbis , and a controphic species; the harlequin fly , Chironomus riparius . The first experiment quantified interactions between predation by Notonecta glauca L . ( Hemiptera: Notonectidae ) and temperature on adult emergence . The second experiment quantified interactions between nutrient additions and temperature on larval mortality and adult emergence . Results indicate that 1 ) irrespective of temperature , predator presence decreased mosquito larval survival and adult emergence by 20–50% , 2 ) nutrient additions led to a 3-4-fold increase in mosquito adult emergence and a 2-day decrease in development time across all temperature treatments , 3 ) neither predation , nutrient additions nor temperature had strong effects on the emergence and development rate of controphic Ch . riparius . Our study suggests that , in addition to of effects of temperature , ecological bottom-up ( eutrophication ) and top-down ( predation ) drivers can have strong effects on mosquito life history parameters . Current approaches to predicting mosquito-borne disease risk rely on large-scale proxies of mosquito population dynamics , such as temperature , vegetation characteristics and precipitation . Local scale management actions , however , will require understanding of the relevant top-down and bottom-up drivers of mosquito populations .
Associations between anthropogenic pressures , disease risk and vector ecology are particularly strong for mosquito-borne infections [1–4] . To date , existing predictive maps of disease risk almost exclusively focus on large-scale drivers of mosquito populations , such as temperature , precipitation , and large scale vegetation properties [5–7] . These efforts have been fuelled by observed and predicted changes of the Earth’s climate [e . g . , 8 , 9] . While temperature has indeed been shown to be a key determinant of mosquito development , survival , and fitness [9–14] it is often not fully appreciated that mosquitoes inhabit complex ecosystems and are exposed to a myriad of local biotic and abiotic factors that likely influence the dynamics of mosquito populations [15–18] . These factors operate on various scales , ranging from local-level pressures ( e . g . pesticides , eutrophication ) to regional ( e . g . land use change ) and global scales ( e . g . climate change ) . Human activities are known to strongly alter these biotic and abiotic factors through nutrient additions , biodiversity declines and climate change [19] . Understanding how these biotic and abiotic factors in turn influence mosquito-borne disease risk requires quantifying how they interact to influence mosquito population dynamics . Local mosquito population dynamics are mainly controlled by bottom-up ( food availability ) and top-down forces ( predator abundance ) [20–23] . Work by Hagstrum and Workman ( 1971 ) [22] suggests that temperature and food availability can jointly impact larval development rates ( Culex tarsalis ) . Temperature-dependent development rates were only observed in treatments with high food availability . Similarly , the effect of predators on mosquito populations may also be mediated by biotic and abiotic factors[17 , 21 , 24] , such as eutrophication , the presence of controphics as alternative prey , habitat structure and pesticide concentrations . However , our current understanding of the factors driving mosquito populations are based on experiments that were carried out under highly simplified lab conditions devoid of abiotic variability and species interactions [8 , 9 , 25] . The relevance of this work under natural environmental conditions as well as the relative importance of the drivers for mosquito populations therefore remains unknown . In this study , we used an outdoor mesocosm setup with Culex pipiens , a confirmed vector for West Nile virus , Usutu and Sindbis , to experimentally test the influence of three likely drivers of mosquito populations , representing three common anthropogenic pressures . Specifically , we manipulated nutrient concentrations , the presence or absence of predators and temperature to explore the consequences of eutrophication , biodiversity loss and climate change on mosquito population dynamics .
Two mesocosm experiments were carried out in the experimental garden at the Hortus Botanicus of the University of Leiden , the Netherlands . The two experiments focused on role of temperature in conjunction with predation of larval mosquito populations or eutrophication of mosquito populated waters . Both experiments were conducted in 65-litre polyethylene tubs filled with 12 litres of rain water , which were set up in a semi latin-square design . In order to prevent excessive heating , each mesocosms was placed into the ground so its rim was approximately ten cm above the surface . To allow for natural colonization of dipterans and standardized timing in the start of the experiment , the mesocosms were left open for 24 hours prior to both experiments . Within a single night , all mesocosms were colonized by two common Diptera species; Culex pipiens , a common mosquito species and Chironomus riparius , a controphic non-biting midge . To standardize the experimental settings , egg rafts were redistributed such that each mesocosm received two egg rafts of Cx . pipiens [in total equalling appr . 440 eggs; 26] and one egg raft of the harlequin fly Ch . riparius [equalling appr . 500 eggs; 27] . These densities were selected based on being within the observed range for Cx . pipiens , which varies widely under natural conditions [28] . Although there may be some variation in the number of eggs per raft , this is unlikely to influence the results because the egg rafts were randomly redistributed over the treatments . Preliminary experiments at this location showed that these two species typically colonize this type of habitat . To confirm that only these two species colonized our mesocosms , keys by Cranston et al . ( 1987 ) [29] for Culicidae and Langton ( 1984 ) [30] for Chironomidae were used . In the first experiment , four mesocosms were additionally colonized by herbivorous beetles , which we removed at the onset of the experiment . All mesocosms were covered with 50% shade cloth nets to prevent heating and animal escapes or introductions . Both experiments used multiple temperature scenarios , for which aquarium heaters were used ( 50W , Aquadistri UK Ltd ) . The heaters were set at 24 , 28 and 32°C in the first experiment and 18 , 22 , 26 , 30°C in the second experiment . Allowing for fluctuating day-night temperature regimes , heaters were only switched on during the daylight hours between 6AM and 10PM , which represent the minimum and maximum daily temperature in the time of year that the experiments were carried out ( S1 Fig ) . To monitor the temperature regimes , the temperature of each mesocosm was measured every 7 days using a portable hq 40d electronic multi-parameter meter ( Hach Ltd , Colorado , US ) at 6:00 AM ( night temperature ) and 12:00 PM ( day temperature ) . The same device was used to record pH and electrical conductivity ( EC ) , which were measured on a weekly basis . The average mesocosm temperature in this experiment was calculated as ( 16*measured day max temperature [measured at 12:00] + 8*minimum night temperature [measured at 6:00] ) / 24 ( Table 1 ) , where 16 and 8 represent the daylight hours and night time hours respectively . This resulted in the following mean temperatures which are presented in the remainder of this manuscript: exp . 1; 22 . 7 , 25 . 3 and 28 . 1°C; exp . 2: 22 . 1 , 24 . 1 , 26 . 1 and 26 . 8°C . Furthermore , no extra food was added to any of the mesocosms to mimic rainwater fed systems and ensure consistent and realistic nutrient concentrations . The first experiment was conducted between 15 May and 20 June 2016 in 42 mesocosms ( S2 Fig ) . Three temperature scenarios with and without predators resulted in six treatments . Each treatment contained 7 replicate mesocosms , which were set up in a modified latin square design ( S2 Fig ) . The effects of predation were investigated by adding one adult Notonecta glauca ( Hemiptera: Notonectidae , collected on the same day from a natural population in a nearby pond within a natural population ) to half of the mesocosms , five days after the experiment started . All Notonecta glauca individuals were added 8 days after the experiment started when all mesocosms had 2nd instar larvae . The temperature regimes were set immediately following egg raft redistribution and predator addition . Two of the most important ecological factors affecting predation that should be considered when designing predation experiments are the predators’ dietary preference for mosquitoes and the abundance of alternative prey for the predators [17] . Notonecta glauca is a common aquatic predator in Europe and is known for its ability to colonize new habitats [31] . Furthermore , N . glauca is a visual hunter and confirmed predator of Cx . pipiens ( S1 Table , S3 Fig ) and Ch . riparius [31] . The effect of the treatments ( predation and temperature ) on three aspects of mosquito ecology were quantified: the cumulative number of emerged adult mosquitoes after 36 days , the eventual number of surviving mosquito larvae and the number of surviving mosquito pupae after 36 days . These dependent variables were uncorrelated and analysed separately . We distinguished between pupae and larvae because the experiment was terminated before all mosquitoes emerged , and we suspected predators to have stronger negative effects on pupae than on larvae because of their relative immobility . For Chironomids , only the number of emerged adults and survival of larvae were determined after 36 days . To quantify adult emergence of both C . pipiens and C . riparius , 10x10 cm Pherocon ( Threce Adair , OK , US ) sticky fly paper sheets with a general insect attractant were fitted below the top net of each mesocosm . These were replaced twice a week and all emerged adult mosquitoes ( both species ) were counted subsequently . This is a low invasive , unbiased method to determine emergence [32] . To quantify pupal and larval survival , the number of larval and pupal dipterans of both species remaining and alive after 36 days were counted . To count the remaining Cx . pipiens pupae and larvae , mesocosms were emptied by filtering the water using a 0 . 5 mm dipping net . For Ch . riparius , only the remaining larvae were counted . The second experiment , focusing on the effect of eutrophication was conducted between 18th of August and the 15th of October 2016 . It used a modified latin square design with 48 mesocosms and six replicates per treatment ( S2 Fig ) . Eutrophication and temperature treatments ( 22 . 1 , 24 . 1 , 26 . 1 and 26 . 8°C ) , were initiated immediately following egg raft redistribution . Eutrophication treatments were applied to half of the mesocosms . An addition of 6 . 16 mL of soluble plant feed ( Nitrogen:Phosphorus:Potassium 7:4:7; Pokon Naturado , The Netherlands ) was added to half of the mesocosms to a final concentration of 9 . 0 mg inorganic nitrogen and 6 . 4 mg inorganic phosphorus per litre . These are typical values for stagnant , eutrophic , freshwater bodies [33] . The other half of the mesocosms received , as a control , a similar amount of untreated rain water . The numbers of larvae of both species were assessed on day 8 of the experiment , by gently filtering the entire volume of each mesocosm through a 0 . 5 mm sieve . This number was used for the larval development rate and survival calculations in this experiment . Emergence of the first Cx . pipiens was observed fourteen days after the experiment started , after which the emergence of adult mosquitoes and chironomids from all mesocosms was recorded daily ( between 6 and 9 AM ) , using a manual aspirator , which was a much quicker method than the sticky trap for daily collections . Newly emerged mosquitoes were sexed and counted . Mean larval development rate was calculated as follows: 1/ ( average number of days between egg and emergence ) and adult survival was calculated as follows: ( number of emerged adults ) / ( number of larvae at day 8 ) . To examine the effect of the nutrient addition on food availability , we measured the electrical conductivity ( EC ) and pH on a weekly basis . Electrical conductivity was used as a measure for the nutrient status [34] . because it reflects the abundance of microorganisms which compose the primary food for mosquito larvae [17] . Additionally , thirty days after the experiment started , chlorophyll A content was determined in each of the mesocosms . For this analysis , a subsample of 15 ml was collected from each of the mesocosms . These samples were filtered onto a Whatmann GF/F filter . Next , the filter was dissolved in 5 ml 90% acetone and allowed to break-down the algal cells for 20 hours at -20°C . Subsequently , samples were centrifuged at 1000G for 15 min at 4°C and supernatants were measured for absorbance at 620 nm using a plate reader . The experiment was terminated after 56 days when there were no more adults emerging from the mesocosms for 2 days . A single replicate mesocosm was colonized by Daphnia magna and excluded from further analysis . First , the effect of the various temperatures in both experiments , top-down and bottom-up treatments on abiotic parameters were explored . Differences in mean day temperature and mean night temperature were tested with a one way ANOVA and a post-hoc Tukey test . The effect of temperature treatments and eutrophication treatments in experiment 2 on biotic ( chlorophyll A ) and abiotic ( pH , EC ) variables were tested using linear models , where temperature was a categorical variable and eutrophication was a binomial variable . For the number of emerged adults , number of surviving larvae and number of surviving pupae at day 35 of Culex and Ch . riparius , linear models with type III sum of squares were used to test the effects of temperature , predator presence and their interaction for experiment 1 . Similarly , the effect of experimental treatments in exp . 1 and their interaction on the number of emerged adults and the number of surviving larvae at day 35 was tested . Likewise , the effects of temperature , eutrophication and their interaction in exp . 2 were tested on larval development rate and the percentage of larvae that survived until emergence . Temperature was a categorical variable with three levels in exp . 1 ( 22 . 7 , 25 . 3 and 28 . 1°C ) and four levels in exp . 2 ( 22 . 1 , 24 . 1 , 26 . 1 and 26 . 8°C ) . Predator presence ( exp . 1 ) and eutrophication ( exp . 2 ) were binomial variables representing top-down and bottom-up effects . As shown in S2 Fig , each treatment in both experiments was included only once in each row and column . Row and column could therefore be included in the model as random effects [35] . To detect the most important abiotic predictors for the abundance of Cx pipiens in exp . 2 , a generalized linear regression model was used . The full model consisted of the following non-collinear main effects: temperature , EC and chlorophyll A . Significant effects ( P < 0 . 05 ) were entered in the models in a forward stepwise fashion , starting with the most significant term . To meet assumptions of normality and homogeneity of variances , all response variables were square root transformed prior to analysis . Statistics were carried out in Statistica 7 . 0 and graphs were made in Sigmaplot 13 . 0 .
In conclusion , our results suggest that , in addition to temperature , ecological bottom-up ( nutrient availability ) and top-down ( predation pressure ) drivers can have strong impacts on mosquito life history parameters . As such , this study presents a case to consider local anthropogenic stressors in concert with climatological conditions to obtain an improved understanding of the factors driving mosquito populations . Our study may have implications for understanding mosquito-borne disease risk . By showing that mosquito survival and development rates are strongly driven by anthropogenic pressures related to global change , our results highlight two potentially important mechanisms driving spatial variation in vector abundance: eutrophication and biodiversity loss . Variation in mosquito abundance is one potentially important driver of variation in disease transmission [12 , 45] , with consequences for the size and speed of an outbreak [46] . Knowledge of the mechanisms driving variation in mosquito abundance in natural settings will be important for managing the disease risks associated with future environmental change . | Human actions have strongly altered ecosystems worldwide , through climate change , eutrophication , and biodiversity loss . The consequences of these global changes for mosquito populations could have important implications for mosquito-borne infections . Previous studies have focused on the effects of temperature from climate change , but we lack a comprehensive understanding of how ecological factors related to global change influence mosquito populations . To this end , we carried out two mesocosm experiments with the common brown house mosquito , a vector for West Nile Virus , Usutu and Sindbis . The first experiment tested how the interaction between predation and temperature affected mosquito emergence from larvae to adults; the second experiment tested how the interaction between nutrient addition and temperature affected mortality and emergence . Our results show that predator presence decreased mosquito survival and emergence , whereas nutrient additions led to an increase in emergence and a decrease in development time . Temperature and competition had no major impact . Our study suggests that , in addition to effects of climate , ecological drivers can have strong effects on mosquito populations known to transmit disease . | [
"Abstract",
"Introduction",
"Methods",
"Discussion"
]
| [
"death",
"rates",
"invertebrates",
"ecology",
"and",
"environmental",
"sciences",
"medicine",
"and",
"health",
"sciences",
"plant",
"cell",
"biology",
"chloroplasts",
"pigments",
"animals",
"developmental",
"biology",
"plant",
"science",
"pupae",
"materials",
"science",
"population",
"biology",
"insect",
"vectors",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"research",
"facilities",
"mesocosms",
"infectious",
"diseases",
"life",
"cycles",
"chlorophyll",
"materials",
"by",
"attribute",
"population",
"ecology",
"organic",
"pigments",
"disease",
"vectors",
"insects",
"arthropoda",
"community",
"ecology",
"population",
"metrics",
"mosquitoes",
"eukaryota",
"plant",
"cells",
"cell",
"biology",
"ecology",
"predation",
"trophic",
"interactions",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"species",
"interactions",
"physical",
"sciences",
"larvae",
"organisms"
]
| 2018 | Eutrophication and predator presence overrule the effects of temperature on mosquito survival and development |
For any organism , population size , and fluctuations thereof , are of primary importance in determining the forces driving its evolution . This is particularly true for viruses—rapidly evolving entities that form populations with transient and explosive expansions alternating with phases of migration , resulting in strong population bottlenecks and associated founder effects that increase genetic drift . A typical illustration of this pattern is the progression of viral disease within a eukaryotic host , where such demographic fluctuations are a key factor in the emergence of new variants with altered virulence . Viruses initiate replication in one or only a few infection foci , then move through the vasculature to seed secondary infection sites and so invade distant organs and tissues . Founder effects during this within-host colonization might depend on the concentration of infectious units accumulating and circulating in the vasculature , as this represents the infection dose reaching new organs or “territories” . Surprisingly , whether or not the easily measurable circulating ( plasma ) virus load directly drives the size of population bottlenecks during host colonization has not been documented in animal viruses , while in plants the virus load within the sap has never been estimated . Here , we address this important question by monitoring both the virus concentration flowing in host plant sap , and the number of viral genomes founding the population in each successive new leaf . Our results clearly indicate that the concentration of circulating viruses directly determines the size of bottlenecks , which hence controls founder effects and effective population size during disease progression within a host .
Virus progression within multi-cellular hosts operates via two distinct mechanisms: cell-to-cell proximal contamination and long-distance migration ( either as free infectious units or in circulating cells ) to colonize new organs and/or tissues . Both animals and plants can be considered as heterogeneous landscapes consisting of an ensemble of very different organs or “territories” , variably distant , and interconnected by a complex vascular system transferring nutrients , metabolic product and wastes , and information . Soon after entry into a healthy host , the vast majority of viruses use this connecting vasculature to travel long distances and expand their populations into virgin territories . In distant susceptible organs , it seems intuitively obvious that the number of initially infected cells , the number of viral genomes entering each of these cells , and thus the number of founders of new viral “colonies” , will depend on the concentration of infectious units transported in the plasma or sap flooding the vasculature . In other words , the virus load in the circulating flux will determine the bottlenecks in a virus population progressing within a host , and hence the effective population size and the pace at which new variants are produced , selected , and emerge . Taking an alternative view , however , one could speculate that the number of “entry points” into various organs of the host might sometimes be extremely limited , and hence the number of founder viral genomes in such territories may always be low or constant , regardless of the circulating virus load . Surprisingly , for both animal and plant viruses , experimental data supporting one or the other of these contrasting scenarios are extremely rare and only fragmentary at best . For animal viruses , many studies have quantified the virus load in the plasma of infected individuals . That this virus load changes drastically during progression of the infection , upon the onset of host defenses or during drug treatments , has been reported for several viruses , e . g . hepatitis B virus ( HBV , [1] ) , hepatitis C virus ( HCV , [2] , [3] ) , human immunodeficiency virus ( HIV , [2] , [4] , [5] ) , simian immunodeficiency virus ( SIV , [6] ) , and poliovirus [7] . The logical speculations that an increase in plasma virus load increases the infection dose in various tissues of the host [1] , [6] , [8] , in some cases allowing viral access to specific organs [7] , [9]–[12] , enlarges the viral population size and most likely augments the number of multiply infected cells , thus favoring recombination [13]–[16] , have been discussed frequently . Interestingly , a very recent study demonstrated an ongoing exchange of HIV genomes between the plasma and CD4 ( + ) T blood cells [17] . However , despite its recognition as a priority question [16] , this expected relationship between circulating virus load , effective viral population size , and the multiplicity of cellular infection ( MOI ) remains to be characterized . In plants , this question has thus far been totally hampered by long-standing technical difficulties with the collection and analysis of pure phloem sap for determining virus titer . Collection of phloem sap exuding from sieve tubes in the veins of severed or wounded stems works in only a few plant species [18] , and the wounded cells surrounding the sieve tubes always contain viruses that can contaminate the exudates . A very sophisticated alternative technique uses aphid stylets as pure-phloem-sap collecting tools [19] , but its implementation is laborious in practice and produces only minute amounts of sap per treated insect . As a result of these technical limitations , despite the fact that all plant viruses travel long distances within their host together with the elaborated sap , the viral titer ( and its putative dynamic fluctuations ) circulating within sieve tubes remains a complete mystery . Recently , we reported monitoring of within-host MOI in successive leaves of a host plant infected by Cauliflower mosaic virus ( CaMV , [20] ) . We demonstrated that the MOI varies greatly among leaf levels , increasing as the infection progresses and later decreasing before flowering and senescence of the host . Although this variation could not be formally explained , we speculated that it could be driven by changes in the virus titer circulating within the plant vasculature . Thus , more viral infectious units may enter each leaf , and each individual cell within these leaves , as the titer in the sieve tubes increases . In addition to an augmented MOI , this would translate into larger effective sizes of within-host viral populations , due to larger numbers of founder genomes in each leaf , and thus into mild ( or no ) bottlenecks at leaf entry . The viral population bottlenecks associated with leaf colonization have been investigated in several plant virus genera—Potyvirus [21] , Tobamovirus [22] , Cucumovirus [23] , [24] , and Caulimovirus [25]—but the bottlenecks detected were either not quantified [23] , [24] , quantified only at the single leaf level [22] , or calculated as a single averaged value over the whole systemically infected host plant [21] , [25] . It is interesting to note that physical host barriers [21]–[23] , [25] , [26] have often been speculated to induce bottlenecks at leaf entry . Most interestingly , however , a recent report is demonstrating that the viral dose could also be a major factor , thus convincingly concluding that it should be considered in future studies on within host population bottlenecks [27] . Here , we report the first quantification of a plant virus titer within the vasculature of its host , and monitoring of this important trait as infection progresses . In addition , we demonstrate that changes in sap virus titer correlate with changes in the size of viral population bottlenecks at leaf entry , strongly suggesting that the circulating virus load is a major factor determining the effective size of within-host viral populations .
We first conducted a pilot experiment to collect pure phloem sap from severed aphid stylets inserted precisely within sieve tubes , as previously described [19] , [28] . Although the considerable technical problems encountered led us to conclude that it was unreasonable to use this technique for further time-course experiments ( for details see online Text S1 ) , we could compile measures from 19 sap samples collected from different leaves at variable stages of development , and obtain an average value of 318 ( SD +/−244 ) viral genomes per nanolitre of sap . The implications of this order of magnitude , i . e . tens to hundreds of CaMV virions per nanolitre , are discussed further below . We next decided to estimate the virus load directly within whole aphids , processed after a 16-hour acquisition period on CaMV-infected turnip plants—a time at which aphids are most often engaged in a phase of sustained sap ingestion from the phloem sieve tubes [29] . As discussed further below , a continuous sap flux transits rapidly within the aphid gut during this sustained phloem-feeding phase , and ingested CaMV is thought to simply follow this flux from ingestion to excretion without entering and accumulating within the aphid body [30] . We also carefully targeted young sink leaves to make sure that we were indeed analyzing virus load in aphids containing phloem sap flowing into the newly developing leaves ( see Materials and Methods for a more detailed explanation ) . Using this technique , we monitored virus load in the sieve tubes at several successive leaf levels appearing on 20 plants infected in parallel , and were able to demonstrate huge temporal variations in the circulating virus titer . We measured an average of 1386 viral copies per aphid in the first systemically infected leaf level ( level 5 ) , with a sharp increase in leaf levels 9 and 14 before reaching a maximum of 11291 viral copies/aphid , and then a decrease back to initial values in leaf level 28 ( Figure 1A ) . The differences observed between successive leaf levels proved highly significant ( linear mixed-effects model , P<0 . 001 ) . The control experiment gave an estimate of what aphids can potentially acquire when feeding superficially only in epidermal and mesophyll cells . Because aphids always conduct test probes in these tissues before settling in the sieve tubes [28] , we needed to assess the number of CaMV copies acquired during these test probes and how much this could affect our measurements of virus load in the sap . Aphids of the species Myzus persicae seldom reach deep phloem tissues before a feeding time of about 30 min has elapsed [29] . We thus performed exactly the same experiment in parallel , on the same plants and the same leaves , but allowing only a short feeding period of 10 min . Figure 1A shows that the virus load within aphids that have fed only in epidermal and mesophyll cells ( dashed line ) is much lower , and can be neglected when estimating the virus load from aphids fed in sieve tubes ( solid line ) . By enlarging the scale ( Figure 1B ) , it can be seen that the time-pattern of virus copy number in aphids fed only for a short time is distinctly different from that in aphids fed for longer , confirming that “long-fed” aphids indeed access a different tissue ( sieve tubes ) . Finally , we also immediately estimated the virus load in cells of leaves used directly for aphid-feeding experiments ( Figure 1C ) . Interestingly , the range of CaMV accumulation among successive leaf levels was of the same order of magnitude ( from ∼10 to ∼40 , Figure 1C ) , suggesting that the efficiency of viral replication in these cells is rather constant . The time-pattern of virus accumulation in cells of successive leaf levels ( Figure 1C ) appeared totally different from that in the sap ( Figure 1A ) , highlighting the absence of correlation of the dynamics of the virus load in these two distinct compartments ( i . e . sieve tubes vs . mesophyll ) . Consistently , the difference in virus load in the sap between two leaf levels was not correlated to the corresponding difference in the mesophyll ( Pearson r values were equal or inferior to 0 . 417 , and p-values equal or superior to 0 . 067 ) . We next assessed whether a higher virus concentration in the circulating sap represents a higher viral dose , which would in turn increase the number of founder viral genomes in newly infected leaves . For this experiment , we conducted a similar longitudinal analysis of infected plants , measuring the size of bottlenecks imposed on CaMV populations at the entry to successive leaf levels . Fifty replicate plants were inoculated with a mixture of two CaMV variants ( Mys4 and Mys7 ) at a 1∶1 ratio , and their relative frequency was monitored in order to evaluate bottleneck sizes upon entry to leaf levels 5 , 16 and 21 , as described in the Materials and Methods . Briefly , Mys4 and Mys7 are equally competitive in doubly infected plants ( Figure 2A ) and the procedure consisted of comparing the variance in the relative frequency of Mys4/Mys7 among the 50 replicate plants , at an “initial” stage ( source population ) and at a “final” stage ( population in a specific leaf level ) . Changes in variance during passage from the initial to the final stage were used to calculate the number of CaMV genomes from the source populations that made it through to each respective leaf level ( according to [25] ) . As leaf level 5 was the first systemically infected level , we considered the corresponding source population to be that present in inoculated leaf level 2 , for reasons described previously [22] . Thus , by analyzing the viral populations in leaves 2 and 5 in each of the 50 replicate plants , we were able to estimate the size of the bottleneck through which the CaMV population passes when leaving the inoculated leaf to initiate systemic infection . This protocol is not destructive since the inoculated leaves are collected just before senescence , and leaves at level 5 are punctured only to extract few leaf discs . In contrast , the initial source populations from which viral genomes in leaf levels 16 and 21 originate are much harder to define because many leaves below them are exporting virions . We made the same assumption as in [25] and considered that , due to anastomosis of the vasculature connecting different leaves , all viral genomes present in all leaves below leaf level 16 ( or 21 ) provide the best estimate of the source population feeding into this leaf level . Because in this case the protocol is destructive ( most infected leaves are finally harvested ) , we could not analyze levels 5 , 16 and 21 with the same set of plants . We thus estimated the bottleneck at entry of leaf levels 5 and 16 with one set of 50 plants , and that at entry of leaf level 21 with a second similar set under the same experimental conditions . The results , summarized in Figure 2B , illustrate the remarkable variation in bottleneck size at the entry to different leaf levels . While the number of founder viral genomes is very low in the first systemically infected leaf ( around ten per leaf ) , it rises by over an order of magnitude in leaf-level 16 , before decreasing back to initial values in leaf level 21 . To confirm the observed changes in bottleneck size over time ( Figure 2B ) , we argued that one could simply track the dynamics of variance in the relative frequency of Mys4/Mys7 among replicate plants at successive leaf levels . As Figure 3A illustrates , if the within-plant effective population size were constant the variance of the marker frequencies between plants would increase monotonically . On the other hand , temporal variations of the within-plant population size are expected to induce a non-monotonic behavior of the among-plant variance . For example , if the population size is first small , then large and then small again , the among-plant variance in marker frequency is expected to first increase rapidly , then increase very slowly and then resume a rapid increase . Thus , simply tracking the between-plant variance , without the need to identify and analyze the initial source populations entering into each leaf level , should allow us to infer the pattern of bottleneck size changes with time , although it cannot allow actual estimation of bottleneck size per se . We first plotted the variances of marker frequency calculated from the two sets of plants used in Figure 2B ( dashed line ) , and verified that it increased when the bottleneck at leaf entry was narrow , whereas it remained approximately constant when the bottleneck was relaxed . Then , we decided to confirm the pattern of bottleneck size changes over time on a new single set of 40 Mys4/Mys7-infected plants . We harvested leaf levels 5 , 10 , 16 , 21 and 27 , leaving all other leaves intact , thus allowing the plants and the infection to grow continuously and develop throughout the experiment . Consistently , Figure 3B shows that the among-plant variance of Mys4 relative frequency increases between leaf levels 2 and 5 , stabilizes in leaf levels 10 and 15 , and again increases in leaf levels 21 and 27 . We verified that the results were better explained by a non-monotonic function , such as a hyperbolic-sinus function , as opposed to a monotonic function , such as a logarithmic function , by comparing their Akaike's Information Criterion ( AIC ) values . The best model ( lowest AIC ) was the hyperbolic-sinus function ( ΔAIChyperbolic-logarithmic>2000 ) , confirming the pattern of bottleneck changes shown in Figure 2B . Remarkably , the pattern of bottleneck size changes at successive leaf levels ( Figures 2 and 3 ) resembles that of the virus load measured in sieve tubes of corresponding leaf levels ( Figure 1 ) , strongly suggesting a correlation between the two . Though stemming from different plant sets , the above results together suggest a match between the pattern of the virus concentration within the sap and that of the size of the bottlenecks endured by CaMV populations when initially colonizing leaves . In order to confirm this relationship in a single experiment , we further analyzed DNA samples used in Figure 1 . As indicated in the Materials and Methods , the plants were infected by a mixture of the two CaMV genotypes ( Mys4/Mys7 ratio 1∶1 ) , allowing the monitoring of the variance of this ratio both within aphids ( so within the phloem sap ) and within the corresponding leaf tissues , thus allowing to calculate the number of viral genomes passing from the sap into the leaf tissues . The data summarized in Figure S1 ( in Text S1 ) consistently confirm an increase of the number of viral colonizers within leaves , when the viral load increases in the sap . Early and late in infection , when viral sap-load is minimum , the size of CaMV population bottlenecks is close to 10 , whereas it rises above 70 as viral sap-load drastically increases . The sizes of CaMV populations colonizing leaves were very similar to those calculated in the other set of plants in Figure 2B .
Unlike the case in mammals , the virus load in the vasculature of plants represents hard-to-access information . There is no reported method of artificially puncturing sieve tubes and aspiring sap; simply severing the stem is , in most cases , totally inefficient as in many plant species ( including turnip as used here ) no liquid exudes from the wound [18] . In addition , the viral content of companion and other cell types that always surround the sieve tubes represents an inevitable source of putative contamination upon wounding . For these reasons , and because they are natural entities capable of pumping pure sap , in this study we attempted to exploit aphids to access this plant compartment . Because of the multiple drawbacks encountered by us ( see Text S1 ) and others [19] , the sophisticated stylectomy technique was implemented only in an exploratory experiment , not differentiating leaf level and leaf age . How the resulting range of tens to hundreds of CaMV genome copies per nanoliter ( average+/−SD 318+/−244 copies/nl ) compares with values quantified from whole aphids ( Figure 1A ) depends on the volume of sap that an aphid actually contains , and whether or not virions are massively degraded in the aphid gut . It can be calculated from [31] that aphids of the species Myzus persicae ( Sulz . ) can individually contain around 25 nl of sap , continuously flowing and rapidly transiting ( about 1 hour ) from ingestion to excretion . This volume would consistently correspond to a range of 55–452 virus copies per nanoliter of sap , depending on the leaf level ( number of copies estimated from pools of five aphids , divided by 5 to provide a per aphid number as in Figure 1 , and by 25 to provide a per nanoliter number ) . The current literature suggests that aphids ( and other hemipteran insects ) readily assimilate free amino acids , but poorly degrade proteins [32—34][35] . However , in additional experiments where aphids were fed with artificial solutions of known virus concentrations , a similar calculation of the virus load in aphids ( as above ) gave values about one order of magnitude lower than that measured in the feeding solutions ( Figure S2 in Text S1 ) . As detailed in the section 6 of the Text S1 , despite clear evidence that aphids feed much less on artificial membranes than on plant leaves ( Figure S3 and Table S5 , in Text S1 ) , we cannot exclude a possible degradation of some virions within the gut . All caution considered , we thus propose the above values ( 55–452 virus copies/nl ) as a conservative estimate of the average CaMV sap load in turnip hosts . In any case , would aphids partly decrease the number of detectable viral copies per nanoliter of sap , they would most likely equally do so for all leaves tested , letting the dynamic pattern in Figure 1 unaffected . The approximation reported here for the viral sap load is certainly specific to CaMV infecting turnip; however , it represents the first estimation of the load of a plant virus circulating within the vasculature of its host . In addition to the cautionary remark mentioned above , to reliably reflect the virus load in plant sap , the technique used in Figure 1 requires the use of an insect species specifically feeding in sieve tubes , and in which the virus does not accumulate . This condition is fulfilled in the case of aphids and non-circulative viruses ( like CaMV ) , because only a few viral particles are specifically retained within the anterior part of the feeding apparatus [36]–[39] ( see also dashed line in Figure 1A ) . Circulative viruses appear more problematic , as they can penetrate the body of their vectors and accumulate , or even replicate , in gut cells , salivary gland cells , or elsewhere [40] . We propose that by using phloem-feeding insect species that are non-vector , and by verifying that they do not accumulate the studied viruses , this technique can be transferred to virtually all plant virus species . It is generally and intuitively assumed that the number of viral particles circulating within the vasculature depends on the replication rate in the cells that shed virions into the plasma or sap , and on the number of such infected cells within the host [6] , [41] , [42] . Accordingly , early studies quantifying the accumulation of Tobacco mosaic virus in different leaves ( reviewed in [43] ) speculated that changes in within-leaf accumulation might correlate with changes in sap load . However , because we found small differences in CaMV accumulation at comparable development stages of successive leaf levels ( cf Figure 1A and 1C ) , and because the CaMV load in the mesophyll does not correlate with that in the sieve tubes , we believe that changes in the CaMV replication rate are not responsible for the variations in the sap observed here . Rather , it is the fact that more cells and leaves are becoming systemically infected , and thus cumulatively shedding virions within the sieve tubes , that accounts primarily for the increase in viral sap load , as previously speculated [20] . In contrast , the drop in virus load late in infection is surprising and invites speculation on several hitherto unreported aspects of plant virus biology , including a possible arrest of virion export from infected leaves , an increased rate of virion degradation within the sap , or massive and rapid storage in unknown plant compartments ( for example roots ) clearing the vascular system . For animal viruses , virion turnover in plasma results from the constant production by infected cells , and rapid degradation due to intrinsic instability and attack by the immune system . The half-life of circulating virions—studied for HCV , HIV and HBV [1] , [2]—has been demonstrated to be extremely short , in the range of one to a few hours . Equivalent studies do not exist in plants , and there is no mention of any possible virion decay and/or turnover within the sap in the available literature . Although CaMV produces very stable virus particles , as demonstrated by purification procedures involving 1 . 5 M Urea , 2% Triton , 1 CMC β-OG and butanol treatments [44] , our results suggest that they might be degraded in ( or removed from ) the sap with an unknown half-life time . Whether the leaves stop exporting virions at some point , or whether unknown mechanisms act to increase degradation , or to sequester virions in an as yet undetermined compartment , is unclear , but such questions set an interesting scene for future investigations . Variation in the size of the bottlenecks undergone by CaMV populations at leaf entry has been analyzed here with previously described procedures . Entry of leaf level 5 was analyzed using a protocol similar to that described by Sacristan and co-workers [22] , which identified precisely the initial ( within leaf 2 ) and final ( within leaf 5 ) populations . In contrast , when quantifying the CaMV genome founders colonizing leaf levels 16 and 21 , the originating population is elusive and has been suggested to be best illustrated by the overall population present within the plant [25] . Because the two different protocols used in these previous studies have been suspected to differentially affect the outcome of the experiments [45] , we developed a third and novel approach ( Figure 3 ) , allowing a similar analysis at all leaf levels in a single set of plants . In this approach , approximately one leaf level out of five is harvested , thus plant development is affected only minimally . While the sizes of the bottlenecks cannot be estimated in this way due to a lack of information on the source population colonizing each individual leaf , the pattern of dynamic changes in bottlenecks can be tracked efficiently together with the spread of infection . This additional experiment demonstrated that bottlenecks are consistently narrow at early and late time points in infection , and relaxed at intermediate stages , thus signifying that the two distinct protocols previously used for bottleneck quantification do not bias the results . Interestingly , the viral population within the sap flowing into leaves is the real source population , and we could use it to calculate and confirm the bottleneck sizes at leaf entry ( Figure S1 , in Text S1 ) . Though in this experiment , the leaves were at early stages of development ( still sink ) , the estimated number of viral colonizers was remarkably similar to that established from fully matured leaves ( compare Figure 2B and Figure S1 in Text S1 ) . The different sets of plants analyzed indicate that the dynamic pattern of virus load within the sap is very similar to that of bottleneck size at the entry to successive leaf levels . This strongly suggests a direct relationship between the two , and we propose that ( i ) in the first systemically infected leaf , the virus titer in the sap is far too low to saturate all existing entry points , and the limiting factor is thus the availability of virions; ( ii ) as more leaves become infected and shed virions within the sap , the virus titer increases and relaxes the bottleneck during disease progression , with no apparent limit due to entry points; and finally ( iii ) the situation reverts later in infection by unknown mechanisms that either halt virus export from infected cells , store virions elsewhere than in the vasculature , or accelerate virion decay in the sap ( or a combination thereof ) , resulting in a drop in the number of circulating virions and thus of the size of bottlenecks . Interestingly , we previously described a very similar pattern of dynamic changes in the MOI of CaMV at different leaf levels , indicating that the CaMV load in the sap probably influences the number of viral genomes entering leaves as well as individual cells within the leaves [20] . In the scenario suggested here for CaMV , the most significant bottleneck encountered during the virus life cycle is certainly that imposed by aphid transmission , during which very few genomes might be inoculated , as suggested by studies with other non-circulative viruses [36] , [38] , [46] . The infection would then evolve with an increasing population size limited only by the number of virions produced and loaded into the vascular system . Whether this scenario holds true for other plant viruses is totally unknown . Interestingly , the procedure developed here to study in parallel the fluctuation of virus sap-load and the number of founders entering each leaf is certainly transferable and may inform on this question .
The genome of CaMV ( genus Caulimovirus ) is a circular dsDNA of approximately 8000 bp ( depending on the strain ) , encoding seven independently translated open reading frames ( ORF I-VII ) [47] . The QuickChange Site-Directed Mutagenesis kit ( Stratagene ) was used to insert 39-bp oligonucleotides between ORF I and ORF II of plasmid pW260 [48] , to be used as genetic markers as previously described [49] . Two distinct clones were engineered with this technique ( primer sequences available upon request ) , containing 39 additional nucleotides ( TCTACATATTCCTGATAACTCAACGGTCGTCGACGGAGT or AGTAAGTGCTGTAAGTATAATAAGGATACTTGTCGACAG ) between the stop codon of ORF I and the start codon of ORF II . These two clones were named Mys4 and Mys7 , respectively , and a real-time PCR protocol was developed for their specific quantification in DNA mixtures . PCR conditions and primers are detailed in the online supporting information ( Text S1 ) . Clones Mys4 and Mys7 were tested for infectivity , and the stability of the introduced genetic markers was confirmed by sequencing viral populations after two serial passages of 21 days each in turnip host plants . The symptoms induced by both Mys4 and Mys7 were similar to those induced by the parental pW260 clone . Virus particles were purified from infected plants , quantified and stored as previously described [49] , [50] . Colonies of the aphid species Myzus persicae were maintained in insect-proof cages on eggplant , in a growth chamber at a temperature of 23/18°C and a photoperiod of 14/10 h ( day/night ) , conditions ensuring that the colonies are reproducing clonally . The colonies were transferred to a new cage and new host plants every two weeks , and aphids were always collected at the moment of the transfer , thus at comparable population densities . Turnip plants ( Brassica rapa cv . “Just Right” ) were maintained in an insect-proof growth chamber under controlled conditions ( temperature 24/15°C , photoperiod 15/9 hours day/night ) . For all experiments , plantlets at the third leaf stage were inoculated mechanically by rubbing a virus suspension containing 400 ng of virus particles and Carborundum abrasive powder , on the second leaf level . All inoculums were prepared by mixing Mys4 and Mys7 purified virus particles at a 1∶1 ratio . The appearance of new leaves ( budding ) on the inoculated plants was monitored and noted daily in order to allow leaf sampling at a precise leaf age . Depending on the experiment , either entire leaves or six leaf discs ( 0 . 8 cm ø ) distributed evenly over the leaf surface were sampled . DNA from each leaf sample was extracted as described [51] . Twenty replicate plants were inoculated and sampled at four time points . The different time points corresponded to leaf levels 5 , 9 , 14 and 28 . The first systemically infected leaves appeared at leaf level 5 , the first visual signs of flowering induction ( changes in leaf morphology ) appeared irregularly between leaf levels 21 and 28 , and the plants slowly entered senescence after leaf level 28 . We originally planned to collect also leaf level 21 , but our aphid colony unfortunately collapsed at this time and recovered only in time to use leaf level 28 . Because of the length of the experiment , different aphid cohorts were used at different dates for different leaf levels , but they all originated from the same clonal rearing , maintained in constant conditions and collected at comparable population densities . It is important to note that the sampling dates were chosen when leaves were in their 5th day of development . At this stage , the sink-to-source transition has not yet occurred and the phloem sap flows into the developing leaves , whereas this flow is inverted after the transition that occurs on approximately the 10th day of leaf development , when each leaf exceeds 1/3 of its final size [52] . For each time point , fifty aphids were confined in a “cage” enclosing the defined leaf level ( see above ) on each of 20 replicate plants . Three groups of five immobile aphids—those most likely feeding—were collected from each leaf after an acquisition period of 10 minutes in epidermal and mesophyll cells [29] , and instantly frozen in liquid nitrogen . The remaining aphids were caged again and left on the leaves overnight ( 16 hours ) . After this overnight period , three groups of five immobile aphids were collected from each leaf and similarly frozen in liquid nitrogen . M . persicae generally settle and feed continuously within the sieve tubes during such a long period [29] , and we thus assumed that most of them would contain sap from which the CaMV genome copy number could be quantified . As CaMV is transmitted in a non-circulative manner , it does not accumulate within the vector body and is believed simply to follow the sap flow in the aphid's gut , from ingestion in the stylets to the excretion of honeydew [30] . Finally , all remaining aphids were discarded and the entire corresponding leaves were immediately collected and stored at −20°C until use . The DNA from each group of 5 aphids was extracted as previously described [53] and analysed by Q-PCR . The total DNA from each leaf was also extracted as described [51] , and analyzed by Q-PCR . The validation of this approach by feeding aphids with artificial suspensions with known virus concentrations is presented in the point 6 of the Text S1 . We estimated the size of the CaMV population bottlenecks during colonization of leaf levels 5 , 16 and 21 , which appear successively on infected plants . In this experiment , leaves were collected on the 13th day of development , after the sink-to-source transition , when colonization by viruses imported from the phloem sap is over [52] . Two sets of 50 plants were inoculated with a mixture of Mys4 and Mys7 purified virions , on leaf level 2 . In both sets , we collected the inoculated leaves just before their death in order to minimally affect ( if at all ) virus exit and migration towards systemically infected leaves . In addition , in one of the sets of plants , we sampled leaf discs on leaf level 5 and let the plants grow until leaf level 16 appeared . Using an approach similar to [25] , we then collected a leaf pool containing all leaves except leaf 16 , which was left to develop for 13 days and finally harvested . The second set of plants was treated similarly for analysis at leaf level 21 , collecting pooled leaves below nascent leaf 21 , which was finally collected 13 days later . DNA was extracted from all leaf samples , and the relative frequency of Mys4/Mys7 was estimated using Q-PCR as indicated above . In the first plant set , we determined the variance of Mys4/Mys7 relative frequency among leaves 2 ( a ) , among leaves 5 ( b ) , among pools of leaves below leaf 16 ( c ) , and among leaves 16 ( d ) . In the second plant set , we determined the variance of Mys4/Mys7 relative frequency among pools of leaves below leaf 21 ( e ) , and among leaves 21 ( f ) . Comparing a–b , c–d , and e–f allowed evaluation of the size of the bottlenecks at the entry of leaf levels 5 , 16 and 21 , respectively , as described [25] . In order to rapidly establish the pattern of bottleneck changes in a viral population invading successive leaf levels , we tracked the among-plant variance in relative marker frequency over time . Severe bottlenecks should increase among-plant variance , while mild bottlenecks should affect it only slightly . We expected that if the viral effective population size was constant throughout the infection the among-plant variance would increase monotonically , while temporal variation in the effective size would induce non-monotonicity in the pattern of variation of among-plant variance . To illustrate the patterns that could arise from such temporal variation in the effective population size , we simulated the evolution of a single diallelic neutral locus in a number of unconnected populations , representing the different host plants . All populations had the same effective size at any given time , but their effective size could vary over time . We computed the among-population variance in allele frequencies over time . The results , e . g . Figure 3A , confirmed our expectation: when the populations evolve under constant effective size , the among-populations variance increases monotonically . When the effective population size varies over time , the increase in among-populations variance is fast when the effective population size is small and slow when it is large . To experimentally investigate how among-plant variance in relative marker frequency evolved over time , 40 plants were inoculated in parallel and sampled at six different time points , corresponding to leaf levels 2 ( inoculated leaf ) , 5 , 10 , 16 , 21 and 27 . Each leaf level ( except leaf level 2 , which was collected just before death ) was sampled when leaves were in their 13th day of development by collecting the entire leaf . The Mys4/Mys7 relative frequency was estimated ( by Q-PCR as above ) in successive leaf levels on the 40 repeated plants , and the function best describing changes of the Mys4/Mys7 variance was determined as follows . To test whether the empirical results were better explained by monotonic or non-monotonic functions , we fitted phenomenological models whose behaviour a priori fitted that of the simulation results in Figure 3A . Thus a logarithmic function ( y = a+b*ln ( x ) ) , and a hyperbolic-sinus function ( y = d* ( b/d+sinh ( c+ ( x−a ) ) ) ) were fitted to the data , plotting changes in the Mys4/Mys7 variance over leaf levels , using a maximum likelihood approach . These two functions were chosen because they could correspond to different possible patterns of changes of the viral population bottlenecks during disease progression . The logarithmic function illustrates cases where the bottleneck size is constant . The hyperbolic-sinus function illustrates the case where bottlenecks are severe at first , then relaxed for a while , and become severe again late in infection ( see text of the Results section ) . The AIC was used to decide which of these functions best explains the observed distribution of the variance values at successive leaf levels . We previously reported two methods for evaluating the size of CaMV population bottlenecks , one based on a simple analysis of changes in the variance of Mys4/Mys7 relative frequency at different leaf levels , and the other using Fst statistics [25] . The application of both methods , including a slight adjustment required for the analysis of bottlenecks at leaf level 16 , to our data sets is provided in the online Text S1 . As already noted when first describing these methods [25] , they provide very similar estimations of the bottleneck sizes , and we used only results obtained with the former in the Results section . Other statistical analyses used various classical tests , which were all performed with the R ( v2 . 11 ) package . The nature of the tests and their results are indicated in the text . | Infecting viruses progress within multi-cellular hosts via two distinct mechanisms: cell-to-cell proximal contamination and long-distance migration to remote organs through the vasculature . In distant susceptible organs , it seems logical that the number of initially infected cells , the number of viral genomes entering each of these cells , and thus the number of founders of new viral “colonies” , depends on the concentration of infectious units transported in the vasculature . For any organism , the number of founders colonizing a “virgin territory” , is of prime importance in determining the forces driving its evolution . This is particularly true for viruses where the so-called founder effect is a key factor in the emergence of new variants with altered virulence . It is surprising to note , however , that whether the circulating virus load directly drives the size of viral populations during host colonization remains elusive . By monitoring for the first time the virus concentration flowing in host plant sap , in parallel with the number of viral genome founders in each successive leaf , we provide unequivocal evidence that the concentration of circulating viruses can directly determine the founder effect and effective population size during disease progression in a eucaryotic host . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"viral",
"transmission",
"and",
"infection",
"plant",
"biology",
"population",
"dynamics",
"population",
"genetics",
"microbiology",
"host-pathogen",
"interaction",
"plant",
"science",
"effective",
"population",
"size",
"plant",
"pathology",
"population",
"biology",
"viral",
"load",
"viral",
"entry",
"biology",
"viral",
"evolution",
"genetic",
"drift",
"virology",
"genetics",
"and",
"genomics"
]
| 2012 | Circulating Virus Load Determines the Size of Bottlenecks in Viral Populations Progressing within a Host |
Acute respiratory distress syndrome ( ARDS ) is the leading cause of death in critical care medicine . The syndrome is typified by an exaggerated inflammatory response within the lungs . ARDS has been reported in many species , including dogs . We have previously reported a fatal familial juvenile respiratory disease accompanied by occasional unilateral renal aplasia and hydrocephalus , in Dalmatian dogs . The condition with a suggested recessive mode of inheritance resembles acute exacerbation of usual interstitial pneumonia in man . We combined SNP-based homozygosity mapping of two ARDS-affected Dalmatian dogs and whole genome sequencing of one affected dog to identify a case-specific homozygous nonsense variant , c . 31C>T; p . R11* in the ANLN gene . Subsequent analysis of the variant in a total cohort of 188 Dalmatians , including seven cases , indicated complete segregation of the variant with the disease and confirmed an autosomal recessive mode of inheritance . Low carrier frequency of 1 . 7% was observed in a population cohort . The early nonsense variant results in a nearly complete truncation of the ANLN protein and immunohistochemical analysis of the affected lung tissue demonstrated the lack of the membranous and cytoplasmic staining of ANLN protein in the metaplastic bronchial epithelium . The ANLN gene encodes an anillin actin binding protein with a suggested regulatory role in the integrity of intercellular junctions . Our study suggests that defective ANLN results in abnormal cellular organization of the bronchiolar epithelium , which in turn predisposes to acute respiratory distress . ANLN has been previously linked to a dominant focal segmental glomerulosclerosis in human without pulmonary defects . However , the lack of similar renal manifestations in the affected Dalmatians suggest a novel ANLN-related pulmonary function and disease association .
Acute respiratory distress syndrome ( ARDS ) is a multifactorial syndrome characterized by rapid-onset respiratory failure resulting from pulmonary inflammation [1] . ARDS is common and leads to substantial mortality in man [1] . Two forms of idiopathic interstitial pneumonia are found in human ARDS: a diffuse alveolar damage ( DAD ) in acute interstitial pneumonia ( AIP ) and an acute exacerbation of usual interstitial pneumonia in idiopathic pulmonary fibrosis [2] . The molecular mechanisms leading to human ARDS remain largely unknown . Candidate gene studies suggest the involvement of inflammatory mediators such as interleukins IL-6 , IL-8 and IL-32 [3 , 4] , pre-B-cell colony-enhancing factor ( PBEF ) [5 , 6] , and angiotensin-converting enzyme ( ACE ) [7] . In addition , the nuclear factor erythroid-derived 2–like 2 ( NFE2L2 ) transcription factor has been identified as a potential mediator of acute lung injury in a mouse model [8] . Spontaneous ARDS has also been described in dogs [9] . We previously described a familial fatal ARDS-like syndrome in young Dalmatian dogs with the main clinical signs including progressive tachypnea and dyspnea leading to severe respiratory distress and euthanasia [10] . The clinicopathological findings were restricted to pulmonary lesions in the majority of the affected Dalmatians , although some of the puppies presented with concurrent unilateral renal aplasia and hydrocephalus [10] . Pulmonary manifestations included multiple foci of marked atypical hyperplasia and squamous metaplasia of the bronchiolar epithelium , patchy ongoing fibrosis with myofibroblastic metaplasia , smooth muscle hyperplasia and occasional honeycombing of alveoli and hyperplasia of type II pneumocytes ( PCII ) along with acute alveolar edema [11] . Exclusion of specific causes like exposure to toxins and viruses and overrepresentation of the cases in the Dalmatian breed suggested a recessive genetic defect [10] , which we aimed to reveal in this study . We describe the identification of a fully penetrant recessive nonsense variant in a novel candidate gene , ANLN . The ANLN gene encodes an anillin actin binding protein which has an important role in the integrity of the epithelial cell organization . The functional defect of ANLN due to early truncation is consistent with the observed histopathology with hyper- and metaplasia of the bronchiolar epithelium , consecutive DAD and clinical ARDS .
To identify the genetic cause of ARDS in Dalmatians , we performed a combined analysis of homozygosity mapping and whole genome sequencing ( WGS ) . The study cohort of eleven Dalmatians including two affected littermates , one healthy obligate carrier , one healthy sibling , one healthy grandparent and six other healthy dogs were genotyped using Illumina’s CanineHD SNP array . Genotype data of two cases was used for homozygosity mapping , which revealed 49 shared homozygous regions ( S1 Table ) . Whole genome sequencing with mean coverage of 16x was performed on one affected dog . The filtering of variants from WGS data under recessive model of inheritance against WGS and exome variant data of 136 unaffected dogs ( S2 Table ) uncovered 16 , 195 case-specific variants of which 98 were exonic ( Table 1 ) . Only 15 out of the 98 coding variants were found in the homozygosity regions of which eight variants were either non-synonymous ( n = 4 ) , frameshift ( n = 3 ) or nonsense ( n = 1 ) ( Table 2 ) . Manual inspection of the two PSMD6 deletions with Integrated Genome Viewer ( IGV ) revealed that the variants were also present in several control genomes ( S2 Table ) , excluding the gene as a candidate for the disease . The other six exonic variants in ANLN , CD302 , GANAB , MUC5B , OR16D05 and ORO8C02 genes ( Table 2 ) were genotyped in a cohort of twelve dogs , including seven cases and five closely related unaffected dogs ( parent , grandparent and three healthy siblings ) . Variants in the CD302 , GANAB , MUC5B , OR16D05 and ORO8C02 genes were excluded as they did not agree with a recessive segregation pattern . A complete segregation was found only for the c . 31C>T variant in the ANLN gene . All seven affected dogs were homozygous , both the parent and the grandparent were heterozygous , and healthy littermates were either heterozygous ( 1/3 ) or wild-type ( 2/3 ) ( Fig 1A and 1B ) . The ANLN c . 31C>T variant is located within a 2 Mb region of continuous homozygosity on chromosome 14 at 46 . 78–48 . 71 Mb ( Fig 1A ) . The 2 Mb haplotype is part of the larger 14 . 6 Mb homozygosity region identified by homozygosity mapping ( S1 Table ) . The entire 14 . 6 Mb homozygosity region had an average 15x coverage ( >90% bases with at least 10x coverage ) and did not contain other case-specific coding homozygous variants . We manually assessed the haplotypes surrounding the ANLN variant in eleven dogs that were genotyped with the SNP array ( Fig 1A ) . The 2 Mb haplotype was homozygous in both cases and in the parent and the grandparent . However , only the two cases were homozygous for the ANLN c . 31C>T variant while the parent and the grandparent were heterozygous , suggesting a recent origin of the risk variant in the pedigree . Further validation of the ANLN c . 31C>T variant in 176 randomly selected unaffected Dalmatian dogs revealed a low 1 . 7% carrier frequency ( 3/176 ) . No new genetically affected dogs were found while three new non-Finnish carriers were identified . The pedigree information of these three dogs was not available and therefore the relationship to the original Finnish discovery population remains unknown . A combined analysis in the entire cohorts of 188 dogs ( 12 +176 Dalmatians ) gave a highly significant association between the T allele and the disease ( p = 3 . 075x10-58 ) . Breed-specificity was studied by screening the ANLN variant in 31 Pointers , which is considered as the closest relative to Dalmatians . None of the Pointers carried the variant . The identified ANLN c . 31C>T nonsense variant is predicted to result in an early truncation of the normal 1121 amino acid ANLN protein after the first ten residues ( Fig 1C ) . This severe truncation very likely completely abrogates the ANLN function . The effect of the truncation could not be assessed at transcriptional level , as fresh RNA samples from the affected dogs were not available . Therefore , the ANLN expression was analyzed by immunohistochemistry , using antibody recognizing residues 1074–1124 of the protein , expected to be lacking in the affected dogs . Various organs from an age- and breed-matched control dog were included , and the staining was compared to that in the lungs and other tissues available from four affected Dalmatians . In addition , the expression pattern in the lungs of the affected dogs was compared to a canine lung affected by diffuse alveolar damage ( DAD ) to include the assessment of PCII cells , which are important for normal regeneration process in the alveolus [9] . The kidneys from seven affected Dalmatians were histologically re-evaluated , since loss of ANLN function has been linked to human focal segmental glomerulosclerosis ( FSGS ) [12] . Glomerular collagen was highlighted by Masson-trichrome staining and basal membranes by periodic acid-Schiff staining ( PAS ) in order to reveal even subtle fibrosis . We found a specific positive membranous ANLN-signal in the epithelial cells lining the terminal bronchioles in the control sample ( Fig 2A ) and in the control lung affected by DAD . In addition , proliferating PCII cells showed a strong cytoplasmic positivity in the canine lung affected by DAD ( Fig 2C ) . In contrast , the membranous bronchiolar epithelial ANLN-signal as well as the cytoplasmic PCII staining were absent in the lung specimens of the affected Dalmatians ( Fig 2B and 2D ) . Few basal bronchial and alveolar interstitial cells showed nuclear positivity in all groups . In other organs , ANLN expression was neither detected in the control dogs , nor in the tested affected dogs . Glomerular tufts appeared histologically normal , with no excessive collagen deposition ( Fig 2E ) and slender glomerular basal membranes ( Fig 2F ) . These results support the conclusion that ANLN is mainly expressed in the canine lung , co-localizes with the histological lesion and is lacking in the affected Dalmatians . Specific ANLN expression was not detected in organs other than lung in the canine control samples . Lesions similar to those described in human FSGS were not present in the kidneys of the affected dogs .
This study reveals the primary cause of ARDS in Dalmatian dogs by identifying a recessive nonsense variant ( c . 31C>T , p . R11* ) in the ANLN gene . Several findings support the causality of the ANLN defect in ARDS . First , the nonsense variant was the only case-specific coding change within the homozygosity regions that fully segregated with the disease ( p = 3 . 075x10-58 ) . Second , the genetic defect results in the very early truncation abolishing the function of the protein . Third , the ANLN protein was found to be expressed predominantly in the lungs , which is the key affected organ in ARDS . ANLN was absent in the lungs of the affected dogs . Finally , ANLN is a relevant functional candidate gene , since it plays a role in cell division and in the assembly of intercellular junctions [12–14] . Histopathology of the affected dogs show a disorganized bronchiolar epithelial regeneration attempt and disturbed alveolar epithelial regeneration [10 , 11] , which could be due to improper ANLN contributions . Therefore , our study has important implications since it uncovers a novel candidate gene for ARDS and sheds new light on the understanding of the underlying pathophysiology . The ANLN gene encodes a 124 kDa intracellular multi-domain protein ( Fig 1C ) that is expressed in several organs , including lungs , kidney and brain [14–16] . ANLN interacts with F-actin and CD2-associated protein ( CD2AP ) [12] and is implicated in cytoskeletal dynamics [14] . As ANLN interacts both with the organizing mitotic spindle during mitosis and the cytoskeletal actin in cellular migration [15] , the absence of the protein and consequent disturbed migration and proliferation of PCII at the alveolar level could trigger the immense bronchiolar epithelial regeneration attempt seen on the histopathology of the affected Dalmatians . One of the key events in the repair of alveolar injury involves the proliferation and migration of ANLN-positive PCII cells [9] . In addition , the atypical , broad based and multinucleated PCII of the affected dogs correlate morphologically with the disturbed cell division , likely caused by the absence of functional ANLN . An alternative or complementary hypothesis is that the loss of ANLN results primarily in the disorganized bronchiolar epithelia in the affected Dalmatians due to improper assembly of intercellular junctions . Hereby the hyper-and dysplastic epithelium acts as a mechanical hurdle at the bronchioalveolar junction during expiration and air is trapped at the alveolar level during passive exhalation , causing over-extension damage to the alveolar wall and ARDS . A comparable pathogenesis leads to ARDS in ventilator-induced lung injury in man and dog , where alveolar over-inflation , with consecutive alveolar edema and alveolar emphysema , progresses into interlobular emphysema and pneumomediastinum [9] . Seven of the affected dogs suffered from marked alveolar edema , five from marked alveolar emphysema [10 , 11] . In addition , three dogs developed pneumomediastium [10] , which is rare in dogs not suffering from perforating trauma of the esophagus , neck or trachea . ANLN has been associated in diverse forms of neoplastic disease in man [15 , 17–20] with a proposed role in regulating intercellular adhesion in the epithelia [21] . Interestingly , another scaffold protein Alix has recently been associated with the maintenance of epithelial cell polarity and assembly of intercellular junctions [16] . Abnormal structure of the choroid plexus epithelium and ependymal in the Alix knockout mouse results in enlargement of the lateral ventricles and hydrocephalus as the homeostasis of the blood-cerebrospinal fluid barrier requires intact tight junctions . Alix , just like anillin , interacts with F-actin and in addition , with tight junction protein ZO-1 , being essential for the maintenance of epithelial cell polarity and barrier . Some of the ARDS-affected Dalmatians manifested also hydrocephalus and renal aplasia [10] , which could be caused by abnormal assembly of intercellular junctions in the epithelium of the choroid plexus and the ureteric bud epithelium during early organogenesis . Mutations in the ANLN gene have been linked to human FSGS and ANLN has been suggested to play a role in retaining the podocyte function in the glomerular filtration barrier [12] . We did not identify ANLN in the normal canine glomeruli by IHC staining . These findings are similar to the IHC findings in man , where ANLN expression was upregulated in the glomeruli affected by FSGS but not detected in normal glomeruli . Apart from unilateral renal aplasia in two affected dogs , serum biochemistry and renal histology of the affected Dalmatians did not reveal other renal disease [10 , 11] . The suggested conclusion [12] that ANLN expression is induced in response to podocyte injury and repair , not in the end-differentiated mature podocyte , remains unverified in dogs as early onset and lethal outcome of the disease prevents further study of a potential renal phenotype . Unilateral renal aplasia may result from a developmental expression of ANLN . Additional evidence for the physiological significance of the loss-of-function variants in genes can be explored utilizing available variant databases . We searched possible loss-of-function variants ( frameshifts and nonsense variants ) for canine and human ANLN in our canine variant database and in public Genome Aggregation Database ( gnomAD ) [22] . No loss-of-function variants were found in 136 unaffected dogs while the exploration of the GnomAD database revealed four heterozygous frameshifts and fourteen heterozygous nonsense variants with very low allele frequencies and most being singletons . It therefore appears that the loss-of-function variants in ANLN are extremely rare across species , which supports the vital role of the gene for survival and is in agreement with the observed lethal disease in the affected Dalmatians . In summary , our study reveals a novel lethal pulmonary disease association with the ANLN gene and suggests that abnormal cytoskeletal dynamics and epithelial regeneration due to lack of functional ANLN result in the hyper- and metaplastic bronchiolar epithelium that predisposes the affected dogs to ARDS . A genetic test can be established to facilitate veterinary diagnostics and to eradicate the detrimental condition in the affected breed .
The experiments performed on dogs were approved by the “Animal Ethics Committee at the State Provincial Office of Southern Finland” ( permits: ESAVI/6054/04 . 10 . 03/2012 and ESAVI/7482/04 . 10 . 07/2015 , expire date 17 . 10 . 2020 ) and by the “University of Helsinki Viikki Campus Research Ethics Committee” ( Statement 4/2014 ) . Altogether 188 Dalmatian dogs and 31 Pointers were included in the study . Samples were obtained from seven affected Dalmatian dogs from four litters presented to the Small Animal Hospital of Helsinki University as described previously [10] . The mean age at the onset of illness in the puppies included in this study was seven months ( range 5–10 months ) . The mean duration of illness varied from one to six weeks , with a mean of three weeks . Four of the puppies were male , three female . DNA was extracted either from formalin fixed paraffin embedded ( FFPE ) tissue samples ( four dogs ) , from bronchoalveolar lavage samples ( two dogs ) or from whole EDTA -blood of unaffected Dalmatian dogs in Finland . All dogs in this study were privately owned pets that were examined with the owners’ consent . Genomic DNA from the FFPE and EDTA samples was extracted using the semi-automated Chemagen extraction robot ( PerkinElmer Chemagen Technologie GmbH , Germany ) . DNA from the BAL samples were extracted using QIAamp DNA Micro Kit ( Qiagen , Germany ) . DNA concentration was determined either with the NanoDrop ND-1000 UV/Vis Spectrophotometer ( Thermo Fisher Scientific Inc . , USA ) or Qubit 3 . 0 Fluorometer ( Thermo Fisher Scientific Inc . , USA ) . Genome-wide SNP genotyping of two affected , one healthy sibling , one obligate ( parent ) and one possible carrier ( grandparent ) and six healthy control Dalmatian dogs was performed at the GeneSeek facility ( Neogen Corporation , USA ) using Illumina’s CanineHD BedChips containing 173 , 662 validated SNPs . Genotypes were stored in BC/Gene database version 3 . 5 ( BC/Platforms ) . The PLINK v 1 . 07 software was used to search for segments of extended homozygosity in the two affected dogs as described previously [23 , 24] . Genotype data was filtered using a SNP call rate of > 95% , an array call rate > 95% and minor allele frequency of > 0 . 05 . The genotype data is available for further use upon request . We performed WGS of one affected Dalmatian dog and used 136 other dog genomes ( 48 whole exome sequences and 88 whole genome sequences ) available as controls ( S2 Table ) . A fragment library was prepared with a 290 bp insert size and collected to a single lane of Illumina HiSeq2000 paired-end reads ( 2 x 100 bp ) . The reads were processed using speedseq align module available in SpeedSeq suite to produce a duplicate-marked , sorted and indexed BAM file . The Genome Analysis Tool kit ( version = 3 . 3 . 0-g37228af ) was used to perform realignment around potential indel sites and base quality score recalibration using the known SNP variation available at the Broad Institute ( https://www . broadinstitute . org/ftp/pub/vgb/dog/trackHub/canFam3/variation ) . Dual algorithms , Samtools mpileup ( version samtools-1 . 2 ) and GATK haplotype caller module were used to detect variants and the variants from both algorithms were merged into variant call format ( VCFv4 . 1 ) . In summary , 98 . 37% of the reads from Dalmatian dog were mapped to the reference genome yielding a genome-wide mean coverage of 16X . We identified 1 , 475 , 318 indels and 4 , 804 , 627 SNPs of which 39 . 91% of the variants were known and 60 . 09% were novel compared to SNPs from Axelsson et al . , Lindblad-Toh et al . and Vaysse et al . and dbSNP build 131 [25–27] . Annovar and SnpEff tools were used to annotate the variants to Ensembl , NCBI and Broad annotation databases to predict the functional effects of the variants . Canine genome build CanFam 3 . 1 was used as a reference sequence . We used PCR and Sanger sequencing to perform targeted genotyping for selected variants in the candidate gene . PCR primers were designed with Primer 3 [28] to assess the prevalence of the mutation by Sanger sequencing in a cohort of Dalmatian dogs . We performed a standard PCR , including 0 . 5 U Biotools DNA Polymerase ( Biotools , Madrid , Spain ) , 2 . 0 mM MgCl2 ( Biotools , Madrid , Spain ) , 200 μM dNTPs ( Finnzymes , Espoo , Finland ) , 1 x Biotools PCR Buffer ( Biotools , Madrid , Spain ) , 0 . 5 μM forward primers and 0 . 5 μM reverse primers ( S3 Table ) . All primers were custom ordered from Sigma Aldrich ( St . Louis , MO , USA ) . Reaction mixtures were subjected to the thermal cycling program of 95°C for 10 min , 35 cycles of 95°C for 30 s , 30 s 57°C , 72°C for 40 s and final elongation state of 72°C for 10 min . Genomic PCR products were sequenced using a 3730xl DNA Analyzer ( Applied Biosystems , Foster City California , USA ) in the core facility , Institute for Molecular Medicine Finland ( FIMM , Technology Centre , University of Helsinki , Helsinki , Finland ) . We analyzed the sequence data with Sequencher 5 . 3 software ( Gene Codes Corp , Ann Arbor , MI , USA ) . The UniProt database ( http://www . uniprot . org ) and SMART tool ( http://smart . embl-heidelberg . de ) were used to confirm the protein domain structure of ANLN [29–30] . The ANLN sequence alignment and cross-species conservation was analyzed with ClustalW2 algorithm ( http://www . ebi . ac . uk/Tools/clustalw2/ ) . All numbering with the ANLN gene correspond to the accessions ENSCAFG00000003243 ( gene ) and ENSCAFT00000005209 ( protein ) . Archived paraffin blocks of autopsy tissue samples from four affected Dalmatian dogs were available for immunohistochemical staining . Autopsy lung samples from a 5-month-old , male Dalmatian , euthanized due to epilepsy and without histopathological changes in the lungs and other internal organs was used as healthy controls . Organs available for assessment of the ANLN expression in the normal dog included lung , kidney , smooth and cross-striated muscle , heart , choroid plexus , testis , liver , spleen , pancreas and lymph node . Lung samples from a 5-year-old Chihuahua male , euthanized due to ARDS with histopathologically confirmed AIP and DAD , were used as comparison of PCII expression pattern in wild-type dogs . Paraffin blocks were sectioned at 4 μm thickness and deparaffinized , antigens were retrieved with 0 . 01M citrate buffer at pH 6 and heat for 20 minutes at 99°C . Overnight- incubation was used for the primary antibody ( rabbit-polyclonal anti-Anillin Antibody aa1074-1124 , LS-C288200 ( LifeSpan BioSciences , Inc . , USA ) . The sections were stained according to the UltraVision Detection System HRP/DAB kit ( Thermo Fisher Scientific Inc . , USA ) . Separate tissue sections from all of the dogs were also stained with hematoxylin and eosin ( HE ) . The hematoxylin-eosin stained histological slides of kidneys from seven affected Dalmatians , including two adults , were histologically re-evaluated for a renal phenotype . Paraffin blocks from four puppies were available for further studies of the kidney and glomerular collagen was highlighted by Masson-trichrome ( MTC ) staining and basal membranes by periodic acid-Schiff ( PAS ) staining in order to reveal even subtle fibrosis . | Acute respiratory distress syndrome ( ARDS ) is characterized by life-threatening impairment of pulmonary gas exchange and leads to substantial mortality in man . Spontaneous ARDS has also been described in dogs including a familial fatal ARDS-like syndrome in young Dalmatian dogs . The main clinical signs include progressive tachypnea and dyspnea leading to a severe respiratory distress and euthanasia . The prominent clinicopathological findings involve pulmonary lesions , although some affected puppies also presented with renal aplasia and hydrocephalus . This study finds the genetic cause of the disease by identifying a recessive nonsense variant in the ANLN gene . The ANLN gene encodes an anillin actin binding protein which has an important role in the integrity of the epithelial cell organization . The functional defect of ANLN due to early truncation and absence from the bronchiolar epithelium is consistent with the observed histopathology with hyper- and metaplasia of the bronchiolar epithelium and clinical ARDS . Our study reveals a novel pulmonary disease association for ANLN , provides new insights to pathophysiology and has enabled the development of a genetic test for breeding purposes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"respiratory",
"failure",
"membrane",
"staining",
"vertebrates",
"dogs",
"animals",
"mammals",
"lungs",
"pulmonology",
"respiratory",
"system",
"genome",
"analysis",
"mammalian",
"genomics",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"genomics",
"biological",
"databases",
"biological",
"tissue",
"homozygosity",
"critical",
"care",
"and",
"emergency",
"medicine",
"animal",
"genomics",
"acute",
"respiratory",
"distress",
"syndrome",
"anatomy",
"heredity",
"database",
"and",
"informatics",
"methods",
"genetics",
"epithelium",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"amniotes",
"genomic",
"databases",
"organisms"
]
| 2017 | ANLN truncation causes a familial fatal acute respiratory distress syndrome in Dalmatian dogs |
Toxoplasma gondii resides in an intracellular compartment ( parasitophorous vacuole ) that excludes transmembrane molecules required for endosome - lysosome recruitment . Thus , the parasite survives by avoiding lysosomal degradation . However , autophagy can re-route the parasitophorous vacuole to the lysosomes and cause parasite killing . This raises the possibility that T . gondii may deploy a strategy to prevent autophagic targeting to maintain the non-fusogenic nature of the vacuole . We report that T . gondii activated EGFR in endothelial cells , retinal pigment epithelial cells and microglia . Blockade of EGFR or its downstream molecule , Akt , caused targeting of the parasite by LC3+ structures , vacuole-lysosomal fusion , lysosomal degradation and killing of the parasite that were dependent on the autophagy proteins Atg7 and Beclin 1 . Disassembly of GPCR or inhibition of metalloproteinases did not prevent EGFR-Akt activation . T . gondii micronemal proteins ( MICs ) containing EGF domains ( EGF-MICs; MIC3 and MIC6 ) appeared to promote EGFR activation . Parasites defective in EGF-MICs ( MIC1 ko , deficient in MIC1 and secretion of MIC6; MIC3 ko , deficient in MIC3; and MIC1-3 ko , deficient in MIC1 , MIC3 and secretion of MIC6 ) caused impaired EGFR-Akt activation and recombinant EGF-MICs ( MIC3 and MIC6 ) caused EGFR-Akt activation . In cells treated with autophagy stimulators ( CD154 , rapamycin ) EGFR signaling inhibited LC3 accumulation around the parasite . Moreover , increased LC3 accumulation and parasite killing were noted in CD154-activated cells infected with MIC1-3 ko parasites . Finally , recombinant MIC3 and MIC6 inhibited parasite killing triggered by CD154 particularly against MIC1-3 ko parasites . Thus , our findings identified EGFR activation as a strategy used by T . gondii to maintain the non-fusogenic nature of the parasitophorous vacuole and suggest that EGF-MICs have a novel role in affecting signaling in host cells to promote parasite survival .
Toxoplasma gondii is an obligate intracellular protozoan parasite that infects around a third of the human population worldwide . T . gondii is of clinical importance because it causes encephalitis in immunocompromised individuals and retino-choroiditis in immunocompetent and immunosuppressed patients . T . gondii can also cause congenital infection that may result in cerebral and ocular disease . Tachyzoites of T . gondii infect virtually any nucleated cell through active invasion . This process is dependent on the parasite actin-myosin motor and sequential secretion of proteins from micronemes and rhoptries , specialized organelles present in the apical end of the parasite [1] . Once secreted , T . gondii micronemal proteins ( MICs ) are expressed at the parasite surface membrane and they interact with host cell receptors [2] . MICs contain adhesive domains such as type I thrombospondin repeats , apple domains , EGF repeats and integrin A domains [3] , [4] . The connection between transmembrane MICs to the actin-myosin motor ( glideosome ) of the parasite together with the binding of host cell receptors by MICs is considered to enable the organism to penetrate host cells [5] , [6] . Following the release of MICs , rhoptries secrete rhoptry neck proteins ( RONs ) that are critical for the formation of a structure called the moving junction ( MJ ) [7] , [8] . The MJ anchors the parasite to the host cell while the parasite penetrates it . The MJ is also believed to function as a sieve that excludes host type I transmembrane proteins from entering the PV membrane ( PVM ) [8] , [9] . The end result is the formation of a parasitophorous vacuole that is devoid of host proteins required for recruitment of endosomes and lysosomes [10] . T . gondii cannot withstand the lysosomal environment . Thus , the non-fusogenic nature of the PV is critical since it allows the parasite to survive and replicate . The immune system can deprive the parasite from this niche by disrupting the PVM through the effects of IFN-γ/Immunity related GTPases ( IRG ) [11] , [12] and by making the PV fusogenic through the effects of CD40 ligation [13]–[15] . CD40 re-routes the PV to the lysosomes through the autophagy machinery [13]–[15] . Autophagy is a conserved cellular mechanism of lysosomal degradation . During autophagy , portions of the cytosol or organelles are encircled by an isolation membrane [16] . The expansion of the isolation membrane results in the formation of a double membrane structure called autophagosome that delivers its contents to the lysosomes for degradation [16] . Autophagy is recognized as a mechanism stimulated by innate and adaptive immune mechanisms to degrade numerous intracellular pathogens [17] . However , various bacteria and viruses have evolved mechanisms to prevent autophagic degradation by targeting autophagy proteins to avoid recognition by the autophagy machinery or prevent the initiation and maturation of autophagosomes [18]–[24] . Much less is known regarding whether pathogens manipulate signaling cascades that regulate autophagy to prevent their degradation . HIV-1 envelope can activate the negative regulator of autophagy mTOR and it has been proposed that this would prevent lysosomal degradation of virions [25] . Bioinformatic analysis of human THP-1 cells infected with Mycobacterium tuberculosis suggested that the pathogen activates host signaling cascades that impair autophagy [26] . The highly successful nature of T . gondii as a pathogen together with evidence that the autophagy pathway can trigger lysosomal killing of the pathogen raise the possibility that T . gondii prevents autophagic targeting of the PV to maintain the non-fusogenic nature of the PV . Moreover , approximately 25–35% of various CD40+ cells subjected to CD40 ligation are unable to kill T . gondii further suggesting that the parasite may utilize mechanism ( s ) to prevent induction of autophagic killing . Here we report that maintenance of the non-fusogenic nature of the PV requires T . gondii-induced activation of EGFR-Akt , a signaling cascade that prevents autophagy protein-dependent vacuole-lysosomal fusion , lysosomal degradation and killing of the parasite . Blockade of EGFR-Akt may prove of therapeutic benefit for toxoplasmosis since it is sufficient to induce killing of the parasite without the need for immune-induced activation of host cells .
We determined whether Akt is quickly activated by T . gondii during infection of various non-hematopoietic cells . Activation of Akt is a multistep process where phosphorylation of Serine 473 results in full activation of the molecule [27] . Primary human brain microvascular endothelial cells ( HBMEC ) were infected with either type I ( RH ) or type II ( ME49 ) strains of T . gondii under conditions that caused synchronized infection . T . gondii infection resulted in an enhanced phosphorylation of Akt Serine 473 as assessed by immunoblot ( Figure 1A ) . Similar results were obtained with a mouse endothelial cell line mHEVc ( Figure 1B ) . T . gondii also caused Akt phosphorylation in a human retinal pigment epithelial ( RPE ) cell line , an effect that decreased at later time points post-infection ( Figure 1C ) . We assessed whether viable parasites are required to induce activation of Akt . HBMEC were challenged with live or killed parasites followed by determination of Akt activation . Viable but not killed tachyzoites induced Akt phosphorylation ( Figure 1D ) . Activation of phosphatidylinositol 3-kinase ( PI3K ) with resulting production of phosphatidylinositol 3 , 4 , 5 trisphosphate ( PIP3 ) production is a major trigger of Akt activation [28] . The amino-terminal pleckstrin homology ( PH ) domain of Akt mediates recruitment of this molecule to plasma membrane containing increased PI ( 3 , 4 , 5 ) P3 or PI ( 3 , 4 ) P2 [29] . Indeed , the PH domain of Akt fused to GFP ( PH-Akt-GFP ) has been used as a probe to examine sites of PIP3 accumulation [30] . HBMEC were transiently transfected with a plasmid encoding PH-Akt-GFP followed by challenge with RH T . gondii that express cytoplasmic RFP ( T . gondii-RFP ) . T . gondii-infected cells exhibited accumulation of PH-Akt-GFP around the parasite ( Figure 1E ) . To examine the role of PI3K in this process , HBMEC were incubated with or without LY294002 , a specific PI3K inhibitor , followed by challenge with the parasite . LY294002 did not affect the percentage of infected cells ( not shown ) . Accumulation of PH-Akt-GFP around T . gondii was ablated by LY294002 ( p<0 . 01 ) ( Figure 1E ) . Moreover , incubation with LY294002 impaired the upregulation of Akt phosphorylation induced by T . gondii , especially in the earlier time points post-infection ( Figure 1F ) . Similarly , Akt phosphorylation during T . gondii infection was impaired in HBMEC transfected with siRNA against the PI3K catalytic subunit p110α ( Figure 1G ) . Taken together , these findings indicate that T . gondii induces rapid Akt activation in non-hematopoietic cells in a manner that is dependent on PI3K . We performed studies to investigate whether blockade of Akt signaling promotes killing of T . gondii . HBMEC were incubated with or without Akt inhibitor IV followed by challenge with T . gondii . The percentage of infected cells at 2 hours and 24 hours post-challenge were determined . Akt inhibitor IV did not impair the percentage of infected cells at 2 h ( Figure 2A ) . However , treatment with Akt inhibitor IV markedly reduced the percentage of infected cells at 24 h ( p<0 . 01 ) ( Figure 2A ) . Changes in the percentage of infected cells were not due to preferential cell loss in Akt inhibitor IV-treated cells since cell densities as determined with an eyepiece grid were similar in all experimental groups and inhibition of Akt did not induce a detectable increase in apoptosis of T . gondii-infected cells ( not shown ) . Akt inhibitor IV not only induced a significant decrease in the numbers of parasites per 100 HBMEC at 24 h but it also caused a profound reduction in the numbers of T . gondii-containing vacuoles per 100 HBMEC ( p<0 . 01 ) ( Figure 2A , Figure S1A ) . Similar results were obtained with mouse endothelial cells ( mHEVc; Figure 2A , Figure S1A ) and human RPE cells ( p<0 . 01 ) ( Figure 2A , Figure S1A ) . Not only pharmacologic inhibition of Akt but also Akt knockdown in HBMEC reduced the parasite load and the number of T . gondii-containing vacuoles ( p<0 . 01 ) ( Figure 2B , Figure S1A ) . The vacuoles that persisted after Akt knockdown had similar numbers of parasites as those from control cells ( Figure S1B ) . These results indicate that blockade of Akt caused parasite killing . T . gondii infection causes Akt activation in macrophages [31] . Similar to endothelial and epithelial cells , treatment with Akt inhibitor IV caused anti-T . gondii activity in the mouse macrophage line RAW 264 . 7 and in mouse microglia line BV-2 ( p<0 . 01 ) ( Figure 2C and not shown ) . These findings revealed an important role of Akt activation in promoting survival of T . gondii within host cells . T . gondii survives within mammalian cells by avoiding delivery of the lysosomal contents into the parasitophorous vacuole [32]–[34] . Akt is a negative regulator of autophagy [35] , a cellular mechanism that results in lysosomal degradation and killing of T . gondii [13]–[15] . First , we examined T . gondii-infected cells after Akt inhibition to determine the distribution of LC3 , a protein associated with the autophagosome membrane . mHEVc-LC3-EGFP cells were treated with or without Akt inhibitor IV and challenged with T . gondii-RFP . Akt inhibitor IV led to significant accumulation of LC3 around the parasite ( p<0 . 01 ) ( Figure 2D ) . Electron microscopy studies were performed since a double membrane isolation membrane that encircles portions of cytoplasm or organelles is formed during autophagy [16] . Indeed , a double membrane structure was noted around the parasitophorous vacuole membrane in HBMEC treated with Akt inhibitor IV ( Figure 2E ) . Next , we examined the effects of Akt inhibition on the distribution of the late endosomal/lysosomal molecule LAMP-1 . Endothelial cells were incubated with or without Akt inhibitor IV , challenged with T . gondii-YFP followed by staining with anti-LAMP-1 mAb . Treatment with Akt inhibitor IV resulted in a remarkable increase in the percentage of parasites surrounded by LAMP-1 ( p<0 . 01 ) ( Figure 2F ) . To explore whether the killing of T . gondii during inhibition of Akt is dependent on the autophagy machinery , we examined the effects of knockdown of the autophagy proteins Beclin 1 or Atg7 on T . gondii survival . Transfection with Beclin 1 siRNA or Atg7 siRNA effectively diminished expression of Beclin 1 or Atg7 respectively ( Figure 2G , H ) . Endothelial cells transfected with Beclin1 siRNA or Atg7 siRNA were incubated with or without Akt inhibitor and challenged with T . gondii . Cells transfected with Beclin1 siRNA ( Figure 2G ) or Atg7 siRNA ( Figure 2H ) were unable to control the parasite in the presence of the Akt inhibitor IV . Since autophagosomes deliver their contents to lysosomes for degradation , we examined the role of lysosomal degradation in killing of T . gondii utilizing the lysosomal protease inhibitors leupeptin and pepstatin . mHEVc and RPE cells were treated with or without Akt inhibitor IV and infected with T . gondii . 1 h post infection cells were treated with or without leupeptin plus pepstatin . Lysosomal protease inhibitors impaired the anti-T . gondii activity induced by Akt inhibition ( p<0 . 05 ) ( Figure 2I and not shown ) . Finally , the anti-T . gondii activity induced by Akt inhibitor IV in mouse microglia and human RPE cells was impaired by 3-methyl adenine , an inhibitor of autophagy ( p<0 . 05 ) ( Figure 2J and not shown ) . Taken together , these results indicate that T . gondii-induced Akt activation is critical to promote parasite survival because it prevents killing of T . gondii dependent on the autophagy pathway and lysosomal protease activity . Akt activation classically occurs downstream of cell membrane receptors that include growth factor receptors , G protein-coupled receptor ( GPCR ) and TLR [36] . To examine the role of GPCR in Akt activation in non-hematopoietic cells , HBMEC were incubated with or without Pertussis toxin ( PTx ) , an inhibitor of GPCR signaling , followed by challenge with T . gondii tachyzoites . PTx did not affect the initial percentage of infected cells ( data not shown ) . Incubation with PTx decreased basal Akt phosphorylation . However , PTx did not prevent the increased Akt phosphorylation induced by T . gondii ( Figure 3A ) indicating that T . gondii can activate Akt independently of GPCR signaling . In contrast , PTx inhibited Akt activation induced by lysophosphatidic acid ( LPA ) , a GPCR ligand [37] ( Figure 3A ) . To examine the potential role of TLR signaling in Akt , MyD88 was knocked-down in HBMEC using siRNA . Knockdown of MyD88 did not affect T . gondii-induced Akt activation ( Figure 3B ) . In contrast , as assessed by FACS , the ICAM-1 upregulation induced by LPS ( 1 µg/ml ) in HBMEC was inhibited in cells transfected with MyD88 siRNA compared to those transfected with control siRNA ( cMFI: Control siRNA = 10 , 682±1 , 053; MyD88 siRNA = 3 , 250±527; p<0 . 05 ) . These studies indicate that GPCR and TLR are unlikely to play a major role in Akt phosphorylation induced by T . gondii in non-hematopoietic cells . Relevant to the possibility of activation of growth factor receptors during T . gondii-host cell interaction is the fact that host cell invasion by T . gondii requires the secretion of parasite micronemal proteins ( MICs ) with the potential to activate such receptors [38] . MICs exist as multiprotein complexes , the most important being MIC1/4/6 , MIC3/8 , MIC2/M2AP , and a complex of the microneme protein TgAMA1 with rhoptry neck proteins RON2/RON4/RON6/RON8 [39]–[41] . MIC3 , MIC6 and MIC8 have multiple domains with homology to EGF [42] and are therefore termed EGF-MICs . As an initial experiment , we examined whether T . gondii induces autophosphorylation at 2 major tyrosine residues of EGFR ( 1068 and 1148 ) . HBMEC were incubated with RH T . gondii tachyzoites followed by determination of EGFR phosphorylation by immunoblot . T . gondii induced activation of EGFR , as indicated by phosphorylation of tyrosine residue 1068 ( Figure 4A ) . Moreover , the parasite caused phosphorylation of tyrosine residue 1148 , a site that appears to be phosphorylated only by ligand binding to EGFR [43] ( Figure 4A ) . Similar results were found using the ME49 strain of T . gondii ( not shown ) . Immunoblot analysis revealed that EGFR activation occurred in HBMEC upon challenge with viable but not killed parasites ( Figure 4B ) . EGFR autophosphorylation was not only observed in endothelial cells but also in human RPE cells and mouse microglia incubated with T . gondii ( Figure 4C , 4D ) . Thus , T . gondii causes EGFR activation in various mammalian cells . Next , we examined whether EGFR signaling is involved in activation of Akt triggered by T . gondii . Endothelial cells were transiently transfected with a plasmid that encodes either control siRNA or EGFR siRNA followed by challenge with T . gondii . The efficiency of EGFR knockdown was confirmed by immunoblot ( Figure 5A ) . EGFR knockdown ablated the ability of T . gondii to induce activation of Akt at all time points tested ( Figure 5A ) . Next , we explored the role of EGFR signaling on Akt activation in professional phagocytes . Mouse microglia were treated with vehicle or AG1478 , a pharmacological inhibitor of EGFR kinase activity , followed by challenge with T . gondii . Inhibition of EGFR kinase activity ablated parasite-induced Akt activation in mouse microglia ( Figure 5B ) . We assessed whether EGFR activation affects T . gondii survival within host cells . HBMEC were treated with vehicle or AG1478 followed by challenge with T . gondii . While AG1478 did not affect the percentage of infected cells at 2 h , AG1478 caused a marked reduction in the percentage of infected cells 24 h post-challenge ( p<0 . 05 ) ( Figure 6A ) . In addition , there was a significant reduction in the numbers of parasites per 100 endothelial cells ( p<0 . 01 ) ( Figure 6A ) . Similar results were obtained whether HBMEC or human retinal endothelial cells were infected with RH or ME49 strains of T . gondii ( not shown ) . The role of EGFR in affecting parasite survival was confirmed with a genetic approach since knockdown of EGFR in human RPE cells resulted in enhanced killing of T . gondii ( p<0 . 01 ) ( Figure 6B ) . Similar to the studies of blockade of Akt , inhibition of EGFR signaling not only reduced the percentages of infected cells but also caused a reduction in the numbers of vacuoles per 100 cells without affecting the numbers of parasites in the vacuoles that persisted after EGFR blockade ( not shown ) . The effects of EGFR signaling inhibition were not restricted to non-hematopoietic cells since mouse bone marrow-derived macrophages also acquired anti-T . gondii activity when treated with AG1478 ( p<0 . 05 ) ( Figure 6C ) . To further explore the role of EGFR in the survival of T . gondii , we took a reverse approach and infected parental CHO cells , known to be EGFR null [44] , and CHO cells expressing human EGFR ( CHO-EGFR ) . A reduction in the percentage of infected cells and a reduction in parasite load at 24 h were observed in parental CHO cells compared to CHO-EGFR cells ( p<0 . 05 ) ( Figure 6D ) . These findings revealed an important role of EGFR in promoting Akt activation and T . gondii survival within host cells . We investigated whether T . gondii killing induced by inhibition of EGFR is dependent on autophagy proteins . Knockdown of EGFR in mHEVc cells or treatment of these cells with AG1478 resulted in an enhanced accumulation of LC3 and LAMP-1 around the parasite ( p<0 . 05 ) ( Figure 6E and 6F ) . Moreover , silencing of Beclin 1 or Atg7 prevented induction of anti-T . gondii activity in endothelial cells subjected to EGFR knock-down or treated with AG1478 ( p<0 . 01 ) ( Figure 6G and 6H ) . Taken together , activation of EGFR signaling promoted survival of T . gondii within host cells by inhibiting autophagy protein-dependent killing of the parasite . EGFR ligands exist as precursors transmembrane proteins that are shed from the plasma membrane by members of the ADAM ( a disintegrin and metalloprotease ) family of zinc-dependent metalloproteases [45] . This results in an autocrine or paracrine EGFR activation , a phenomenon that explains how proteins such GPCR activate EGFR [45] . We explored whether EGFR activation triggered by T . gondii could be due to this mechanism of autocrine/paracrine signaling . HBMEC were treated with GM6001 , a broad spectrum ADAM inhibitor , followed by challenge with T . gondii . GM6001 did not affect the percentage of infected cells ( data not shown ) and did not prevent the ability of T . gondii to induce EGFR activation ( Figure 7A ) . Moreover , EGFR phosphorylation after T . gondii infection took place despite incubation with PTx ( Figure 7B ) . These findings suggest that ADAM- and GPCR-dependent EGFR activation do not play a major role in EGFR phosphorylation induced by T . gondii . As stated above , MIC3 , MIC6 , MIC8 have multiple domains with homology to EGF [42] . MIC7 and MIC9 also express EGF-like domains but these MICs have poor or no expression in tachyzoites [42] . We examined the effect of deficiency of MICs on the ability to induce activation of EGFR and Akt . HBMEC were infected with wild type ( WT ) , MIC1 ko ( lacks MIC1 , resulting in deficient secretion of MIC6 [46] ) , MIC3 ko ( lacks MIC3 ) , MIC1-3 ko ( lacks MIC6 secretion and MIC3 ) parasites followed by determination of EGFR and Akt activation . These MIC ko parasites still express MIC8 ( MIC8 deficiency results in parasites that are unable to infect mammalian cells ) . The multiplicity of infection was adjusted so that the initial percentages of infected HBMEC were similar for all strains of the parasite ( Figure 8A ) . Compared to WT T . gondii , MIC1 ko and MIC3 ko parasites caused a partial reduction in EGFR and Akt phosphorylation ( p<0 . 05 ) ( Figure 8B , 8C ) . MIC1-3 ko parasites caused further decrease in EGFR and Akt phosphorylation compared to MIC1 ko and MIC3 ko parasites ( p<0 . 05 ) ( Figure 8B , 8C ) . However , even in cells infected with MIC1-3 ko parasites the reduction in EGFR and Akt phosphorylation was not complete . MIC1-3 ko parasites still express MIC8 , a molecule that has EGF-like domains . We used conditional MIC8 knockout T . gondii previously generated using a tetracycline-inducible system to explore the potential role of MIC8 in signal activation [47] . Incubation of these parasites with anhydrotetracycline ( ATc ) results in almost complete ablation of MIC8 [47] . Parasites previously grown in the absence or presence of ATc were incubated with HBMEC . We could not detect an appreciable decrease in Akt phosphorylation in cells exposed to MIC8 deficient parasites ( Figure S2 ) . To further explore the role of MICs in the activation of EGFR and Akt , HBMEC were incubated with Pichia pastoris-derived MIC3 . Although the EGF-like domains alone do not appear to promote the adhesion of MIC3 to mammalian cells [48] , it was still possible that MIC3 could cause EGFR and Akt activation . Indeed , compared to recombinant MIC4 ( a control that does not express EGF-like domains ) incubation with recombinant MIC3 caused enhanced phosphorylation of EGFR and Akt in HBMEC ( Figure 8D , 8E ) . Moreover , incubation with E . coli-derived MIC6 but not M2AP caused EGFR-Akt phosphorylation ( Figure 8D , 8E ) . This response was unlikely to be mediated by LPS since M2AP and MIC6 preparations had similar concentrations of LPS ( 12 ng/ml and 12 . 4 ng/ml respectively ) . In addition , LPS at concentrations between 10–1 , 000 ng/ml failed to induce EGFR phosphorylation in HBMEC ( not shown ) . Taken together , EGF-MICs ( MIC3 and MIC6 ) can induce EGFR-Akt activation and parasites deficient on these MICs have diminished capacity to activate EGFR and Akt . Cells stimulated with CD154 ( CD40 ligand ) exhibit accumulation of LC3 around T . gondii and killing that is dependent on autophagy proteins [13]–[15] . We examined whether targeting of the parasite by LC3+ structures in CD154-treated cells can be affected by EGFR signaling . Endothelial cells were treated with or without CD154 followed by challenge with T . gondii in the presence or absence of EGF . EGF did not affect the initial percentage of infected cells ( not shown ) . As previously reported [15] , CD154 caused accumulation of LC3 around T . gondii ( Figure 9A ) . Targeting of parasites by LC3+ structures was inhibited in cells that were exposed to EGF ( p<0 . 05 ) ( Figure 9A ) , The effect of EGF was specific since addition of AG1478 to cells treated with EGF restored LC3 accumulation around T . gondii ( Figure 9A ) . Similar results were obtained using rapamycin , a well-described stimulator of autophagy ( Figure 9B ) . Next , we explored the role of MICs on the distribution of LC3+ structures in endothelial cells treated with CD154 . Endothelial cells were treated with or without CD154 followed by challenge with WT , MIC1 ko , MIC3 ko , MIC1-3 ko and their respective complemented parasites . Infection with MIC1 ko , MIC3 ko or MIC1-3 ko parasites induces a partial decrease in EGFR-Akt activation ( see Figures 8B , 8C ) . Indeed , in control endothelial cells ( no CD154 treatment ) there were no differences in the low level LC3 accumulation around the parasites ( Figure 9C ) . After treatment with CD154 , enhanced accumulation of LC3 around the parasites was similar in endothelial cells infected with WT , MIC1 ko or MIC3 ko parasites ( Figure 9C ) . In contrast , cells infected with MIC1-3 ko parasites ( the strain that was the weakest inducer of EGFR-Akt activation ) exhibited a significant further increase in LC3 accumulation ( p<0 . 05 ) ( Figure 9C ) . These results were specific because the phenotype was lost in the complemented strain of T . gondii ( MIC1-3 ko+MIC1-3 ) ( Figure 9C ) . Examination of the parasite load revealed that the loads of MIC1 ko , MIC3 ko and MIC1-3 ko parasites were not significantly different from those of WT parasites in control endothelial cells ( no CD154 treatment ) ( Figure 9D ) . When cells were treated with CD154 , MIC1-3 ko T . gondii displayed increased susceptibility to CD154-induced anti-T . gondii activity ( p<0 . 05 ) ( Figure 9D ) . Similar to the studies of LC3 expression , the phenotype of MIC1-3 ko parasites was lost in the complemented strain ( MIC1-3 ko+MIC1-3 ) ( Figure 9D ) . Next , we examined whether increased killing of MIC1-3 ko parasites was observed in cells treated with another autophagy inducer ( rapamycin ) or in cells treated with IFN-γ , a cytokine that triggers anti-T . gondii activity independently of autophagic degradation [13]–[15] . Similar to CD154-stimulated cells , MIC1-3 ko parasites were more susceptible to rapamycin-induced killing ( p<0 . 05 ) ( Figure 9E ) . Moreover , in contrast to the results obtained with CD154-stimualtion , anti-T . gondii activity induced by IFN-γ/TNF-α was similar in all parasite strains tested including MIC1-3 ko T . gondii ( Figure 9F ) . Finally , we explored the effects of recombinant MICs on CD154-induced killing of MIC1-3 ko T . gondii . In initial experiments , recombinant MICs did not affect the load of T . gondii in non-activated ( control ) endothelial cells or cells treated with IFN-γ/TNF-α ( not shown ) . Next , control or CD154-activated endothelial cells were challenged with WT or MIC1-3 ko parasites in the presence of absence of recombinant MICs . Whereas treatment of endothelial cells with MIC4 and M2AP did not affect the load of WT or MIC1-3 ko parasites in CD154-activated cells , treatment with MIC3 or MIC6 inhibited CD154-induced T . gondii activity ( p<0 . 05 ) ( Figure 9G ) . Moreover , the phenotype of MIC1-3 ko parasites of increased susceptibility to CD154-mediated anti-T . gondii activity was lost in the presence of either MIC3 or MIC6 since the loads of WT and MIC1-3 ko parasites were no longer different in cells treated with these EGF-MICs ( Figure 9G ) . Taken together , our findings indicate that EGFR , MIC3 and MIC6 negatively regulate autophagic killing of T . gondii .
Avoidance of lysosomal degradation is pivotal for the survival of numerous intracellular pathogens including T . gondii . Our studies indicate that , in addition to exclusion of type I transmembrane proteins from the PVM , T . gondii also activates EGFR-Akt signaling in the host cell to prevent targeting of the parasite by LC3+ structures and pathogen killing that is dependent on autophagy proteins and lysosomal protease activity . Thus , these studies identified EGFR-Akt signaling as a pathway critical for pathogen survival . In addition , they suggest that EGF-MICs may be involved in pathogen virulence not only by allowing parasite invasion of host cells but also by activating host cell signaling that counter-regulates autophagy . Various bacteria and viruses encode virulence factors that impair the function of autophagy proteins and as a result , avoid their degradation via the autophagy pathway [18]–[24] . It has been suggested that HIV-1 and M . tuberculosis may prevent autophagic degradation by affecting signaling cascades that regulate the autophagy pathway [25] , [26] . Our studies indicate that indeed a pathogen can act at the level of a regulatory pathway to avoid its degradation by the autophagy machinery . Relevant to our findings is the report that HIV-1 tat impairs autophagy by stimulating counter-regulatory cascades ( Akt and STAT3 ) , although these studies did not examine whether these pathways would prevent lysosomal degradation of the virions [49] . Our studies indicate that T . gondii-induced EGFR activation is a major event upstream of Akt phosphorylation in endothelial and RPE cells , a finding consistent with the important role of EGFR and other growth factor receptors as activators of Akt signaling [36] , [50] . PI3K is a classical link between growth factor receptors and Akt activation . However , in contrast inhibition of EGFR signaling , the effect of PI3K inhibition on Akt activation appeared to be more transient . These findings may be explained by the fact that , besides PI3K , there are additional activators of Akt that might be engaged by growth factor receptors [51] . T . gondii has been reported to activate Akt in macrophages , a phenomenon that was inhibited by PTx [31] . Our studies indicate that EGFR also contributes to Akt activation in macrophages/microglia since the parasite caused EGFR autophosphorylation and inhibition of EGFR signaling impaired parasite-induced Akt activation . Moreover , not only activation of Akt but also activation of EGFR in endothelial cells , RPE cells and macrophages/microglia prevented killing of T . gondii dependent on autophagy proteins and lysosomal enzymes . The fact that Akt activation has been linked to inhibition of apoptosis of T . gondii-infected cells [31] raises the possibility that parasite-induced EGFR - Akt signaling may not only promote parasite survival by preserving the non-fusogenic nature of the PV but also by avoiding death of infected cells subjected to pro-apoptotic signals . While EGFR is a central mediator of Akt activation in the early stages after T . gondii , Akt phosphorylation has recently been reported at 24 h post-infection with the parasite [52] . This raises the possibility that T . gondii may also activate Akt through additional mechanisms besides parasite engagement of EGFR . Although T . gondii causes EGFR - Akt activation and these signaling molecules have been shown to inhibit autophagy [35] , [53] , [54] , T . gondii does not appear to prevent autophagosome formation in infected cells . Indeed , large LC3+ structures were readily detected within infected cells during early stages post-infection ( see Figure 2D ) , a finding previously reported in host cells at 24 h post-infection [55] . Moreover , there is no decrease in the levels of LC3 II ( the lipidated form of LC3 that associates with the autophagosome membrane ) during the early stages of infection ( Muniz-Feliciano and Subauste , unpublished observations ) . In fact , T . gondii has been reported to increase LC3 II levels and autophagosome formation in host cells at 24 h post-infection , presumably as an attempt to gain access to nutrients [55] . Our studies indicate that while global autophagy did not appear to be inhibited by T . gondii , engagement of EGFR impaired targeting of the PV by LC3+ structures . Future studies that identify how autophagosomes target the PV will likely shed light on the molecular mechanism by which EGFR - Akt diminish autophagic targeting of the parasite . Various pathogens can target EGFR . Pseudomonas aeruginosa and Helicobacter pylori can cause EGFR phosphorylation that is mediated by the release of membrane-bound EGF ligands and transactivation of EGFR [56] , [57] . Klebsiella pneumonia causes EGFR activation that appears to be dependent on bacterial capsule polysaccharide engagement of TLR4 and subsequent Src-dependent EGFR activation [58] . In addition , proteins from oncogenic viruses activate EGFR to mediate transformation [59] . Much less is known on whether microbial products can directly engage and activate EGFR . It has been suggested that H . influenza may activate EGFR through the presence of bacterial-derived molecules with EGF-like properties [60] . Uptake of Influenza A virus causes EGFR activation , a process that may be dependent on multivalent binding of hemagglutinin to sialic acids present on EGFR or ganglioside GM1 leading to aggregation of rafts , clustering of EGFR and its activation [61] . Our studies suggest that EGF-MICs play a role in mediating EGFR-Akt activation of host cells and prevention of parasite killing since: recombinant EGF-MICs ( MIC3 and MIC6 ) induce EGFR-Akt activation while MICs that lack EGF domains do not cause appreciable phosphorylation of EGFR and Akt; EGFR signaling inhibits LC3 accumulation around T . gondii; parasites deficient in 2 EGF-MICs ( MIC3 and MIC6: MIC1-3 ko parasites ) cause markedly impaired EGFR-Akt activation and exhibit increased encasing by LC3+ structures as well as killing in cells treated with autophagy stimulators; MIC3 and MIC6 impair parasite killing mediated by the autophagy pathway . It was interesting to note that MIC1-3 ko parasites are not targeted by LC3+ structures and are not more likely to be killed in unstimulated cells despite the markedly weakened EGFR-Akt signaling . MIC1-3 ko parasites only display increased susceptibility to autophagic targeting and killing when autophagy is stimulated by CD154 or rapamycin . Of relevance to our findings , other studies support the existence of signaling thresholds that need to be achieved in order for autophagy to take place [62] , [63] . For example , in Drosophila both the Ret-like receptor tyrosine kinase Stitcher ( Stit ) and insulin receptor ( InR ) are required for cell growth and proliferation through the PI3K-I/TORC1 pathway in the wing disc [63] . A decrease in either Stit or InR signaling diminishes TORC1 activity and suppresses growth [63] . However , this decrease in TORC1 activity is not sufficient to trigger autophagy in the wing [63] . Autophagy only takes place when both Stit and InR are impaired [63] . It was proposed that the simultaneous inactivation of Stit and InR reduces PI3K-I activity and TORC1 signaling below a critically low level at which autophagy in the wing can no longer be prevented [63] . Given that the EGFR-Akt pathway inhibits autophagy by regulating TORC1 activity , a similar phenomenon could be at play in the case of T . gondii infection . The reduction in EGFR-Akt observed in cells infected with MIC1 ko or MIC3 ko parasites does not translate in increased autophagic killing of these parasites either in unstimulated cells or in cells treated with stimulators of autophagy . The further reduction in EGFR-Akt signaling observed in cells infected with MIC1-3 ko may still be sufficient to prevent autophagic killing in unstimulated cells but results in enhanced killing in cells treated with autophagy stimulators . Finally , further inhibition of EGFR-Akt signaling ( by genetic or pharmacological approaches ) triggers autophagic targeting of T . gondii even in unstimulated cells . Thus , our studies suggest that the effects of MIC deficiency on the levels of EGFR-Akt activation likely explain the differences in outcome observed after infection . Taken together , in addition to being key for invasion of host cells , EGF-MICs ( MIC3 and MIC6 ) contribute to the induction of a signaling cascade within these cells that is required to avoid lysosomal degradation of the parasite . While MIC1-3 ko parasite exhibited a marked defect in EGFR-Akt activation in host cells , phosphorylation of these molecules still took place . Although we cannot rule out a role of MIC8 in activation of this cascade , it appears that the residual ability of MIC1-3 ko parasites to activate EGFR-Akt may not be explained by their expression of MIC8 ( an EGF-MIC ) . Conditional MIC8 ko parasites did not exhibit a noticeable defect in signal activation in host cells . These findings are likely explained by the fact that MIC8 ko parasites do not exhibit defects in attachment to host cells and they secrete MICs [47] . The presence of an additional mechanism of EGFR-Akt activation that normally cooperates with MIC-dependent EGFR signaling may explain why MIC1-3 ko T . gondii have residual capacity to activate the EGFR-Akt pathways . T . gondii is very successful as a pathogen and utilizes various strategies to manipulate host cell signaling to ensure its survival [64]–[67] . Here we report that the parasite activates EGFR - Akt to maintain the non-fusogenic nature of PV a process that appears to be dependent at least in part on EGF-MICs . These findings may be of therapeutic relevance since various inhibitors of EGFR are being used for treatment of cancer . The fact that EGFR inhibition induced parasite killing in cells not treated with immune activators , raises the possibility that this approach may be effective even in immunocompromised hosts .
Primary human brain microvascular endothelial cells ( HBMEC ) were obtained from ScienCell Research Laboratories ( Carlsbad , CA ) and cultured in fibronectin-coated tissue culture flasks and basal medium supplemented with Endothelial Cell Growth Supplement ( ECGS ) and 5% fetal bovine serum ( FBS ) all from ScienCell . The mouse high endothelial venule cell line ( mHEVc ) ( gift from Joan Cook-Mills , Northwestern University , Chicago , IL ) and mHEVc cells stably expressing LC3-EGFP construct ( mHEVc-LC3-EGFP ) or hmCD40 plus LC3-EGFP ( hmCD40 mHEVc-LC3-EGFP ) [15] were cultured in DMEM plus 10% FBS ( HyClone; Logan , UT ) . A human RPE cell line ( ARPE-19; American Type Culture Collection , Manassas , VA ) , a mouse macrophage line ( RAW 264 . 7 ) and mouse microglia line ( BV-2 ) were cultured in DMEM plus 10% FBS . Mouse bone marrow-derived macrophages were obtained as described and cultured in DMEM plus 30% L929-conditioned medium , 10% FBS and 5% horse serum [68] . Parental Chinese Hamster Ovary ( CHO ) cells and CHO cells expressing human EGFR ( CHO-EGFR ) were cultured in MEM plus 10% FBS . Experiments were conducted using tachyzoites of the RH strain of T . gondii ( Type I strain ) , RH that express cytoplasmic YFP [69] or cytoplasmic DsRed ( RFP ) [69] , tachyzoites of the ME49 ( type II strain ) , transgenic parasites deficient in micronemal proteins MIC1 ko , MIC3 ko , MIC1-3 ko and the complemented strains ( MIC1ko+MIC1 , MIC3 ko+MIC3 and MIC1-3 ko+MIC1-3; gift from Maryse Lebrun , Universite de Montpellier 2 , France ) [39] , as well as conditional MIC8 ko parasites ( gift from Markus Meissner , University of Glasgow ) . Parasites were maintained in human foreskin fibroblasts following standard procedures [70] . In order to deplete MIC8 , conditional MIC8 ko parasites were cultured in HFF in the presence of anhydrotetracycline ( 1 µg/ml ) for 48 h . T . gondii tachyzoites were killed by incubation in 1% paraformaldehyde in PBS . A potassium buffer shift was used to synchronize T . gondii invasion of serum-starved ( 0 . 1% FBS ) mammalian cells as described [71] . Briefly , freshly egressed tachyzoites were resuspended in Endo buffer and incubated with cells for 20 minutes at 37°C . The Endo buffer was replaced for a low-potassium permissive medium to allow parasite invasion . In certain experiments , mammalian cells were incubated with Akt inhibitor IV ( 1 . 25 µM; EMD Millipore , Billerica , MA ) , PI3K inhibitor ( LY294002; 20 µM; Sigma-Aldrich; St . Louis , MO ) , EGFR inhibitor ( AG1478; 1 µM; EMD Millipore ) , a broad spectrum ADAM inhibitor ( GM6001; 10 µM; EMD Millipore ) ( all 1 h prior to challenge with T . gondii ) , Pertussis Toxin ( PTx; 100 ng/ml; EMD Millipore; 4 h prior to challenge ) , leupeptin ( 10 µM; EMD Millipore ) and pepstatin ( 10 µM; EMD Millipore; both 1 h after challenge with T . gondii ) , 3-methyl adenine ( 3MA; 10 mM; Sigma Chemical ) and rapamycin ( 1 µM; EMD Milipore; both 2 h after challenge with T . gondii ) or vehicle . To induce CD40 signaling , mHEVc cells were treated with cell-free supernatants containing either multimeric human CD154 or a non-functional CD154 mutant [72] ( T147N; both obtained from Dr . Richard Kornbluth , Univ . of California San Diego , current address Multimeric Biotherapeutics Inc . , La Jolla , CA ) for 18 h at 37°C as previously described [73] prior to challenge with parasites . Monolayers were fixed at indicated time points and stained with Diff-Quick ( Dade Diagnostics , Aguada , Puerto Rico ) . The percentage of infected cells , the numbers of tachyzoites and vacuoles per 100 cells as well as the numbers of parasites per vacuole were determined by light microscopy by counting at least 200 cells or 200 vacuoles per monolayer as previously described [15] . For expression of MIC3 and MIC4 in P . pastoris , amplified DNA fragments were cloned into a pPICZα A vector ( Invitrogen; Carlsbad , CA ) . The pPICZα A vector contains the S . cervisiae α-factor secretion signal that allows for the secretion of folded proteins from P . pastoris . Cells were grown in BMGY media , washed and resuspended in BMMY media for an initial OD600 of 20–40 . The culture was then incubated in a 28°C incubator with vigorous shaking . The culture was then grown for 1–5 days depending on the optimal period of expression . Inhibition of glycosylation during culture required the addition of 20 µg/ml of tunicamycin . The supernatant is then passed through a HiTrap Q HF Column ( GE Healthcare; Little Chalfont , UK ) . The eluted fraction was buffer exchanged into nickel-column binding buffer ( 50 mM Tric-HCl , pH 8 . 0 , 50 mM NaCl ) . If needed further protein purification was achieved by a further nickel affinity step and gel filtration . M2AP and MIC6 encompassing residues 87 to 197 ( including EGF2 and EGF3 motifs ) were generated using a pET32 Xa/LIC plasmid ( Novagen , EMD Millipore ) in the Origami ( D3 ) Escherichia coli strain ( Stratagene ) [74] , [75] . Expressions of the fusion protein was induced by adding 1 mM IPTG and harvested after overnight culture at 18°C . The cells were collected by centrifugation and lysed by French Press . The fusion protein incorporating a hexahistidine tag was purified by bench top chromatography using a nickel-nitrilotriacetic acid resin ( QIAGEN ) . The fusion partner of protein was cleaved by factor Xa and removed by an additional chromatography step and the factor Xa was removed by agarose resins ( Novagen ) . Protein samples were concentrated to 0 . 5 mM in 50 mM NaCl , 50 mM potassium phosphate and 5% D2O at pH 5 . 8 . Endotoxin concentrations were similar in MIC3 and MIC4 as well as in M2AP and MIC6 . Cells were transiently transfected with a plasmid that encodes Akt-PH-GFP [76] , human PI3K p110α siRNA [77] , human Akt siRNA [78] , mouse Beclin1 siRNA [79] , mouse Atg7 siRNA [79] , human MyD88 siRNA [80] , human EGFR siRNA [81] or control siRNA ( Dharmacon ) using Lipofectamine 2000 ( Invitrogen ) or an Amaxa nucleofector as described [13] , [15] . siRNA against mouse EGFR was synthesized using siRNA construction kit ( Ambion ) [82] following manufacturer's recommendation and used for mouse EGFR knock-down after transfection using Lipofectamine 2000 . To assess for LC3 accumulation around the parasite , mHEVc-LC3-EGFP cells were cultured with or without Akt inhibitor IV or transfected with either control siRNA or EGFR siRNA or treated with or without EGF ( 50 ng/ml; PeproTech ) . Monolayers were challenged with RH T . gondii that express cytoplasmic RFP ( T . gondii-RFP ) . Five hours post-challenge , monolayers were fixed with 4% paraformaldehyde , slides were mounted using Fluoromount G and assessed for LC3-EGFP accumulation around T . gondii as described [13] , [15] . In certain experiments , hmCD40 mHEVc-LC3-EGFP cells treated with or without CD154 were infected with WT , MIC1 ko , MIC1 ko+MIC1 , MIC3 ko , MIC3 ko+MIC3 , MIC1-3 ko or MIC1-3 ko+MIC1-3 tachyzoites . Monolayers were fixed with 4% paraformaldehyde , permeabilized with 0 . 1% Triton X-100 and incubated with rabbit anti-T . gondii Ab ( BioGenex; San Ramon , CA ) for 30 minutes . Monolayers were then washed with PBS and incubated for 1 h at room temperature with goat anti-rabbit Alexa 568-conjugated secondary antibody ( Jackson ImmunoResearch Laboratories Inc . , West Grove , PA ) . HBMEC transfected with a plasmid encoding PH-Akt-GFP were cultured with or without LY294002 followed by challenge with T . gondii-RFP . Distribution of PH-Akt-GFP was examined 5 min . post-challenge . In certain experiments , endothelial cells were treated with either Akt inhibitor IV or AG1478 and challenged with T . gondii-YFP were fixed with 4% paraformaldehyde at 8 h post-infection , permeabilized with ice-cold methanol . Monolayers were incubated overnight at 4°C with either mouse anti-human LAMP-1 or rat anti-mouse LAMP-1 ( all from Developmental Studies Hybridoma Bank; Iowa City , IA ) . Monolayers were washed with PBS plus 1% BSA , then incubated for 1 h at room temperature with Alexa 568-conjugated secondary antibodies ( Jackson ImmunoResearch Laboratories Inc . ) . Specificity of staining was determined by incubating monolayers with secondary antibody alone . Slides were analyzed using a Leica DMI 6000 B automated microscope equipped for epifluorescence microscopy . Experimental groups had triplicate samples and at least 100 cells per sample were counted . Endothelial cells were seeded onto a sterilized Aclar Embedding Film ( Electron Microscopy Sciences , PA ) and incubated with or without T . gondii tachyzoites in the presence of Akt inhibitor IV or vehicle . At 5 h post-challenge , the Aclar sheets with their attached cells were fixed as described [83] . After a soak in acidified uranyl acetate , the specimen was dehydrated in ethanol , passed through propylene oxide , and embedded in Poly/Bed ( Polysciences , PA ) . Sections were cut in a horizontal plane parallel to that of the Aclar film to provide panoramic views of the endothelial cells . Thin sections were stained with acidified uranyl acetate in 50% methanol followed by triple lead stain of Sato . These sections were examined in a JEOL 1200 EX electron microscope ( Tokyo , Japan ) . Cells were lysed in buffer supplemented with protease and phosphatase inhibitors ( Cell Signaling ) . Equal amounts of protein were subjected to either 7 . 5% or 10% SDS-PAGE ( Bio-Rad ) and transferred to PVDF membranes . Membranes were probed with either antibody to total Akt ( Cell Signaling ) , phospho-Ser473 Akt ( Cell Signaling ) , total EGFR ( Santa Cruz Biotechnology ) , phospho-tyrosine 1068 EGFR ( Invitrogen ) , phospho-tyrosine 1148 EGFR ( Cell Signaling ) , Atg7 ( Cell Signaling ) , Beclin 1 ( BD Biosciences ) , PI3K p110α ( Cell Signaling ) or MyD88 ( Cell Signaling ) followed by incubation with secondary antibody conjugated to horseradish peroxidase ( Santa Cruz Biotechnology ) . Bands were visualized by using enhanced chemiluminescence kit ( Pierce Bioscience ) . Intensities of phospho-Akt and phospho-EGFR were calculated using ImageJ ( NIH ) and normalized against total Akt and total EGFR respectively . Results from pooled experiments were analyzed for statistical significance using 2-tailed Student's t test and ANOVA ( InStat version 3 . 0 , GraphPad; La Jolla , CA ) . Differences were considered statistically significant when P was<0 . 05 . | Toxoplasma gondii resides in a parasitophorous vacuole that excludes transmembrane proteins required for recruitment of endosomes and lysosomes and thus , does not follow the path of classical lysosomal degradation . However , the non-fusogenic nature of the vacuole can be reverted when autophagy , a pathway to lysosomal degradation , is upregulated through the immune system or pharmacologically . Maintenance of the non-fusogenic nature of the vacuole is central to parasite survival . Thus , in addition to preventing degradation through a classical lysosomal pathway , T . gondii may also deploy strategies to prevent constitutive levels of autophagy from targeting the pathogen and causing its lysosomal degradation . We report that T . gondii accomplishes this task by causing EGFR activation in host cells . In cells that were not subjected to immune or pharmacologic upregulation of autophagy , blockade of EGFR resulted in parasite encasing by structures that expressed the autophagy protein LC3 , vacuole-lysosomal fusion and autophagy protein-dependent killing of the parasite . Moreover , EGFR signaling also impaired targeting of the parasite by LC3+ structures in cells treated with stimulators of autophagy . Studies with T . gondii deficient in EGF domain containing-micronemal proteins ( EGF-MICs ) and recombinant EGF-MICs support the concept that these parasite adhesins contribute to EGFR activation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2013 | Toxoplasma gondii-Induced Activation of EGFR Prevents Autophagy Protein-Mediated Killing of the Parasite |
The ontogeny of large-scale functional organization of the human brain is not well understood . Here we use network analysis of intrinsic functional connectivity to characterize the organization of brain networks in 23 children ( ages 7–9 y ) and 22 young-adults ( ages 19–22 y ) . Comparison of network properties , including path-length , clustering-coefficient , hierarchy , and regional connectivity , revealed that although children and young-adults' brains have similar “small-world” organization at the global level , they differ significantly in hierarchical organization and interregional connectivity . We found that subcortical areas were more strongly connected with primary sensory , association , and paralimbic areas in children , whereas young-adults showed stronger cortico-cortical connectivity between paralimbic , limbic , and association areas . Further , combined analysis of functional connectivity with wiring distance measures derived from white-matter fiber tracking revealed that the development of large-scale brain networks is characterized by weakening of short-range functional connectivity and strengthening of long-range functional connectivity . Importantly , our findings show that the dynamic process of over-connectivity followed by pruning , which rewires connectivity at the neuronal level , also operates at the systems level , helping to reconfigure and rebalance subcortical and paralimbic connectivity in the developing brain . Our study demonstrates the usefulness of network analysis of brain connectivity to elucidate key principles underlying functional brain maturation , paving the way for novel studies of disrupted brain connectivity in neurodevelopmental disorders such as autism .
Understanding the development of human brain organization is critical for gaining insight into brain organization and functions in adulthood as well as for investigating disorders such as autism spectrum disorders ( ASD ) and attention-deficit/hyperactivity disorder ( ADHD ) , where normal developmental processes are disrupted . Neuroimaging studies of development have primarily focused on structural changes from childhood , to adolescence , and into adulthood . These studies have reported age-related changes in ( 1 ) overall brain volumes [1] , [2] , ( 2 ) volumes of individual brain areas [3] , [4] , ( 3 ) regional cortical thickness [5] , [6] , as well as ( 4 ) regional and global grey-matter and white-matter densities [7]–[9] . Collectively these studies have suggested that the human brain undergoes vast developmental changes in grey and white matter structure between childhood and adulthood . These changes are thought to reflect synaptic pruning and myelination observed at the neuronal level [8] , [9] . More recently , diffusion tensor imaging ( DTI ) studies investigating the development of white-matter pathways have shown increase in anisotropy [10]–[12] , decrease in overall diffusion [13] , and maturation in major white-matter fiber tracts [14]–[19] , with age . In spite of growing evidence from these studies for patterned brain development , the functional organization of the human brain in childhood is not well understood and it is also not clear how the above structural changes translate to differences in functional brain organization between children and adults . Task-free ( resting-state ) functional connectivity MRI is a useful technique for investigating the functional organization of the human brain . This method detects interregional correlations in spontaneous blood oxygen level-dependent ( BOLD ) signal fluctuations [20] , [21] , and has been used to investigate brain networks involved in motor [20] , sensory [22] , attention [23] , salience and cognitive control [24] , and memory [25] , [26] systems . However , only a small number of studies have examined developmental changes in functional brain organization . A few recent studies have examined developmental changes in functional connectivity of brain regions involved in attention and cognitive control [27] and the default mode network ( DMN ) [28] , as well as in functional connectivity of anatomical structures such as the anterior cingulate cortex [29] . To our knowledge , the developmental changes in the functional organization of large-scale networks at the whole-brain level have not yet been investigated . Here we use a graph theoretical approach to examine developmental changes in the large-scale functional organization of the human brain . Graph metrics such as the clustering coefficient and the characteristic path length [30] , [31] have been shown to be useful measures of organization of large-scale networks . Briefly , graphs are data structures that have nodes and edges between the nodes [32] . The clustering coefficient is a measure of local network connectivity . A network with a high average clustering coefficient is characterized by densely connected local clusters . The characteristic path length is a measure of how well a network is connected . A network with a low characteristic path length is characterized by short distances between any two nodes . Many biological systems have small-world network properties , characterized by a high clustering coefficient and a low characteristic path length [30] , [33] . These graph-theoretic metrics have also proven useful in modeling the large-scale functional and structural organization of the human brain [34]–[37] . In a graphical representation of a brain network , a node corresponds to a brain region while an edge corresponds to the functional connectivity between two brain regions . Functional connectivity networks of the human brain derived from electroencephalograms ( EEGs ) , magnetoencephalograms ( MEGs ) , and task-free functional magnetic resonance imaging ( fMRI ) data have been shown to exhibit small-world characteristics [35] , [38] , [39] . These studies suggest that small-world metrics are suited to quantify the global topological properties of large-scale organization of the human brain . Recently , in addition to small-world metrics , Bassett and colleagues used graph theoretic metrics such as hierarchy to characterize local topological properties of large-scale organization of the human brain . Using structural brain imaging data and modeling of interregional covariance in cortical thickness , they reported that hierarchical organization in anatomical human brain networks is characterized by the presence of frontal hubs [40] . A recent study of aging by Meunier and colleagues investigated the modular organization of large-scale functional brain networks using Newman's graph-based modularity metric . They reported that while both young and older adults showed modularity of network organization , the topological roles of the specific brain regions as well as the intermodular connectivity was significantly different between the two groups [41] . The use of small-world metrics along with more advanced graph theoretic metrics to characterize local organization of complex networks provides a new approach for investigating large-scale functional organization of the human brain at multiple levels of granularity . We investigated developmental changes in the functional organization of large-scale brain networks at multiple levels by ( 1 ) creating whole-brain functional connectivity networks from task-free fMRI data , ( 2 ) characterizing the organization of these networks using metrics of global and local brain organization ( including small-worldness and hierarchy , as defined in the Methods section ) , and ( 3 ) comparing these metrics of global and local brain organization between healthy children ( ages 7–9 y ) and young-adults ( ages 19–22 y ) . In older adults ( age ±40 y ) it is now well established that large-scale brain networks have a small-world architecture that reflects a robust and efficient , nonrandom , functional organization [34] , [35] , [39] . Whether children and younger adults have a similar functional brain organization is currently not known . This question is important from a developmental perspective because the brain undergoes vast changes in structural connectivity during adolescence [9] . We hypothesized that the global functional organization of brain networks would be characterized by nonrandom , efficient , small-world characteristics in both subject groups , but that young-adults would show higher small-worldness compared to children , on the basis of previous neurobiological studies in humans and animals suggesting that developmental changes improve efficiency of information processing in the brain [42]–[44] . We further predicted that local organization patterns would be significantly different in children , reflecting a process of continuing structural maturation during the period between childhood and young adolescence . To further characterize developmental changes in the global and local functional organization of brain networks , we used the parcellation scheme of Mesulam [45] to examine functional organization in five key subdivisions: primary sensory , subcortical , limbic , paralimbic , and association areas . Additionally , developmental changes in the connectivity between these subdivisions ( hereafter referred to as interregional connectivity in the manuscript ) were examined . Lastly , to characterize the underlying developmental processes that produce these changes in the global and local functional organization of large-scale brain networks , we examined changes in functional connectivity as a function of DTI-based wiring distance between distinct brain regions . The formation of brain networks during development is thought to arise from a dual process of integration and segregation [27] , [46]–[49] . Accordingly , we investigated whether there is in vivo developmental evidence for the emergence of functional segregation and integration in large-scale brain networks .
Demographic and cognitive profile data for the child and young-adult groups are shown in Table 1 . The two groups were well-matched and did not differ in IQ ( p = 0 . 93 ) or gender ( p = 0 . 75 ) . We first examined graph-theoretic metrics obtained for the functional brain networks constructed by thresholding ( threshold values ranged from 0 . 01 to 0 . 99 , with an increment of 0 . 01 ) the wavelet correlation matrix at three different frequency scales . Scale 1 encompassed 0 . 13–0 . 25 Hz , scale 2 encompassed 0 . 06–0 . 12 Hz , and scale 3 encompassed 0 . 01–0 . 05 Hz . As shown in Figure 1 , for both the children and young-adult groups , the mean degree—the average number of edges incident on a node belonging to the network—was highest at scale 3 for a wide range of correlation thresholds ( 0 . 01<R<0 . 8 ) . The mean characteristic path length ( λ ) for both groups , when controlled for the degree of the network ( 1<λ<1 . 57 ) , showed similar trends at all three scales . The clustering coefficient ( γ ) for both groups , when controlled for the degree of the network , was highest at scale 3 . Due to higher mean γ values , the small-world measure σ ( γ/λ ) , when controlled for degree of the network , was highest at scale 3 for both groups . The small-world property ( σ>1 ) showed a monotonic increase in small-worldness as the threshold increased and the degree decreased . σ values for higher correlation thresholds are difficult to interpret because at higher threshold values , graphs of functional brain networks have fewer edges ( smaller degree ) and tend to split into isolated subgraphs . Graph metrics such as clustering coefficient , characteristic path length , and small-world property do not meaningfully characterize network structures that are not composed of a single , large group of interconnected nodes [30] . Since functional connectivity and small-world properties were highest ( p<0 . 01 , Kolmogorov-Smirnov test ) at lower-frequencies ( scale 3: 0 . 01– 0 . 05 Hz ) for both children and young-adults , we focus on this frequency interval in subsequent analyses , consistent with other recent studies [34] , [35] . We examined path length ( λ ) , clustering coefficient ( γ ) , and small-worldness ( σ ) values in the two groups in scale 3 ( 0 . 01–0 . 05 Hz ) . For group comparisons , we controlled for the average correlation value ( r ) , as it varies considerably across individuals . Thus , for a given correlation threshold , the number of edges in the graph are likely to be different , resulting in different λ and γ values . To ensure that graphs in both groups had the same number of edges , individual correlation matrices were thresholded such that the resultant graph had on average K′ edges per node . K′ is the average number of edges per node in the graph obtained by thresholding individual correlation matrices with R = ri ( ri is the average correlation value for subject i , i = 1–45 ) , averaged across subjects . This procedure not only ensured that both groups had the same number of edges , but also selected a conservative K′ such that the networks generated were not disconnected . This is particularly important for network characterization because graph metrics are not interpretable when the network is disconnected . The value of K′ selected according to this procedure was 48 for both the groups . Thus , every network generated by using this degree preserving threshold will have exactly 2 , 160 ( = 48×90/2 ) edges , which is equivalent to a network cost of 0 . 54 ( = 2 , 160/4 , 005 ) . Network cost is defined and calculated as the ratio of number of edges in the network to the maximum possible edges in the network [50] . Mean λ , mean γ , and mean σ values for the networks of each group were derived by thresholding the correlation matrices such that the network has on average K′ ( = 48 ) edges per node . Using this approach , no significant differences in the mean λ , γ , and σ values were observed between children and young-adults . Global efficiency , the harmonic mean of the minimum path length between each pair of nodes , is an alternative measure of connectivity of the network . This measure overcomes some of the limitations of the original measure of network connectivity , characteristic path length , which is susceptible to disconnected nodes . We examined global efficiency ( Eglobal ) values obtained for the functional brain networks constructed by thresholding ( threshold values ranged from 0 . 01 to 0 . 99 with an increment of 0 . 01 ) , the wavelet correlation matrix at each of the three scales . The mean Eglobal for both groups , when controlled for the degree of the network , was low ( 0 . 7<Eglobal<1 ) and showed similar trends at all three scales . In the frequency interval 0 . 01–0 . 05 Hz ( scale 3 ) , mean Eglobal values for the two groups , obtained by thresholding the correlation matrices such that the network has on average K′ ( = 48 ) edges per node , which is equivalent to a network cost of 0 . 54 were compared . No significant differences in the mean Eglobal values were observed between the two groups . We examined differences in whole-brain functional connectivity patterns between children and young-adults . The connectivity patterns , - correlation values of 4 , 005 pairs of anatomical regions , were used as features in a support vector machine ( SVM ) classifier ( see Text S1 ) . We found that connectivity patterns in children could be distinguished from those in young-adults with accuracies ranging from 89% to 91% , with the highest accuracy in scale 3 ( see Table 2 ) . This suggests that functional connectivity patterns at the whole-brain level in children are significantly different from those in young-adults . We report below the nature of these developmental changes in the context of hierarchical and regional organization of brain connectivity . Hierarchy ( β ) is a measure of the relationship between the clustering coefficient and number of nodes in the network . Networks with higher hierarchy values are characterized by high degree nodes , which exhibit low clustering , and vice versa . These hierarchical networks contain small densely connected clusters; these clusters combine to form large less-interconnected clusters , which combine again to form larger lesser-interconnected clusters [51] . We examined β values obtained for the functional brain networks constructed by thresholding ( threshold values ranged from 0 . 01 to 0 . 99 with an increment of 0 . 01 ) the wavelet correlation matrix at scale 3 ( 0 . 01–0 . 05 Hz ) . As shown in Figure 2A , the β values for both groups , when controlled for the degree of the network , were significantly higher ( −7 . 5<β<2 . 5 ) than β values obtained from random networks ( p<0 . 01 ) . Furthermore , β values in the young-adult group were significantly higher than in the child group ( p<0 . 001 , Kolmogorov-Smirnov Test ) . The mean β value for the two groups , obtained by thresholding the correlation matrices such that the network has on average K′ ( = 48 ) edges per node , which is equivalent to a network cost of 0 . 54 , was significantly higher in young-adults than in children ( p<0 . 01 ) , as shown in Figure 2B . We then examined regional differences in network organization of five major divisions—association , limbic , paralimbic , primary , and subcortical areas [45]—with the rest of the brain . Figure 3 shows a plot of degree , path length ( λ ) , efficiency , and clustering coefficient ( γ ) values for each of the five areas , for children and young-adults , as a function of the correlation threshold . In the subcortical division , the fitted growth curve of degree and efficiency values was significantly higher ( p<0 . 01 ) while the curve of λ values was significantly lower ( p<0 . 01 ) in children , compared to young-adults , reflecting higher connectivity , higher efficiency values , and lower path length for a range of threshold values from 0 . 1 to 0 . 6 . A similar analysis in the association , limbic , paralimbic , and primary sensory areas , revealed no significant differences in the degree , λ , efficiency , and γ values . Across the five divisions , no significant differences in the degree , λ , efficiency , and γ values were observed for correlation threshold values >0 . 6 , mainly due to the large variance observed at higher threshold values . We next examined the degree , λ , efficiency , and γ values for each of the 90 anatomical ROIs , for the two groups . Consistent with the above findings , a significant number of subcortical areas ( six out of eight; p<0 . 01 ) showed differences between the two groups in at least one of the four metrics ( degree , λ , efficiency , and γ ) , whereas only two out of eight regions in the primary sensory , 17 out of 44 regions in association , three out of six regions in limbic , and 12 out of 24 regions in the paralimbic areas , showed differences ( see Table S1 for regions that showed significant differences in degree , λ , efficiency , and γ values between the two groups ) . We next examined connectivity differences within each of the five functional subdivisions . Connectivity differences here reflect the change in the strength of interregional correlations in spontaneous blood oxygen level-dependent fluctuations . The functional connectivity within the paralimbic areas was significantly higher in the young-adults , compared to children ( p<0 . 001; p<0 . 01 , false discovery rate ( FDR ) -corrected for multiple comparisons ) . There were no differences in functional connectivity within the association , limbic , primary , and subcortical areas . To further investigate regional differences in network organization , we examined interregional connectivity differences between the two groups . We found that the subcortical areas had increased connectivity with the primary sensory , association , and paralimbic areas in children , compared to young-adults . Young-adults , on the other hand , had increased connectivity between paralimbic and association areas , between paralimbic and limbic areas , and between limbic and association areas ( p<0 . 001; p<0 . 01 , FDR-corrected for multiple comparisons ) ( Figure 4A ) . The classification analysis of interregional connectivity showed complementary set of findings ( see Text S1 for details ) . The interregional connectivity patterns in children could be distinguished from those in young-adults with accuracies ranging from 44% to 91% , with high accuracy values observed for connectivity patterns between subcortical areas and the primary sensory ( 91% ) , association ( 90% ) and paralimbic ( 83% ) areas , and between paralimbic and association ( 80% ) areas ( see Table 3 ) . Figure 4B shows a graphical representation of developmental differences in functional connectivity along the posterior-anterior and ventral-dorsal axes , highlighting greater subcortical connectivity in children and greater paralimbic connectivity in young-adults . Figure S1 shows separate group-averaged functional connectivity matrices for children and young-adults , and Text S1 provides information about interparticipant variability in these matrices . Lastly , we investigated whether development is associated with simultaneous emergence of functional segregation and integration at the whole-brain level . For each pair of ROIs we first computed the wiring distance using DTI-based fiber tracking . We computed the fiber length in a common Montreal Neurological Institute ( MNI ) space rather than individual subject space to rule out any potential confounding effects of developmental changes in interregional fiber length on our findings . We then examined developmental changes in functional connectivity in relation to the wiring distance between them . We found that functional connectivity between more proximal anatomical regions were significantly higher in children , whereas functional connectivity between more distal anatomical regions were significantly higher in young-adults ( p<0 . 0001 ) , as shown in Figure 5 . This suggests a pattern of higher short-range functional segregation in children and higher long-range functional integration in young-adults . To further examine the robustness of our findings , we repeated our functional connectivity versus wiring distance analysis using Euclidean distance instead of DTI-based wiring distance . The results were highly consistent with those reported above: functional connectivity between more proximal anatomical regions in Euclidean space was significantly higher in children , whereas functional connectivity between more distal anatomical regions in Euclidean space was significantly higher in young-adults ( p<0 . 0001 ) .
A small-world network is characterized by a high clustering coefficient and a low characteristic path length . Functional brain networks in both children and young-adults showed small-world properties ( σchildren>1 , σyoung-adults>1 ) suggesting the presence of subnetworks of densely connected nodes , mostly connected by a short path . Similar findings were observed when clustering coefficient and global efficiency were used as alternative measures of small-worldness . Small-world characterization is well-suited for analyzing functional brain networks at the systems level because these networks are complex and optimally connected to minimize information processing costs [36] , [52] . Functional connectivity networks of the human brain constructed from EEG as well as MEG data have also been shown to have small-world architecture [38] , [39] . Salvador et al . [53] examined connectivity in task-free functional MRI data with the same 90 ROI parcellation scheme used in our study and they reported small-world architecture in this network . This finding was replicated by Achard et al . , who also reported that small-world properties were salient in the low frequency interval 0 . 03–0 . 06 Hz [35] in adults ( ages 25–35 y ) , and by Supekar et al . in older adults ( ages 37–77 y ) [34] . These findings , primarily derived from functional data obtained from middle-age to older adults , suggest that the functional organization of the brain has a small-world architecture , a characteristic that may assist in robust and dynamic information processing . Our finding that large-scale brain networks in children showed small-world properties that were very similar to young-adults , together with the above observations , suggests that key aspects of functional brain organization are conserved throughout the developmental process—from early childhood to young adulthood and into older adulthood . Critically , despite the fact that the brain undergoes vast structural reorganization at the neuronal level in the form of myelination and synaptic pruning throughout development , key global properties of functional organization appear to be conserved . Notwithstanding similarities in global , whole-brain , small-world network properties , functional connectivity patterns in children were significantly different from those in young-adults . SVM-based pattern classification analysis showed that connectivity patterns in children could be distinguished from those in young-adults with an accuracy of over 90% . Accuracy was highest ( 91% ) for connectivity patterns in the low frequency interval ( scale 3; 0 . 01–0 . 05 Hz ) . Previous studies have reported that resting-state functional connectivity is most robust at frequencies below 0 . 1 Hz [20] , [54] , [55] and that these low frequency fMRI fluctuations are related to interregional coupling of local field potentials in the gamma band [56] , [57] . Overall , these findings suggest that observed developmental changes in the functional connectivity measured by fMRI resting state signals are likely to reflect underlying differences in coupling of neuronal signals . We discuss below the nature of developmental changes in the context of hierarchical and regional organization of brain connectivity . Our data provide new evidence that large-scale brain networks in children and young-adults differ in their hierarchical organization . Children showed significantly lower ( p<0 . 001 ) levels of hierarchical organization than young-adults . Hierarchical networks are characterized by the presence of small densely connected clusters; these clusters combine to form large less-interconnected clusters , which combine again to form larger lesser-interconnected clusters [51] . Hierarchical organization has been discovered in the World Wide Web and several biological networks [40] , [58] , [59] . In a recent study , Bassett and colleagues reported significant levels of hierarchical organization in anatomical human brain networks based on interregional correlations in cortical thickness [40] . Our study extends these findings to the realm of hierarchical organization in functional human brain networks in not only young-adults but also in children . Hierarchical networks are optimally connected to support top-down relationships between nodes and minimize wiring costs , but are vulnerable to attack on hubs [51] . The presence of hierarchical organization in the large-scale brain networks of children and young-adults suggests efficient functional connectivity patterns within these networks at the expense of higher vulnerability to attacks . Lower levels of hierarchical organization in children may therefore be protective to such vulnerability , allowing for more flexibility in network reconfiguration on the basis of individual differences in cognitive experience and reserve . How modularity and hierarchy emerge in functionally meaningful ways is an important topic for future research , but the important finding here is that quantitative measures of hierarchy can be used to examine the emergence of functional hierarchy in the developing brain . We used the parcellation scheme of Mesulam to examine developmental changes in the functional connectivity of five major functional divisions of the human brain . Briefly , the primary sensory division consists of unimodal regions for processing visual , auditory , somatosensory , olfactory , and gustatory signals . The subcortical division includes deep brain nuclei , notably the basal ganglia and thalamus , and the association division comprises higher order multimodal regions , including the lateral prefrontal , parietal , and temporal cortices . The paralimbic division consists of the insula , anterior cingulate cortex , posterior cingulate cortex and the orbitofrontal cortex , and the limbic division includes the amygdala and hippocampus . Together , these divisions map the external world into brains' internal sensory , attentional , mnemonic , emotional , and motivational systems [60] . Graph-theoretical analysis identified subcortical regions as a major locus of between-group differences in brain connectivity . More specifically , subcortical connectivity was characterized by higher degree , lower path length and higher efficiency in children ( Figure 3 ) . Node wise analysis showed that the caudate , putamen , and thalamus all showed higher degree , lower path length , and higher efficiency in children . The globus pallidus was the only subcortical region that did not differ in these network metrics between children and young-adults . Further analysis of functional connectivity with the other four subdivisions revealed that subcortical areas were more strongly correlated with primary sensory , association , and paralimbic areas in children , as shown in Figure 4A . These results suggest that subcortical-cortical connections are both more profuse and stronger in children and that the functional development of subcortical connectivity is characterized by both changes in wiring and strength of connections . We also detected significant differences in cortical connectivity but in this case the pattern of age-related differences was reversed , with children showing significantly weaker connectivity between paralimbic , association , and limbic areas ( Figure 4A ) . Graph-theoretical measures of degree , efficiency , and path length of the four cortical subdivisions did not differ between the two groups ( Figure 3 ) . This suggests that key aspects of cortico-cortical wiring are similar in children and young-adults but the strength of the connections is weaker in children . These developmental changes converge on and extend findings from structural neuroimaging studies that have shown protracted age-related structural differences in the regional gray- and white-matter [6]–[9] , [15] . Our findings of differences in subcortical connectivity is consistent with reports that these areas undergo massive structural rewiring characterized by progressive myelination of axons that emanate from these regions followed by extension of these myelinated axons into the cortex during development [11] , [15] . The later teen years , which span an interval in between childhood and young-adulthood is a period of significant brain maturation [61] . In particular , caudate , putamen , and thalamus regions of the subcortical division show some of the largest changes in fractional anisotropy of white-matter tracts , increasing almost 30% to 50% from 5 to 25 years of age . In contrast , major cortico-cortico tracts show a more modest increase of 8% to 20% [15] . Taken together , these results suggest that changes in interregional functional connectivity parallel changes in maturation of white-matter tracts between childhood and young-adulthood . Critically , our data provide novel evidence for a process of rewiring and pruning of subcortical-cortical connectivity accompanied by increased cortico-cortical connectivity at the functional level . Subcortical areas , comprising the basal ganglia and the thalamus , are important for adaptive processing of distributed information in a manner that facilitates the transformation of sensory input and cognitive operations into behavior [62] . More specifically , the basal ganglia link signals in distinct functional networks during different phases of cognitive information processing [63] . Neurophysiological models and anatomical tracing studies have provided evidence for parallel motor , limbic , and prefrontal cortico-basal-ganglia loops [64] , [65] , which funnel large-scale cortical activity into behaviorally relevant motor output . In humans , these circuits are characterized by segregated and overlapping connectivity patterns and a complex pattern of hierarchically organized frontal inputs [66] , [67] . These patterns support the parallel flow of cortical signals inputs into the basal ganglia , where multiple reward related signals are integrated in ways that facilitate incentive learning over short time scales and habit formation over long time scales [68] , [69] . There have been few studies of how these loops develop in children , but the pattern of changes in subcortical-cortical functional connectivity observed in our study suggest a process of pruning at the systems-level . This form of pruning is characterized by weakening of specific subcortical links , leading to longer path lengths similar to those seen in young-adults . Exactly how these links result in the formation of parallel and integrative loops , which support large-scale neuronal networks for learning and memory [63] , [70] remains to be investigated . Changes in paralimbic connectivity were the cornerstone of developing cortico-cortico connectivity . Paralimbic areas play a major role in detection of salient environmental events [71] , in facilitating flexible behaviors in response to risk , reward , and punishment [72] , [73] , and in goal directed behavior [74] . Converging evidence from a number of brain imaging studies across several task domains suggests that the insula and the anterior cingulate cortex respond to the degree of subjective salience , whether cognitive , homeostatic , or emotional [75] , [76] . These paralimbic areas play a causal role in activating attentional and memory systems within association areas to facilitate controlled processing of stimuli during cognitively demanding tasks [71] . Paralimbic and association areas also moderate emotional reactivity to stimuli in limbic areas [77] , [78] . These core motivational and regulatory processes are known to undergo significant changes during adolescence [79] , a time when coordinated interaction of emotion , reasoning , and decision-making becomes increasingly important [80] , [81] . The tighter integration of paralimbic with association and limbic areas revealed by our study may underlie the large-scale functional changes that facilitate this critical developmental process . Our analysis of functional connectivity changes with wiring distance provides strong evidence that development is characterized by simultaneous reduction of short-range connectivity and strengthening of long-range connectivity . This suggests a process of greater functional segregation in children and greater functional integration in young-adults at the whole-brain level , not just in circumscribed nodes of the attentional control [27] and default node networks [28] . In contrast to the 90 cortical and subcortical nodes , based on whole-brain parcellation [82] , used in our study , Fair and colleagues [27] , [28] focused their analysis on 39 distinct cortical regions involved in task-control and default-mode networks . Whereas the lack of correspondence between specific brain regions in the two studies makes a detailed comparison difficult , our findings are , nevertheless , consistent with distributed changes in these two large-scale networks reported by Fair and colleagues . Methodologically , our studies are an improvement over prior studies because we used continuous resting state fMRI data , rather than resting state data extracted from intertask rest periods , uncontaminated by cognitive tasks . Furthermore , our findings indicate that simultaneous weakening of short-range connections and strengthening of long-range connections changes with actual anatomical ( physical ) distance , derived from DTI data , rather than the Euclidean distance , between nodes . Our findings provide new and more direct evidence that dual changes in functional integration and segregation with wiring distance reflects a general developmental principle that operates at the level of the whole brain . Two neurobiological processes are likely to contribute directly to these observed effects . One , systematic pruning of local connections with age are likely to result in weakening of local connections and formation of more localized and specialized processing nodes . These changes are known to occur prenatally , in childhood and in adolescence [83] . In parallel , increased myelination of axonal fiber tracks with age , also contribute to strengthening of long-range connectivity [84] . Both these processes are likely to be influenced by experience dependent Hebbian plasticity , leading to selective strengthening and weakening of connections [85] . This selective strengthening and weakening of connections may be additionally influenced by developmental changes in interregional wiring distance . For example , on the basis of previous findings of an inverse relationship between strength of functional connectivity and wiring distance in adults [53] , [86] , the observed age-related decrease in subcortical-cortical functional connectivity may be due to age-related increases in subcortical-cortical wiring distance . In our analysis , however , we controlled for any confounding influences of changes in physical wiring distance by computing functional connectivity and wiring distance in a common MNI space . Further studies that examine both functional connectivity and wiring distance in native image space are needed in order to investigate the influence of age-related changes in wiring distance on the observed age-related changes in functional connectivity . More generally , the manner in which these structural and functional changes in connectivity influence the development of large-scale functional organization in the human brain is an important topic for future research . Recent studies do , however , suggest that intrinsic resting-state functional connectivity in the human brain reflects anatomical connectivity at both short and long spatial scales [87] , [88] . Taken together , these findings suggest that the development of large-scale functional connectivity is related to ongoing developmental changes in structural connectivity . Our findings suggest that large-scale brain networks derived from task-free fMRI have a robust functional organization in 7–9-y-old children . Importantly , we show that the dynamic process of over-connectivity followed by pruning , which rewires connectivity at the neuronal level [89] , also operates at the systems level and helps reconfigure and rebalance subcortical and paralimbic connectivity in the developing brain . Our study demonstrates the usefulness of network analysis of functional connectivity in elucidating the principles underlying brain maturation . Furthermore , our study shows how quantitative analysis of anatomical connectivity , and in particular the computation of wiring distance between brain regions , allows us to link changes in functional networks to the maturation of white matter tracts . Such multimodal analysis of structural and functional brain connectivity will prove useful in helping us better understand the network architecture that shapes and constrains cognitive development . More generally , our findings provide a framework for examining how fundamental aspects of large-scale organization are disrupted in neurodevelopmental disorders . Previous work has suggested that resting-state functional connectivity can be used to assess disrupted connectivity between specific brain regions that are relevant to the disease-specific pathology in neurodevelopmental disorders such as autism spectrum disorders [90] , and attention-deficit/hyperactivity disorder [91] , disorders that are thought to be characterized by disruptions in synaptic pruning and myelination at the neuronal level [92]–[94] . The methods and results developed here provide a template for a more detailed investigation of disruptions in the large-scale organization of brain networks in these and other developmental brain disorders .
Twenty-three children and 22 IQ-matched young-adult subjects participated in this study after giving written , informed consent . For those subjects who were unable to give informed consent , written , informed consent was obtained from their legal guardian . The study protocol was approved by the Stanford University Institutional Review Board . The children subjects ( 10 males , 13 females ) ranged in age from 7 to 9 y ( mean age 7 . 95 y ) with an IQ range of 88 to 137 ( mean IQ: 112 ) ; the young-adult subjects ( 11 males , 11 females ) ranged in age from 19 to 22 y ( mean age 20 . 4 y ) with an IQ range of 97 to 137 ( mean IQ: 112 ) . The subjects were recruited locally—children from local schools and young-adults from Stanford University and neighboring community colleges . Eleven of 23 children subjects were 2nd graders and the rest of the children subjects were 3rd graders; the young-adult subjects had 13 to 16 y of education ( mean years of education 14 . 5 ) . For the task-free scan , subjects were instructed to keep their eyes closed and try not to move for the duration of the 8-min scan . Functional Images were acquired on a 3T GE Signa scanner ( General Electric ) using a custom-built head coil . Head movement was minimized during scanning by a comfortable custom-built restraint . A total of 29 axial slices ( 4 . 0 mm thickness , 0 . 5 mm skip ) parallel to the AC-PC line and covering the whole brain were imaged with a temporal resolution of 2 s using a T2* weighted gradient echo spiral in-out pulse sequence [95] with the following parameters: TR = 2 , 000 msec , TE = 30 msec , flip angle = 80 degrees , 1 interleave . The field of view was 20 cm , and the matrix size was 64×64 , providing an in-plane spatial resolution of 3 . 125 mm . To reduce blurring and signal loss arising from field inhomogeneities , an automated high-order shimming method based on spiral acquisitions was used before acquiring functional MRI scans . A high resolution T1-weighted spoiled grass gradient recalled ( SPGR ) inversion recovery 3D MRI sequence was acquired to facilitate anatomical localization of functional data . The following parameters were used: TI = 300 msec , TR = 8 . 4 msec; TE = 1 . 8 msec; flip angle = 15 degrees; 22 cm field of view; 132 slices in coronal plane; 256×192 matrix; 2 NEX , acquired resolution = 1 . 5×0 . 9×1 . 1 mm . Structural and functional images were acquired in the same scan session . Data were preprocessed using statistical parametric mapping ( SPM5 ) software ( http://fil . ion . ucl . ac . uk/spm ) . The first eight image acquisitions of the task-free functional time series were discarded to allow for stabilization of the MR signal . Each of the remaining 232 volumes underwent the following preprocessing steps: realignment , normalization to the MNI template , and smoothing carried out using a 4-mm full-width half maximum Gaussian kernel to decrease spatial noise . Excessive motion , defined as greater than 3 . 5 mm of translation or 3 . 5 degrees of rotation in any plane , was not present in any of the task-free scans . The preprocessed task-free functional MRI datasets were parcellated into 90 cortical and subcortical regions using anatomical templates defined by Tzourio-Mazoyer et al . [82] . A task-free fMRI timeseries was computed for each of the 90 regions by averaging all voxels within each region at each time point in the time series , resulting in 232 time points for each of the 90 anatomical regions of interest . These regional fMRI time series were then used to construct a 90 node whole-brain task-free functional connectivity network for each subject . Wavelet analysis was used to construct correlation matrices from the regional fMRI time series data . These matrices described frequency-dependent correlations , a measure of functional connectivity , between spatially distinct brain regions . Correlation matrices were then thresholded to generate a whole-brain functional connectivity network . Wavelets are mathematical functions that transform the input signal into different frequency components [96] . Wavelets methods have previously been applied in the analysis of task-based as well as task-free fMRI signal [35] , [97] . In our study , we applied a maximum overlap discrete wavelet transform ( MODWT ) to each of the 90 regional time series from each subject to obtain the contributing signal in the following three frequency components: scale 1 ( 0 . 13–0 . 25 Hz ) , scale 2 ( 0 . 06–0 . 12 Hz ) , and scale 3 ( 0 . 01–0 . 05 Hz ) . To account for a relatively small number ( 232 ) of data points per time series for low frequency correlation analysis , the vector representing the time series beyond its boundaries ( <0 and >232 ) was assumed to be a symmetric reflection of itself . At each of the three scales , wavelet correlations between signals in the 90 anatomical regions were determined by computing the correlation coefficient between the transformed signals at that scale . For each subject , a 90-node , scale-specific , undirected graph of the functional connectivity network was constructed by thresholding the wavelet correlation matrix computed at that scale . If the wavelet correlation value between two anatomical regions represented by nodes i and j in the network exceeded a threshold , then an edge was drawn between node i and node j . There is currently no formal consensus regarding threshold selection , so we computed networks for threshold values from 0 . 01 to 0 . 99 with an increment of 0 . 01 . Once a whole-brain functional connectivity network was constructed from the correlation matrix , we characterized this network using graph theoretic metrics of global and local brain organization including small-worldness and hierarchy . Small-world properties of a network are described by the clustering coefficient and the characteristic path length of the network . The clustering coefficient and characteristic path length of functional brain networks generated from the task-free fMRI data obtained from 23 children and 22 young-adults were computed . The clustering coefficient of every node was computed as the ratio of the number of connections between its neighbors divided by the maximum possible connections between its neighbors . The clustering coefficient ( C ) of the network was calculated as the mean of the clustering coefficients of all the nodes in the network . The mean minimum path length of a node was computed as the average of minimum distances from that node to all the remaining nodes in the network . The characteristic path length ( L ) of the network was the average of the mean minimum path lengths of all the nodes in the network . The clustering coefficient and path length of nodes completely disconnected with the network were set as 0 and Inf respectively , and these nodes were excluded while computing C and L . To evaluate the network for small-world properties , we compared the clustering coefficient and the characteristic path length of the network with corresponding values ( Cran , Lran ) obtained and averaged across 1 , 000 random networks with the same number of nodes and degree distribution [98] . The degree of every node ( a measure of its connectivity ) was calculated by counting the number of edges incident on that node . The mean degree of the network was the average of the degree of all the nodes in the network . Small-world networks are characterized by high normalized clustering coefficient γ ( C/Cran ) >1 and low normalized characteristic path length λ ( L/Lran ) ≈1 compared to random networks [99] . A cumulative metric σ—the ratio of normalized clustering coefficient ( γ ) to the characteristic path length ( λ ) , a measure of small-worldness—is thus greater than 1 for small world networks . Small-world networks are characterized by high clustering coefficient and low characteristic path length . These small-world metrics , particularly the path length , are not meaningful when the graph contains disconnected nodes . To address this issue , we ensured that only small-world metrics computed on connected graphs were considered in our analysis . Specifically , the algorithm used to choose the correlation threshold ( R ) guaranteed that disconnected graphs were excluded from the analysis . Also , in the node-wise clustering coefficient comparison analysis , we only considered thresholds from 0 . 1 to 0 . 6 . We chose these thresholds because beyond 0 . 6 the network gets divided into disconnected subset of nodes . To determine if our characteristic path length findings were robust and reliable , we computed the efficiency of functional brain networks . It has been previously reported that efficiency as a graph metric ( 1 ) is not susceptible to disconnected nodes , ( 2 ) is applicable to unweighted as well as weighted graphs , and ( 3 ) is a more meaningful measure of parallel information processing than path length [50] . Efficiency of a graph ( Eglobal−net ) [100] is the inverse of the harmonic mean of the minimum path length between each pair of nodes , Lij , and was computed as , ( 1 ) To evaluate the network for its global efficiency of parallel information processing , we compared the global efficiency of the network ( Eglobal−net ) with corresponding values ( Eglobal−ran ) obtained and averaged across 1 , 000 random networks with the same number of nodes and degree distribution . A network with small-world properties is characterized by a global efficiency value that is lower than the random network: Eglobal ( Eglobal−net/Eglobal−ran ) <1 . We evaluated the hierarchical nature of the large-scale whole-brain functional connectivity network by the β parameter [51] . β measures the extent of the power-law relationship between the clustering coefficient ( C ) and the degree ( k ) : C≈k−β . The clustering coefficient ( Ci ) and the degree ( ki ) of every node was computed; β of the network was calculated by fitting a linear regression line to the plot of log ( C ) versus log ( k ) . The human brain can be divided into five major divisions—association , limbic , paralimbic , primary , and subcortical—each of them having a distinct function [45] . We assessed the network organization of these cortical divisions and how it differs in development by examining the regional profile of metrics ( degree , λ , Eglobal , and γ ) at the divisional level . The 90 anatomical regions of our network were grouped into these five cortical divisions . The association division consists of 44 regions , the limbic division consists of 12 regions , the paralimbic division consists of 24 regions , the primary division consists of eight regions , and the subcortical division consists of eight regions ( see Table S1 for region-wise division assignment ) . The graph metrics ( degree , λ , Eglobal , and γ ) of 90 regions were aggregated into five divisions and the aggregated metrics in the two subject groups were compared using growth curve modeling , with an intercept , linear , and quadratic terms . In the aggregation step , the graph metric value at a correlation threshold of a division for a subject group was computed by averaging the corresponding metric values across regions belonging to that division . The aggregated metric values for threshold values from 0 . 1 to 0 . 6 were compared . We chose these thresholds because beyond 0 . 6 the network divides into disconnected subsets of nodes and small-world metrics are no longer meaningful [30] . This analysis was performed using the Mplus software ( http://www . statmodel . com ) . Growth curve models describe change ( growth ) with respect to a control variable . They are well-suited for analyzing group-level differences in biomedical data , particularly in cases where capturing and analyzing individual growth trajectories is important . Furthermore , for group comparisons , growth curve models alleviate the problem of multiple comparisons as fitted-curve coefficients are compared in contrast to traditional approaches where multiple individual points along the curve are compared . In our study , the growth trajectories of graph metric values of a subject carry important information about the variance within the group and needs to be incorporated in the model . The coefficients of growth curve models capture the baseline performance , instantaneous growth rate , and the acceleration of the variable of interest . We next examined degree , λ , Eglobal , and γ values for each of the 90 anatomical ROIs , for the two groups , as a function of the correlation threshold . The metric values for threshold values from 0 . 1 to 0 . 6 in the two subject groups were compared using growth curve modeling , as described above . To further characterize regional differences in network organization , we examined the regional connectivity at divisional level: association , limbic , paralimbic , primary , and subcortical . Differences in mean correlation coefficients for 4 , 005 pairs were aggregated into 15 pairs and the resulting differences were then normalized . ( see also [101] ) . First , interregional pairs that showed statistically significant ( p<0 . 01 , FDR corrected ) increased or decreased functional connectivity in young-adults group compared to child group were identified as ( +1 ) or ( −1 ) , respectively . Second , the number of decreased ( −1 ) or increased connectivities ( +1 ) for each of the 15 pairs was counted . For example , to identify differential connectivity between the association division and the subcortical division , the number of decreased or increased connectivities between all pairs of subregions belonging to the association division and subcortical division was counted . Finally , since each brain region has a different number of subregions , the aggregated differential connectivity count was normalized by the number of possible connections between pairs of subregions belonging to the two divisions under investigation . Next we examined regional correlation values ( connectivity ) in the two groups . We compared regional correlation values aggregated across the 4 , 005 pairs of anatomical regions , between young-adults and children . No significant between-group differences in the aggregated correlation values were observed . On the basis of this observation , subsequently , individual regional correlation values were z-transformed followed by centering of the distribution around zero mean . These normalized correlation values were compared between the two subject groups . t-Test with a false discovery rate of 0 . 005 was used to test for significant differences . We next examined the relationship between differences in regional correlation values ( connectivity ) in the two groups and the interregional wiring distance as determined using DTI . The wiring distance between two regions was computed by measuring the average length of the fiber tracks , in the MNI space , connecting those regions ( see Text S1 for details ) . | The disruption of normal brain organization in humans is believed to underlie a number of behavioral conditions , such as autism spectrum disorders ( ASD ) and attention-deficit/hyperactivity disorder ( ADHD ) . To gain insight into how normal brain organization develops , we mapped functional brain connectivity in children and young adults , and used a network analysis to characterize and compare the organization of brain networks . Comparison of network properties revealed that while children and young adults' brains have similar organization at the global level , there were several key differences in connectivity . For example , children's brains had less of a hierarchical organization than young-adults . Most importantly , we show that the dynamic process of over-connectivity followed by pruning , which rewires connectivity at the neuronal level , also operates at the systems level , reconfiguring and rebalancing subcortical and paralimbic connectivity in the developing brain . Our findings demonstrate the utility of using network analyses of multimodal brain connectivity to study maturation of brain circuits , and suggest new avenues for future research on neurodevelopmental disorders such as ASD and ADHD . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"neuroscience/neurodevelopment",
"neuroscience/cognitive",
"neuroscience",
"radiology",
"and",
"medical",
"imaging/magnetic",
"resonance",
"imaging",
"computational",
"biology/computational",
"neuroscience"
]
| 2009 | Development of Large-Scale Functional Brain Networks in Children |
The phenotypic heterogeneity that characterizes human cancers reflects the enormous genetic complexity of the oncogenic process . This complexity can also be seen in mouse models where it is frequently observed that in addition to the initiating genetic alteration , the resulting tumor harbors additional , somatically acquired mutations that affect the tumor phenotype . To investigate the role of genetic interactions in the development of tumors , we have made use of the Eμ-myc model of pre-B and B cell lymphoma . Since various studies point to a functional interaction between Myc and the Rb/E2F pathway , we have investigated the role of E2F activities in the process of Myc-induced lymphomagenesis . Whereas the absence of E2F1 and E2F3 function has no impact on Myc-mediated tumor development , the absence of E2F2 substantially accelerates the time of tumor onset . Conversely , tumor development is delayed by the absence of E2F4 . The enhanced early onset of tumors seen in the absence of E2F2 coincides with an expansion of immature B lineage cells that are likely to be the target for Myc oncogenesis . In contrast , the absence of E2F4 mutes the response of the lineage to Myc and there is no expansion of immature B lineage cells . We also find that distinct types of tumors emerge from the Eμ-myc mice , distinguished by different patterns of gene expression , and that the relative proportions of these tumor types are affected by the absence of either E2F2 or E2F4 . From these results , we conclude that there are several populations of tumors that arise from the Eμ-myc model , reflecting distinct populations of cells that are susceptible to Myc-mediated oncogenesis and that the proportion of these cell populations is affected by the presence or absence of E2F activities .
A hallmark of human cancer is genetic complexity , reflecting the acquisition of multiple mutations and gene rearrangements that give rise to the tumor phenotype . Indeed , recent large-scale DNA sequencing efforts have provided direct evidence for this complexity , revealing large numbers of alterations that characterize various tumor types [1]–[4] . Undoubtedly , this genetic complexity of cancer underlies much of the challenge in developing effective therapeutic strategies . Not only is it likely that combinations of drugs will be necessary to match the complexity and effectively treat these tumors but equally important is the ability to identify subgroups of cancers that represent more homogeneous mechanisms of disease . An ability to model the complexity that gives rise to the tumor heterogeneity seen in human cancers would clearly enhance the understanding of the oncogenic process but also would enable the development and testing of combination therapeutics that might match this complexity . Mouse models of cancer have generally employed the use of an activated oncogene or the disruption of a tumor suppressor gene to initiate the oncogenic process . Although this represents a defined genetic alteration , it is also true that in most instances this single event is not sufficient to allow for tumor development . This can be seen in the often protracted latency of tumor development as well as the identification of specific additional genetic alterations that appear in these tumors . An example of a well-studied genetic model for the analysis of pre-B and B cell lymphoma is the Eμ-myc transgenic mouse . In the Eμ-myc transgenic mouse c-myc is constitutively expressed in the B lineage [5] , [6] . The resulting polyclonal expansion of pre-B cells is initially limited by increased apoptosis [7] . Additional mutations , many of which inactivate the p53 tumor suppressor pathway [8] , then arise . This leads to the emergence of a clonal pre-B or B cell lymphoma by six months of age in mice of a mixed C57Bl/6 and 129 strain background . Myc has been shown to induce a large number of genes that contribute to cell proliferation . These include the direct transcriptional activation of D cyclin genes , the cdk4 gene encoding the kinase partner for cyclin D , and the Cdc25A gene encoding the phosphatase that removes negative regulatory phosphates from the Cdks . The induction of Cyclin D/cdk4 activity leads directly to the phosphorylation of Rb and thus activation of E2Fs . Numerous studies have demonstrated a central role for the Rb-E2F pathway in the regulation of cellular proliferation . The majority of genes encoding DNA replication and mitotic activities are under the control of E2F proteins . Indeed , recent experiments provide evidence for a role for E2Fs in coordinating transcriptional regulatory events at G1/S and G2/M [9]–[11] . Other work has shown that E2Fs also link this critical proliferative pathway with the p53 response through a capacity to induce the p19ARF/Mdm2 pathway leading to the accumulation of p53 protein [12] , [13] . As such , E2Fs provide a mechanism to directly link the control of cell proliferation with the determination of cell fate . In addition to the connection between Myc and E2F in the control of cellular proliferation , Myc expression couples cellular proliferation with the induction of apoptosis under specific growth conditions where survival growth factors are limiting . Myc-induced apoptosis is largely dependent upon p53 signaling and , similar to E2F1 , involves the induction of p19ARF , inhibition of Mdm2 , and elevated p53 [14] . The shared functional properties of the Myc and E2F transcription factors , coupled with the finding that Myc can induce E2F gene expression [15] , [16] , raise the possibility that Myc function might be mediated , at least in part , through the action of the E2F transcription factors . Indeed , work by the Bernards laboratory revealed that in addition to targeting p27Kip1 , the mitogenic activity of Myc likely involves regulation of E2Fs [17] . This possibility has been more directly assessed using mouse embryo fibroblasts ( MEFs ) from embryos deleted for specific E2F genes to evaluate the functional relationship between Myc and various E2F proteins [18] . Experiments using these E2F-deficient MEFs showed that the ability of Myc to induce S phase in the absence of other mitogens is severely impaired in MEFs deleted for E2f2 or E2f3 , but not E2f1 or E2f4 . In contrast , Myc induced apoptosis in primary serum-deprived MEFs was delayed in cells deleted for E2f1 , but not affected by E2f2 or E2f3 deletion . Thus , at least in cell culture , the induction of specific E2F activities is an essential downstream event in the Myc pathway that controls cell proliferation versus apoptosis , and some of the functions of Myc , such as the induction of p19ARF and p53 could be explained , at least in part , with one pathway leading through E2F activation . To address the significance of the Myc-E2F connection in a relevant , in vivo setting , we have made use of a series of E2F-deficient mouse strains , in combination with the Eμ-myc transgenic model of lymphomagenesis ( MGI:2448238 ) , to investigate whether deficiencies in E2F1 , E2F2 , E2F3 or E2F4 ( MGI:1857424 , 2179111 , 2177428 and 1888951 ) can influence Myc's oncogenic potential . We find that there is a critical role for two E2F activities in affecting the potential for Myc-induced oncogenesis .
Male Eμ-myc transgenic mice ( backcrossed and maintained in the C57BL/6 strain ) were bred into the four different E2F-deficient mouse lines . E2F2 , E2F3 and E2F4 cohorts were maintained as C57BL/6×129 while the E2F1 cohort was predominantly C57BL/6 in background . In particular , the E2F3 and E2F4 cohorts required maintenance on a mixed , rather than inbred , background because the yield of the E2f3-null and E2f4-null mice was severely compromised upon inbreeding ( data not shown ) . Sibling mice , wild type , heterozygous or null for a particular E2F gene , and bearing the Eμ-myc transgene , were examined weekly for any sign of lymphoma emergence . Each mouse was checked for enlarged lymph nodes , a swollen abdomen , a hunched posture , ruffled fur and/or tachypnea [22] . Upon the appearance of any of these symptoms , the mouse was sacrificed , dissected to identify any lymph node enlargement , and tumor tissue harvested for analysis . For studies assessing the pre-tumor phenotype , mice were sacrificed within three to five weeks after birth and bone marrow and spleen recovered . Such samples were characterized as pre-neoplastic only if lymph node and spleen enlargement was nil or modest at the time of dissection and/or Southern analysis of B lineage cell DNA revealed no specific heavy chain rearrangements indicative of the emergence of tumor clones . Tumor emergence was evaluated in the four E2F cohorts ( Figure 1 ) , as well as for our Eμ-myc C57BL/6 congenic stock mice ( Figure S2A ) . When the wild type mice in each cohort were compared to the stock transgenic mice , median onsets did diverge ( Figure S2B ) , with earlier onsets associated with greater 129 strain contribution based on breeding history . In spite of this , the overall appearance of each of the wild type curves was similar , with some mice succumbing early and others succumbing late . As shown in Figure 1A , the loss of E2F1 function did not alter the timing of lymphoma appearance; there was no statistical difference between tumor onset curves when comparing E2f1+/+ ( n = 37 ) , E2f1+/− ( n = 91 ) and E2f1−/− ( n = 40 ) mice . The failure of E2F1 status to influence lymphomagenesis conflicts with the earlier finding that E2F1 deficiency delays lymphoma development in Eμ-myc mice [23] . That study attributed the delay to a defect in p27Kip1 degradation and reduced Myc-induced proliferation when E2F1 is reduced or absent . In our assessments the level of p27Kip1 protein in splenic B lineage cells did not vary according E2F1 status but rather with progression to disease: p27Kip1 was highest in cells isolated from non-transgenic siblings , reduced in healthy Eμ-myc transgenics , and lower still in very sick mice and tumors ( Figure S3A ) . In addition , the accelerated proliferation induced by expression of the Eμ-myc transgene [24] was unaffected by E2F1 deficiency: splenic B lineage cells isolated from E2f1 wild type , heterozygous and null Eμ-myc transgenic mice all exhibited the same dramatically higher proliferative index when compared to that of cells isolated from non-transgenic siblings ( Figure S3B ) . We note that , analogous to our results , E2F1 deficiency did not alter Myc-induced T cell lymphomagenesis [25] . Given that the timing of Eμ-myc-driven tumor development can be influenced by strain background [26] and breeding strategy , we can only surmise that these potential differences or the specific E2f1-null allele [27] , [28] used in our studies versus that of Baudino and colleagues are sufficient to account for the discrepant effects of E2F1 deficiency . As shown in Figure 1B , lymphoma onset was also not appreciably influenced by E2F3 status . While the number of E2f3-null animals was low in this study , reflecting the low number of E2f3-null mice born , there is nevertheless no suggestion that E2F3 loss was protective as several E2f3-null mice died before the average age of onset for their wild type siblings . In contrast to the results seen with the E2f1 and E2f3 knockout animals , a deficiency of E2F2 dramatically accelerated the appearance of lymphoma ( Figure 1C ) . The E2f2−/− mice were prone to early tumor onset with tumors appearing on average 60 days earlier than in their wild-type siblings and there was a significant difference ( p<0 . 0001 ) between the E2f2+/+ and E2f2−/− tumor curves . Notably , the E2f2+/− mice exhibited a median tumor-free span of 92 days and a significantly accelerated course of disease compared to wild-type siblings ( p<0 . 0001 ) . The intermediate phenotype of the E2f2 heterozygotes suggests a degree of haploinsufficiency . In addition , the E2f2+/− lymphomas showed no loss of heterozygosity demonstrating that E2F2 does not behave like a classic tumor suppressor in the Eμ-myc context ( data not shown ) . Finally , a deficiency in E2F4 also had a dramatic effect on Myc-induced lymphoma development ( Figure 1D ) . A comparison of E2f4+/+ , E2f4+/− and E2f4−/− mice revealed that E2f4−/− mice remained tumor-free for significantly longer than siblings ( p<0 . 0001 ) with a median tumor-free span of 375 days past birth . Lymphoma onset may be modestly delayed in E2f4+/− mice ( p = 0 . 0465 ) . Taken together , these data would suggest roles for two E2F proteins , both positive and negative acting , in affecting the onset of Myc-mediated lymphomas . A role for these two E2F family members also coincides with the prominent expression of these proteins in hematopoetic tissues ( Figure S1 ) . Previous work has shown that in the B cell lineage Myc induces proliferation and apoptosis and retards differentiation [24] , [29] . As such , the effects of E2F loss of function on Myc oncogenesis could result from alterations in one or more of these processes . To address the potential for differential effects on proliferation , weanling mice were injected with BrdU , three hours later bone marrow was isolated , and the cell cycle distribution of B lineage cells ( B220+ CD19+ ) determined . As shown in Figure 2A , proliferation of B lineage cells was similar for non-transgenic E2f2 wild type and null cells . Importantly , the effect of Myc on cell cycle entry of B lineage cells , with increased S-phase cells and reduced G0/G1 cells , was independent of E2F2 status . As shown in Figure 2B , E2f4 wild type and null B lineage cells were similarly proliferative in non-transgenic mice . As well there was still the expected increase in proliferation associated with bearing the Eμ-myc transgene for E2f4-null mice ( E2f4−/− compared to E2f4−/− Eμ-myc+: p = 0 . 0002 ) ( Figure 2B ) . It appears that the acceleration of cell cycle progression driven by Myc in B lineage cells is not significantly affected by the loss of E2F2 or E2F4 activities . An alteration in the apoptotic potential of Myc could account for the differences in Myc-initiated tumor onset among wild type , E2f2-null and E2f4-null animals . Possibly in E2f2−/− mice apoptosis is reduced whereas in E2f4−/− mice apoptosis is potentiated . Freshly isolated bone marrow B lineage cells ( B220+ ) from E2f2−/− Eμ-myc transgenics and E2f4−/− Eμ-myc transgenics were found to have comparable percentages of activated caspase-3 positive cells as their wild type and heterozygous Eμ-myc siblings , around 0 . 6% , and all the non-transgenic siblings had less than half this percentage of apoptotic cells ( data not shown ) . Since divergent viability may be masked by clearance in vivo , the survival of B lineage cells under culture conditions where deregulated Myc induces apoptosis was assessed [30] , [31] . The bulk of B lineage cells , excepting progenitors upstream of small pre-B cells , was enriched by negative selection from the bone marrow , spleen and mesenteric lymph node . The resulting population of small pre-B cells , immature B cells and mature B cells from each mouse was cultured for eight hours in medium lacking cytokines . The cultured cells were sampled at two-hour intervals , and B220+ cells assessed for viability based on activated caspase 3 and 7-AAD staining ( Figure S4 ) . As expected , the decline in viability was faster for cells from Eμ-myc transgenics than from non-transgenics ( Figure S4 and Figure 3A and 3B ) . Notably , when mice in the E2F2 cohort were compared , cells from E2f2−/− Eμ-myc mice lost viability to a similar extent over eight hours as cells from wild type and heterozygous siblings ( Figure 3A ) . Likewise , in an experiment assessing E2F4 cohort mice , cells from E2f4−/− Eμ-myc mice declined in viability similarly to those from their wild type and heterozygous siblings ( Figure 3B ) . Thus , the faster tumor onset for E2f2-deficient mice appears not attributable to a general apoptotic deficiency and the slower tumor onset for E2f4-null mice unlikely the result of increased apoptosis , at least in the small pre-B cells and more mature stages assessed here that constitute the large majority of B lineage cells . Myc-mediated tumor emergence is almost invariably associated with a disabling of the ARF-p53 tumor suppressor pathway [8] , [14] . Mutations that are indicative of pathway disruption include: p53 deletion , ARF deletion , or overexpression of ARF , Mdm2 or mutant p53 . Examples of these disruptions in a sampling of tumors are shown in Figures S5 , S6 , and S7 . Consistent with the earlier studies , tumors from E2F wild type Eμ-myc mice showed evidence of disruption of the ARF-Mdm2-p53 pathway ( Figure 3C ) . The largely late-onset lymphomas from E2f4-null mice also demonstrated disruptions in this pathway . Importantly , the spectrum and overall incidence of defects in the E2f2+/− and E2f2−/− lymphomas were very similar to that shown by the E2f2+/+ lymphomas . In contrast , other modifiers of Myc-induced lymphomagenesis such as Bim and Bax relieve or modify the strong selective pressure for functional inactivation of this pathway [32] , [33] . That loss of ARF-p53 function was still associated with development of tumors in the E2f2−/− mice further supports the evidence that E2F2 deficiency does not compromise Myc-induced apoptosis ( Figure 3A ) . Given that E2F2 deficiency does not alter the proliferation or apoptosis of pretumorous B lineage cells in response to Myc , we focused on the possibility that there may be a different underlying mechanism driving the accelerated tumor emergence , one involving development of the B lineage and the response to Myc . As noted in one of the earliest descriptions of the Eμ-myc model , it is possible that different onsets could reflect different extents of lineage expansion in response to Myc and therefore numbers of vulnerable cells [24] . Overall viable white blood cell number in the bone marrow did not change with E2F2 status while Eμ-myc positive mice had modestly higher counts ( data not shown ) . As expected , the B cell lineage expanded as a proportion of the bone marrow in response to the Eμ-myc transgene in E2f2+/+ mice ( Figure 4A ) and the expansion favored less mature over more mature B cells across all three genotypes ( Figure 4B ) . Importantly , there was a similar degree of expansion for the E2f2+/− and E2f2−/− mice . Recent studies have suggested , however , that lymphomagenesis likely initiates in B lineage progenitor cells making the effects of various mutations on progenitor populations particularly relevant [34] . For instance , the Eμ-myc bcl2−/− mice develop tumors at the same rate as Eμ-myc bcl2+/+ mice despite decreased pre-B , immature B and mature B lymphocytes; significantly , they do have similar numbers of pro-B cells [34] . Similarly , Eμ-myc/max41 mice develop lymphoma almost as quickly at Eμ-myc mice despite a severe deficit in more mature , peripheral B cells [35] . Conversely , the increased progenitor population of early/large pre-B stage cells exhibited by Phospholipase Cγ2-deficient mice is associated with accelerated lymphomagenesis [36] . As shown in Figure 4C , CFU pre-B colony assays using bone marrow from non-transgenic E2F2 cohort mice indicate that there were more B lineage progenitors in E2f2−/− marrow ( P<0 . 0001 ) and E2f2+/− marrow ( P = 0 . 0406 ) than wild type marrow . The proportion of early B lineage cells ( B220+ CD43+ ) identified by flow cytometry was also greater in non-transgenic E2f2−/− marrow than in non-transgenic E2f2+/+ marrow ( P = 0 . 0008; Figure 4D ) . The increased proportions of progenitors in E2f2-deficient mice extended into pre-tumorous Eμ-myc positive mice: there was the trend , although not statistically significant , for more immature B lineage cells as a proportion of the total bone marrow in E2f2−/− ( P = 0 . 0769 ) and E2f2+/− Eμ-myc transgenics ( P = 0 . 1393 ) compared to E2f2+/+ Eμ-myc transgenic mice ( Figure 4D ) . Additional analysis revealed that a significant proportion of the E2f2−/− Eμ-myc lymphomas were not monoclonal . Assessment of Igh locus rearrangement patterns by Southern analysis indicated that almost 40% of the E2f2−/− tumors were biclonal or oligoclonal whereas tumors from other genotypes , whether in the E2F2 cohort or in other cohorts , were predominantly monoclonal , in agreement with past studies [5] ( Figure 5A and Figure S8A and S8B ) . This degree of oligoclonality was , however , less extensive than that which occurs with the homozygous Eμ-myc/Eμ-myc mice [37] and when retrovirally-expressed myc was expressed in mice reconstituted with p53−/− hematopoietic stem cells [38] . In addition , several tumors were analyzed by flow cytometry for isotypic surface marker expression ( Figure 5B ) . Seven out of ten tumors emerging from E2f2−/− mice displayed a complex pattern of isotypic surface markers . Such complexity , while not uncommon for Eμ-myc lymphomas in general [22] , is consistent with multiple clones . Of note , two out of eight tumors from E2f2+/− mice were similarly complex . In contrast , the majority of tumors from other Eμ-myc mice displayed single , homogeneous patterns of surface markers and could be clearly classified as being either pro/pre-B or immature B lymphomas . Taken together , these findings support the conclusion that in the E2f2−/− Eμ-myc mice there is an increased population of B lineage cells susceptible to lymphomagenesis resulting in the occasional emergence of more than one independent tumor . Using a different Myc transgenic system , Cory and colleagues noted the emergence of mixed T cell tumors in their study and concluded that such mixed tumors originated as separate clones and could be expected with a high rate of tumorigenesis [39] . Also consistent with the hypothesis that the enhancement of early onset tumors in E2f2−/− mice is the consequence a larger pool of susceptible cells , rather than of differently behaving cells , is the finding that the lymphomas that emerged were very similar to lymphomas that arose early in E2f2+/+ Eμ-myc mice . For instance , when mice showing signs of illness were sacrificed the degree of splenomegaly was comparable ( Figure S9A ) and the histopathology of the lymphomas was similar . The majority of early-onset tumors , either from E2f2+/+ or E2f2−/− mice , featured high mitotic indices and extensive apoptosis with tingible body macrophages and a starry sky appearance similar to that of human Burkitt lymphoma ( data not shown ) . In addition , B lineage cells ( B220+ ) isolated from E2f2+/+ and E2f2−/− lymphomas exhibited comparable rates of proliferation and apoptosis ( Figure S9B and S9C ) . It has recently been shown that E2f4-deficient mice have defects that extend from early hematopoietic progenitor cells , through common lymphoid precursors and into the B and T lineages [40] . Specifically in the B lineage , E2f4−/− mice exhibit a partial block early in B lineage development prior to immunoglobulin gene rearrangement that results in a deficiency in the most mature pro-B subpopulation and a reduction in more mature B lineage cells [19] ( Glozak et al . , manuscript in preparation ) . Total viable white blood cell counts in the bone marrow were modestly higher for Eμ-myc E2f4+/+ and E2f4+/− mice than for non-transgenic siblings . In the case of E2f4−/− mice , there was no increase associated with the Eμ-myc transgene and Eμ-myc E2f4−/− mice had about half as many cells as Eμ-myc wild type siblings ( p = 0 . 0269 ) . As expected , within the bone marrow , non-transgenic E2f4−/− mice had a lower proportion of B lineage cells compared to non-transgenic E2f4+/+ mice ( P = 0 . 0133 , Figure 6A ) . Strikingly , the usual expansion of the B lineage in response to the myc transgene failed to occur in the E2f4−/− mice . As a consequence , the proportion of B lineage cells in the bone marrow was significantly less in E2f4−/− Eμ-myc mice than in E2f4+/+ Eμ-myc mice ( P = 0 . 0019 ) . Myc did , however , elicit the usual reduction in the relative proportion of mature to less mature B lineage cells in the E2f4-deficient mice as in E2f4 wild type mice ( Figure 6B ) . Motivated by the hypothesis that Eμ-myc lymphomas originate in early stage B lineage cells [34] , we assessed progenitor populations in the E2f4−/− mice compared to their siblings . In CFU pre-B colony assays using bone marrow from non-transgenic E2F4 cohort mice , there were significantly fewer progenitors in E2f4−/− marrow than wild type marrow ( data not shown; Glozak et al . , manuscript in preparation ) . Notably , the pre-tumorous E2f4−/− Eμ-myc mice exhibited no expansion of immature B lineage cells ( B220+ CD43+ ) as a proportion of the total bone marrow compared to their non-transgenic E2f4−/− siblings ( Figure 6C ) . We suggest that the Myc transgene fails to overcome the inefficient developmental progression of the B lineage in E2f4-deficient mice , there is a reduced number of susceptible progenitor cells , and consequently delayed tumor emergence . As indicated by Figure 1D , E2f4−/− Eμ-myc mice displayed a much delayed tumor onset . Possibly because of this delay , of the twenty-six mice assessed , nine died before lymphoma emergence or were still healthy at analysis . The three mice that developed lymphoma early ( within 150 days of birth ) and one older mouse exhibited the standard Eμ-myc lymphoma phenotype , characterized by an enlarged spleen and multiple enlarged lymph nodes . Thirteen mice developed lymphoma very late in life . Three of these mice exhibited lymphoma with modest spleen enlargement and isolated lymph node enlargement , similar the uncommon late onset lymphomas that occasionally develop in Eμ-myc mice wild type for E2F . The ten remaining E2f4−/− Eμ-myc mice displayed an atypical tumor phenotype that was only rarely noted in E2F wild type Eμ-myc mice ( 10 of 17 E2f4−/− mice compared to only 3 of 79 E2f4+/+ mice ) . These atypical tumors featured a loose tumor mass of multiple small nodules in the mediastine with little or no associated spleen or peripheral lymph node enlargement . Despite their unusual appearance , the atypical tumors that were tested demonstrated Igh gene rearrangement confirming their B lymphoid origin . Overall , eleven E2f4−/− tumors , including examples of standard , late , and atypical types , were assessed for clonality and all proved to be monoclonal . In summary , along with the general delay in tumor onset there was also a difference in the predominant site of lymphomagenesis and gross morphological appearance of tumors in the E2f4−/− Eμ-myc mice . As a further basis for exploring the effects of E2F loss of function on the development of Myc-induced tumors , we have made use of genome-scale gene expression profiles to characterize the tumors arising in the E2f2-null and E2f4-null Eμ-myc mice . Our recent work has identified expression profiles that distinguish different tumor types within the Eμ-myc mice including a cluster characterized by generally early onset and pre-B markers as well as three distinct clusters characterized by late onset and different sets of more differentiated B lineage markers [41] . Examples of wild type tumors exhibiting this early and late onset pattern are shown in Figure 7A . Analysis of the tumors from the E2f2−/− mice indicated that they were relatively homogeneous with respect to their expression profiles and reflected the characteristics of the “early” category of wild type tumors . In contrast , the tumor types from the E2f4−/− mice were heterogeneous with a distribution across both broad categories of the wild type tumors . The distribution of the E2f4−/− tumors corresponded with their dissection phenotypes - standard , late and atypical - described above . Three early-onset E2f4−/− tumors , all with the standard morphological phenotype , clustered with the “early” wild type and E2f2−/− tumors . Three more E2f4−/− tumors , all with the late morphological phenotype , clustered alongside a group of wild type tumors that overexpress genes characteristic of plasmacytomas . These particular E2f4−/− tumors shared marginally decreased myc mRNA and low Myc protein compared to other E2f4−/− tumors ( data not shown ) . The final six E2f4−/− tumors were all of the atypical phenotype and segregated in the “early” category despite being late onset chronologically . These tumors shared qualities with the “early-standard” tumors in that they featured high levels of myc mRNA and Myc protein ( data not shown ) . Notably , these tumors fell at the extreme end of the early cluster and beside a rare group of wild type tumors that had similarly modest spleen enlargement and late chronological onset . In fact , these tumors highlight a significant subgroup within the “early” category that we have designated “early-atypical” . This tumor subgroup was notable for a high incidence of p53 deletion or mutation ( 67% of tumors assessed versus 18% of other wild type tumors assessed; by Fisher's exact test P = 0 . 0061 ) . Among the genes that characterized each tumor cluster , increased expression of number of genes distinguished this subgroup from both “early-standard” and “late” clusters ( Figure S10 ) . Cdkn2a , the locus that encodes the two tumor suppressors p16 ( INK4a ) and p19 ( ARF ) , was preferentially expressed in these tumors . Given that most of these tumors were mutant for p53 , the increased expression may be a consequence of a role for p53 in negatively regulating the expression of ARF [42] . Other genes that were particularly highly expressed in the “early-atypical” tumors included Dlk1 , a member of the epidermal growth factor-like family that influences B lineage differentiation [43] , [44] and Fzd6 , a receptor for Wnt signaling and a frizzled family member [45] . To further characterize the distinctions in the Eμ-myc tumors , we have made use of signatures of various cell signaling pathways that have previously been shown to distinguish human Burkitt lymphoma ( BL ) from diffuse large cell B lymphoma ( DLBCL ) [46] . These include signatures for Myc pathway activity , the expression of a subgroup of germinal-center B-cell genes , the expression of MHC class I genes , and NFκB pathway activity . An analysis of tumors from the E2f2-null and E2f4-null mice using these pathway signatures is shown in Figure 7B . Consistent with the analysis of whole genome gene expression data in Figure 7A that revealed distinct types of B lymphoma , the analysis using pathway signatures also revealed that the tumors from the E2f2-null mice exhibited a pattern similar to the “early” tumors , characterized by high Myc and germinal center signatures , whereas the tumors from E2f4-null mice were heterogeneous with respect to these patterns . The atypical E2f4−/− tumors did not feature the usual “late” characteristics but instead had elevated Myc and germinal center signatures and low MHC Class I and NF-KB signatures . These tumors highlight the existence of certain tumors of late chronological onset , whether E2F wild type or E2f4−/− , that were unusual in their behavior .
Our results indicate that E2f2 deficiency enhances the emergence of the “early-standard” form of lymphoma likely because the absence of E2F2 activity expands the population of cells that is the usual target for the oncogenic process in the Eμ-myc model . As a result , the population of E2f2−/− tumors is also more homogeneous with respect to their phenotype , as reflected by the gene expression profiles . In contrast , we propose that the loss of E2f4 results in the decrease of this population of cells and thus the frequency of appearance of standard morphology early onset tumors . There is not a complete absence of these cells since a few tumors do arise in the absence of E2F4 that cluster with the “early-standard” wild type and E2f2−/− tumors . But the consequence of this depletion is enrichment for tumors with a late chronological onset , whether to the extreme of the “early” cluster or in the “late” cluster , likely due to an opportunity for these tumors to develop because of the reduced frequency of the “early-standard” variety . Taken together , these results point to a role for E2F activities in determining the population of B lineage cells that contribute to the development of tumors and highlights the interplay between two cell regulatory activities , E2F and Myc , in determining the outcome of the oncogenic process .
Mice were housed in a Duke University Medical Center Division of Laboratory Animal Resources facility and experiments approved by the Duke University Institutional Animal Care and Use Committee . The generation of the specific lines of E2F-deficient mice has been previously described [19] , [28] , [53] . The original 129 substrain background was 129/SvJae for the E2F1 , E2F2 and E2F3 cohorts and 129/OlaHsd for the E2F4 cohort . Based on breeding history , the E2F1 cohort mice used in this study were predominantly C57BL/6 ( backcrossed five generations into C57BL/6 ) while the E2F2 , E2F3 and E2F4 cohorts were mixed C57BL/6×129 . The four E2F cohorts were maintained separately and breeding involved crossing heterozygous mice to yield wild type , heterozygous and null mice in each generation . The Eμ-myc transgenic mouse line 292-1 [5] extensively backcrossed into C57BL/6 and originally from Dr . Alan Harris ( Walter and Eliza Hall Institute , Melbourne , Australia ) , was kindly provided by Dr . Scott Lowe ( Cold Spring Harbor Laboratory ) . For each E2F cohort , stock Eμ-myc positive C57BL/6 congenic males were bred to E2Fn+/− females and of the progeny only the Eμ-myc positive E2fn+/− males , designated the F1 males , were kept . These F1 males , E2fn+/− myc+ , were then bred to E2fn+/− myc− females . The Eμ-myc positive progeny of this cross , E2fn+/+ , E2fn+/− , and E2fn−/− were then compared . Because maternal transmission is associated with reduced latency [34] , transmission of the Eμ-myc transgene was exclusively paternal in this breeding scheme . Eμ-myc negative siblings were also kept as a source of related mice that lacked myc transgene effects . Eμ-myc positive mice were monitored weekly to identify any mice with malignant disease . Mice were evaluated for any visible or palpable lumps , a hunched posture , tachypnea , a swollen belly , or ruffled fur and sacrificed promptly upon the appearance of any such symptoms . Lymphomas that emerged were dissected from sacrificed mice , washed in PBS , and frozen in liquid nitrogen or processed for flow cytometric analysis . The frozen tissue provided material for Southern and western analysis . Tumor onset data refer to the time in days between birth and the first indication of illness . Using GraphPad's Prism program , the time values were plotted to generate Kaplan-Meier survival curves and the curves compared by a logrank test . For comparisons of means and standard deviations , the paired student t-test was performed and statistical significance was determined if the p<0 . 05 . To assess alterations in p19ARF , Mdm2 and p53 protein expression , lymphoma samples were dissected from morbid mice and immediately frozen . Samples were then weighed , ground to a powder in liquid nitrogen and resuspended in 60 mM Tris ( pH 6 . 8 ) /1% SDS at 1 ml per 0 . 2 g tumor weight , boiled , sonicated , and any remaining debris removed by centrifugation . In parallel , whole cell extracts were made from mouse embryonic fibroblasts infected with the indicated adenoviruses for controls . Protein was quantitated using the BCA Protein Assay Reagent Kit ( Pierce ) . Samples ( 150 µg ) were boiled in sample buffer and subjected to SDS-PAGE on 8 . 5% polyacrylamide gels for p53 and Mdm2 assessment and 15% gels for ARF assessment . Western analysis was performed as previously described [54] . The blots were probed with antibodies specific for p53 ( monoclonal antibody Ab-1 OP03 at 1∶1000 , Calbiochem ) , p19ARF ( polyclonal antibody Ab-1 PC435 at 1∶10 , 000 , Calbiochem ) , and Mdm2 ( polyclonal antibody C-18 sc812 at 1∶1000 , Santa Cruz Biotechnology ) . Equal protein loading was verified by staining blots with Ponceau Red ( 0 . 2% ponceau red in 3% trichloroacetic acid ) . Genomic DNA was isolated from lymphomas , normal spleen cells , tail samples and MEFs of specified genotypes . DNA ( 10 µg ) was digested with BamH1 ( for the p53 locus ) , AflII ( for the p19ARF locus ) or EcoRI ( for the heavy chain locus ) . The restricted DNA was separated by agarose gel electrophoresis ( 0 . 8% gels ) , transferred to Hybond N+ membrane , and probed . The p53 probe was a human cDNA fragment ( 686 base pair DrdI-StuI fragment extending from exon 4 to exon 10 ) . The ARF probe was the exon 1B portion of the ARF cDNA ( kindly provided by Charles Sherr ) . The heavy chain locus probe was the heavy chain J3-J4 joining region genomic fragment [37] . On occasion , to verify that weanling mice were essentially tumor-free , genomic DNA isolated from B lineage cells was assessed by Southern analysis for any emergence of clonal heavy chain rearrangements . Mononuclear cells were harvested from the bone marrow of 3-week-old littermates and from lymphomas that arose . Cells were stained with various combinations of antibodies to IgD ( 11-26c . 2a ) , IgM ( R6-60 . 2 ) , CD19 ( 1D3 ) , B220 ( RA3-6B2 ) , CD43 ( S7 ) , BrdU , and active caspase-3 . All antibodies and staining reagents were from BD Biosciences . Cell staining procedures were performed either manually or using a Biomek 2000 robotic fluid handler ( Beckman Instruments , Schaumburg , IL using a series of mini-programs developed with BioWorks software ( Beckman Instruments ) . FACS analysis was performed on a FACSCalibur device equipped with a 488 nm argon laser and a ∼635 nm red dye laser ( Becton Dickinson ( BD ) , San Jose , CA ) . Data was analyzed using FlowJo Software ( TreeStar , Palo Alto , Ca ) . Three hours before analysis , mice were injected with 100 mg/kg BrdU . Bone marrow mononuclear cells were collected and stained with the B220 and CD19 antibodies to identify B lineage cells , with 7-AAD , and with anti-BrdU antibodies . BrdU Flow Kit reagents and directions were followed ( BD/Pharmingen ) . The proportion of cells that had proceeded through S-phase , or resided in G0/G1 or in G2/M phases was determined . Hematopoietic cells were harvested from the bone marrow , spleen and mesenteric lymph node , combined and enriched for B lineage cells using negative selection ( SpinSep Mouse B Cell Enrichment Cocktail , Stem Cell Technologies ) . The antibodies used to label unwanted cell types for depletion were directed against CD4 , CD8 , CD11b , CD49b , Gr-1 , TER119 and CD43 . The approach yielded a subset of B lineage cells from the small pre-B stage through more mature stages . These cells were cultured at 37°C for eight hours in DMEM plus 10%FCS/100 µM L-aspargine/50 µM 2-mercaptoethanol at a concentration of 4×106 cells/ml . At selected time points cells were removed and stained with 7-AAD and B220 antibody , fixed and permeabilized , stained with activated caspase-3 antibody and analyzed by flow cytometry . Viable cells were negative for both 7-AAD and activated caspase-3 . Equivalent numbers of bone marrow cells from non-transgenic 4–6 week old mice were resuspended in Methocult M3630 ( Stemcell Technologies ) according to manufacturer's specifications to assay for pre-B cell colonies . This media , formulated for the detection and counting of mouse pre-B progenitors in bone marrow , is comprised of methylcellulose in Iscove's MDM supplemented with recombinant IL-7 , 2-Mercaptoethanol , L-glutamine , and fetal bovine serum . All samples were assayed in duplicate . After seven days colonies were counted using an inverted microscope . The count was based on the manufacturer's description of expected colony appearance - namely that colonies are composed of at least 30 cells and that individual cells are tiny and irregular to oval in shape . RNA was extracted from lymphoma samples using Qiagen RNeasy Kits ( Qiagen ) . RNA sample integrity was verified by agarose gel electrophoresis or using an Agilent 2100 Bioanalyser . We prepared the targets for DNA microarray analysis and hybridized to Affymetrix Mouse 430 2 . 0 GeneChip arrays according to the manufacturer's instructions and as previously published . To allow merging of expression array results from samples arrayed independently , some duplicate samples were arrayed to provide reference samples and the expression values standardized using ComBat [55] . The method for cross-platform comparison across different versions of Affymetrix GeneChip arrays was described previously [56] . Hierarchical clustering and visualization were performed using Gene Cluster 3 . 0 ( http://bonsai . ims . u-tokyo . ac . jp/~mdehoon/software/cluster/ ) and Java TreeView ( http://jtreeview . sourceforge . net/ ) . Genes and tumors were clustered by average linkage using uncentered correlation as the similarity metric . Analysis of expression data was described previously [56] . In summary , we collected training sets consisting of gene expression values of samples where the pathway activity was known . We created gene expression signatures by choosing the genes whose expression profiles across the training samples most highly correlated with the phenotype . Then , to predict the status of the phenotype on a tumor expression dataset , we fit a Bayesian probit regression model that assigned the probability that a tumor sample exhibited evidence of the phenotype , based on the concordance of its gene expression values with the signature . The Supporting Materials and Methods are available in Text S1 . | The diversity of human cancers reflects the variety of genetic changes that cause tumors to emerge and progress . Even for mice engineered with a specific cancer-causing mutation , the resulting tumors are often divergent , reflecting different additional mutations . We wanted to investigate how activities that work together can collaborate in tumorigenesis . Specifically , we are interested in Myc and the E2F family of proteins , intersecting activities that influence a cell's decision to replicate , rest , or die . We made use of an engineered mouse that develops pre-B and B cell lymphoma initiated by Myc and tested whether the loss of particular E2F family members influences these lymphomas . We found that tumor emergence was accelerated by E2F2 loss and delayed by E2F4 loss . We attributed these results to the finding that the mice lacking E2F2 have a greater proportion than usual of the most susceptible , early-stage B lineage cells and the mice lacking E2F4 have fewer of these cells . Distinct tumor types emerged with their relative proportions influenced by E2F2 and E2F4 status . We conclude that the variety of tumors probably reflect different stages of B lymphoid development that respond to Myc and that E2F proteins can influence the proportions of these different stages . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"and",
"genomics/disease",
"models",
"genetics",
"and",
"genomics/cancer",
"genetics",
"genetics",
"and",
"genomics/gene",
"expression"
]
| 2009 | A Role for E2F Activities in Determining the Fate of Myc-Induced Lymphomagenesis |
Proteasomes recognize and degrade poly-ubiquitinylated proteins . In infectious disease , cells activated by interferons ( IFNs ) express three unique catalytic subunits β1i/LMP2 , β2i/MECL-1 and β5i/LMP7 forming an alternative proteasome isoform , the immunoproteasome ( IP ) . The in vivo function of IPs in pathogen-induced inflammation is still a matter of controversy . IPs were mainly associated with MHC class I antigen processing . However , recent findings pointed to a more general function of IPs in response to cytokine stress . Here , we report on the role of IPs in acute coxsackievirus B3 ( CVB3 ) myocarditis reflecting one of the most common viral disease entities among young people . Despite identical viral load in both control and IP-deficient mice , IP-deficiency was associated with severe acute heart muscle injury reflected by large foci of inflammatory lesions and severe myocardial tissue damage . Exacerbation of acute heart muscle injury in this host was ascribed to disequilibrium in protein homeostasis in viral heart disease as indicated by the detection of increased proteotoxic stress in cytokine-challenged cardiomyocytes and inflammatory cells from IP-deficient mice . In fact , due to IP-dependent removal of poly-ubiquitinylated protein aggregates in the injured myocardium IPs protected CVB3-challenged mice from oxidant-protein damage . Impaired NFκB activation in IP-deficient cardiomyocytes and inflammatory cells and proteotoxic stress in combination with severe inflammation in CVB3-challenged hearts from IP-deficient mice potentiated apoptotic cell death in this host , thus exacerbating acute tissue damage . Adoptive T cell transfer studies in IP-deficient mice are in agreement with data pointing towards an effective CD8 T cell immune . This study therefore demonstrates that IP formation primarily protects the target organ of CVB3 infection from excessive inflammatory tissue damage in a virus-induced proinflammatory cytokine milieu .
Unfolded or misfolded proteins are potentially harmful to cells and have to be efficiently eliminated before they intoxicate the intracellular environment . This is of particular importance during proteotoxic stress as a consequence of intrinsic or extrinsic factors when the levels of misfolded proteins are transiently or persistently elevated ( Dantuma , 2010 #1 ) . In viral infection cytokine exposure and inflammation induce the generation of reactive oxygen species in both immune and non-immune cells [1] , [2] with concomitant oxidant-protein damage and proteotoxic stress . An important defence mechanism is the specific destruction of these proteins by the ubiquitin-proteasome system ( UPS ) [3] . The UPS is among others involved in the regulation of protein quality control in cardiovascular pathologies [4]–[6] , in neurodegenerative disorders and other human pathologies [7] , [8] . The UPS with the 26S proteasome as central proteolytic unit represents the major ATP-dependent degradation system in eukaryotes responsible for the maintenance of protein homeostasis and the generation of the vast majority of antigenic peptides that are presented by MHC class I molecules to CD8+ T cells in infectious disease [9] . Short-lived regulatory proteins involved in cell differentiation , cell-cycle regulation , transcriptional regulation , or apoptosis , but also aberrant proteins are directed to proteasomal degradation through conjugation with the small protein modifier ubiquitin via a cascade of E1 , E2 , and E3 enzymes , thus forming poly-ubiquitinylated ( poly-ub ) proteins [10] . Poly-ub proteins are substrates for 26S proteasomes which are formed through the association of two 19S regulator complexes with the catalytic core complex , the 20S proteasome , that hydrolyzes proteins into shorter peptide fragments [11] , [12] . Peptide hydrolyzing activity of the 20S core is restricted to three β-subunits , i . e . β1 , β2 , and β5 , located in the two inner heptameric β-rings of the 20S proteasome [13] . Upon interferon ( IFN ) -exposure of cells or tissues , alternative catalytically active β subunits , i . e . β1i/LMP2 , β2i/MECL-1 , and β5i/LMP7 , are induced . These so called immunosubunits are incorporated into newly formed 20S immunoproteasomes ( IP ) in a process that is driven by β5i/LMP7 [14] . β1i/LMP2 and β5i/LMP7 are encoded within the major histocompatibility II region and their incorporation into IPs induces altered proteolytic characteristics that result in many cases in more efficient liberation of MHC class I epitopes [15]–[17] particularly within the early phase of antiviral immunity [18] , [19] . This increase in MHC class I peptide supply by IPs appears to be important for triggering an effective early CD8 T cell response [20]–[23] . However , controversial findings about the association between IP function and CD8 T cell priming raised some doubts with regard to the in vivo impact of these data . In fact , an alternative physiological function of IPs has been demonstrated recently by our group in that IPs protect cells against cytokine induced oxidative damage , thus preserving protein homeostasis . Substrate modification of oxidant-damaged proteins with poly-ubiquitin results in protein degradation particularly by IPs [24] . Nevertheless , conclusive studies investigating the role of IP in response to viral infection beyond the analysis of specific T cell immunity have not been performed . Also , the importance of this regulated protease in cardiac disease remains to be elucidated . Within the context of the murine model of ongoing coxsackievirus B3 ( CVB3 ) -myocarditis , we recently reported on cardiac IP formation early upon infection in mice being resistant to chronic disease . The remarkably delayed induction of cardiac IPs in susceptible mice pointed towards a potential disease-modifying effect of this finding [19] . Here , we show that cardiac IP prevent exacerbation of acute CVB3-induced myocardial destruction and possess a protective function in viral heart disease expanding their role to the protection of cells against inflammation induced toxic effects thereby stabilizing cell viability in viral infection .
One well-established model to study myocardial inflammation is the induction of murine myocarditis with coxsackievirus B3 ( CVB3 ) in C57BL/6 mice leading to acute heart muscle injury at d8 p . i . [19] . Here , we have challenged both β5i/LMP7+/+ and β5i/LMP7-/- mice on a C57BL/6 background with CVB3 . Cardiac 20S proteasomes isolated from naive mice contain only very small amounts of β1i/LMP2 , β5i/LMP7 and β2i/MECL-1 [19] , [25] . To test whether β5i/LMP7 deficiency was indeed sufficient to negatively affect the incorporation of all three inducible catalytic subunits into cardiac IP in vivo , heart 20S proteasomes were isolated from naive mice and from CVB3-infected mice at the early stage of disease ( d4 p . i . ) and at the acute stage of myocarditis ( d8 p . i . ) from both β5i/LMP7+/+ and β5i/LMP7-/- mice . Whereas mRNA expression of β1i/LMP2 and β2i/MECL-1 was induced in both hosts ( Fig . 1A ) , immunoblot analysis revealed strongly impaired incorporation of both β1i/LMP2 and β2i/MECL-1 into cardiac 20S proteasomes in acute myocarditis in β5i/LMP7-/- mice ( Fig . 1B ) . Likewise , cytokine stimulation of primary cardiomyocytes from IP-competent mice with IFN-γ resulted in the efficient induction of all three IP subunits , whereas as expected incorporation of β1i/LMP2 , β2i/MECL-1 and β5i/LMP7 was impaired in cardiomyocytes isolated from IP-deficient mice ( Fig . 1C ) . To obtain quantitative information on the content of each proteasome subunit within the 20S core complex in vivo , 20S proteasomes were subjected to reverse phase nano HPLC and LTQ-Orbitrap mass spectrometry ( MS ) analysis . All three inducible subunits β1i/LMP2 ( 3 . 5-fold induction; p<0 . 05 ) , β2i/MECL-1 ( 2 . 6-fold induction; p<0 . 05 ) and β5i/LMP7 ( 2 . 0-fold induction; p = 0 . 10 ) were enhanced in CVB3-challenged hearts as early as at d4 p . i . in β5i/LMP7+/+ mice . In contrast , no improved incorporation of either β1i/LMP2 ( 1 . 3-fold; p = 0 . 67 ) or β2i/MECL-1 ( 1 . 5-fold; p = 0 . 38 ) was observed in cardiac 20S proteasomes from β5i/LMP7-/- mice at this stage of disease . Table 1 indicates relative quantitative expression levels of all three inducible subunits in cardiac proteasomes in wildtype mice with respect to IP-deficient mice . No significant differences were detected in the content of constitutive proteasome subunits as exemplarily shown in Table 1 for catalytic subunits β1 , β2 , and β5 , and for constitutive non-catalytic subunits α2 and β4 , respectively . Notably , in agreement with previous studies [26] , β5i/LMP7-/- mice show baseline deficits in cardiac β2i/MECL-1 incorporation being aggravated during inflammation ( Table 1 ) . Therefore , β5i/LMP7-/- mice encounter a substantial impairment of IP assembly in inflammation-challenged hearts . This appears to be crucial for the interpretation of our data in terms of IP deficiency showing a severe impairment in the incorporation of all three inducible subunits in acute disease in vivo . Histological analysis of acute myocarditis was done as previously published [27] defining acute myocarditis by lymphocytic infiltrates in association with myocyte necrosis , which we also see in patients with acute myocarditis [28] . CVB3-myocarditis was evaluated in IP-deficient mice at early stages of heart muscle infection ( d4 p . i . ) and at acute stages of myocarditis ( d8 p . i . ) . Except for some scattered macrophages , no foci of inflammatory infiltrates were observed in the myocardium of both mouse strains at d4 p . i . ( Fig . 2B ) . At this time point severe inflammation of the pancreas , the primary organ of viral replication , was observed in both β5i/LMP7+/+ and β5i/LMP7-/- mice ( Fig . 2A ) . Pointing towards accelerated organ destruction in β5i/LMP7-/- mice , representative images of the pancreas at d8 p . i . illustrate final pancreatic islet destruction with fibrous and fatty tissue organ replacement in β5i/LMP7-/- mice , with necrotic cells and massive inflammation still being present in β5i/LMP7+/+ mice ( Fig . 2A ) . Clinically , myocardial tissue damage in CVB3-infection is utmost important since acute myocardial injury may result in severe acute arrhythmia and heart failure . Focussing on myocardial damage comprising cardiac cell necrosis and inflammation in acute myocarditis , IP-deficiency was found to be associated with severe acute myocarditis . As representatively depicted in Fig . 2B and 2C for independent experiments by HE staining , at the acute stage of disease large foci encompassing inflammatory cells and cardiomyocyte necrosis were detected in β5i/LMP7-/- mice , which is in striking contrast to small areas of myocardial inflammation and tissues damage in β5i/LMP7+/+ mice . To obtain quantitative information on heart muscle injury , myocarditis scores were determined yielding a score of 3 . 1±0 . 3 in β5i/LMP7-/- mice vs . 2 . 2±0 . 2 in β5i/LMP7+/+ mice ( p<0 . 05 , Fig . 2C ) . Macrophages and CD3+ T lymphocytes represent the major fraction of invading inflammatory cells in acute CVB3-myocarditis in mice [27] , [29] . To address the inflammatory infiltrate in our model in detail , CD3+ T lymphocytes , B220+ B lymphocytes and Mac-3+ macrophages were studied by immunohistology . As demonstrated in Fig . 3 , the inflammatory infiltrate was primarily comprised of Mac-3+ macrophages and to a lesser extent of CD3+ T lymphocytes . B cells were scattered throughout the inflammatory lesions without significantly contributing to the invading cellular infiltrate . In agreement with myocarditis scores ( Fig . 2C ) , quantification of invading cells revealed significantly increased macrophages and T lymphocytes in CVB3-infected β5i/LMP7-/- mice ( Fig . 3D ) . As suggested by quantitative mRNA expression of CD8 and CD4 molecules in the infected myocardium , both CD4+ and CD8+ T lymphocytes were increased in CVB3-infected β5i/LMP7-/- mice ( mRNA expression of CD3 revealed the same result , data not shown ) . In contrast , expression levels of NKP46 , a marker for natural killer ( NK ) cells , did not differ in both hosts suggesting invasion of NK cells in both hosts to the same extent ( Fig . 3E ) . In summary , immunohistological characterization of myocardial inflammation revealed that cardiac IP formation protected CVB3-challenged hearts from exacerbation of acute heart muscle injury . The presence of infected cardiomyocytes adjacent to foci of mononuclear cell infiltrates is pathognomonic in viral myocarditis . Indeed , CVB3 in situ hybridization-positive cardiomyocytes were found within inflammatory lesions in acute heart muscle injury in both β5i/LMP7+/+ and β5i/LMP7-/- mice ( Fig . 4A ) . However , despite the severity of myocardial tissue damage in CVB3-challenged β5i/LMP7-/- mice , scoring of CVB3 in situ hybridization-positive cardiomyocytes pointed towards equal viral replication in both hosts . Also , the titers of cardiac infectious viral particles were found to be within the same range in both β5i/LMP7+/+ and β5i/LMP7-/- mice in acute disease ( Fig . 4B ) . To further investigate viral replication within the context of IP-deficiency , primary cardiomyocytes from β5i/LMP7+/+ and β5i/LMP7-/- mice were infected with CVB3 in vitro and CVB3 replication was determined by quantitative real-time PCR . These experiments were also carried out in the presence of type I IFN-stimulation to mimic the in vivo cytokine milieu in acute heart muscle injury . As shown in Fig . 4C , IP-deficiency revealed no influence on CVB3 replication in vitro . In line with the finding of identical viral load in CVB3-challenged cardiomyocytes from β5i/LMP7+/+ and β5i/LMP7-/- mice , efficient virus elimination was observed at the chronic stage of disease at d28 p . i . revealing no relevant signs of ongoing disease in both hosts ( data not shown ) . These findings pointed towards efficient induction of both innate and adaptive immunity also in mice lacking IP . To address this issue in detail , cytokine responses were determined in acute heart muscle injury ( Fig . 5A ) . Our data demonstrate increased cardiac expression of pro-inflammatory cytokines as shown here exemplarily for TNF-α , IFN-β , IL-6 and IFN-γ in acute myocarditis in both β5i/LMP7+/+ and β5i/LMP7-/- mice . Also , the expression of type I IFN-induced antiviral pathways like the 2′ 5′-oligoadenylate synthetase-like protein-2 ( OASL-2 ) , the IFN-stimulated gene 15 ( ISG15 ) , the Myxovirus resistance protein ( Mx ) and the protein kinase K ( PKR ) pathway was efficiently induced in CVB3-challenged hearts from β5i/LMP7-/- mice ( Fig . 5A ) . To determine whether IP expression is required for maintaining homeostatic levels of B and T cells , we counted B220+/CD19+ B cells and CD4+ and CD8+ T cells in spleens isolated from β5i/LMP7+/+ and β5i/LMP7-/- mice . Hensley et al . recently reported on different absolute B cell and T cell numbers and impairment in humoral immunity in β1i/LMP2-deficient mice [30] . Here , we detected equal CD4+ T cell and B cell numbers in both CVB3-challenged β5i/LMP7+/+ and β5i/LMP7-/- mice ( Fig . 5B , C ) . To further test humoral immunity in β5i/LMP7-/- mice , anti CVB3 IgG titers were determined at different time points p . i . CVB3-antibody titers in β5i/LMP7-/- mice were found to be within the range of anti CVB3 IgG in respective wildtype controls ( Fig . 5B ) . In accordance with data by Fehling et al . [21] , CD4+ and CD8+ T cells were not affected by β5i/LMP7 deletion in naive mice ( Fig . 5C ) . We observed an overall decrease in splenic T cell levels early upon disease potentially reflecting migration of these cells to secondary lymphoid organs and to the target organ of infection; however , host-specific effects were precluded . Also , we did not observe any differences in the absolute number of splenocytes in both hosts in acute myocarditis . Since β5i/LMP7-deficiency has been attributed to impaired CD8 T cell responses [21] , [31] , we also investigated CD8 T cell populations from both β5i/LMP7+/+ and β5i/LMP7-/- mice in vivo . Despite the fact that absolute frequencies of CVB3 VP2 [285–293]-specific CD8 T cells were rather low , no remarkable differences were detected in the absolute VP2 [285–293]-peptide specific CD8 T cell numbers in CVB3-infected β5i/LMP7+/+ and β5i/LMP7-/- mice ( Fig . S2A ) . Given that CD8 T cells are crucial in the control of CVB3-myocarditis [29] , [32] , absolute CD8 T cell numbers for CVB3 epitopes are low and detection of specific CD8 T cell frequencies is limited in this infection model , adoptive transfer studies of CVB3-memory CD8 T cells from IP-deficient and IP-competent mice appeared to be the most reliable approach to address the effect of IP-deficiency on CD8 T cell function . Take of donor CD8 T cells from CVB3-infected mice is shown in Fig . S2B . To preclude effects of IP-deficiency on CD8 T cell survival , CD8 T cells were isolated from CVB3-challenged β5i/LMP7+/+ and β5i/LMP7-/- mice ( both CD45 . 2 ) at d8 p . i . and transferred into naive B6 . SJL-Ptprca Pepcb/BoyJ mice ( CD45 . 1 ) . The amount of transferred CD8 T cells was assessed at d8 after adoptive T cell transfer revealing no impairment of IP-deficiency on CD8 T cell survival ( data not shown ) . Also , transfer of β5i/LMP7-deficient CD8 T cells ( CD45 . 2 ) into CVB3-infected B6 . SJL-Ptprca Pepcb/BoyJ mice ( CD45 . 1 ) and vice versa revealed comparable CD8 T cell survival rates ( Fig . 5D left panel ) . Next , CD8 T cells were isolated from CVB3-infected β5i/LMP7+/+ and β5i/LMP7-/- mice ( both CD45 . 2 ) at d8 p . i . These cells were transferred into naive β5i/LMP7+/+ and β5i/LMP7-/- mice , which were then infected with CVB3 and sacrificed at d8 p . i . to assess myocarditis scores . Following adoptive T cell transfer of IP-deficient CD8 T cells into either IP-deficient or IP-competent recipients , we observed no effect on acute heart muscle inflammation in comparison to adoptive T cell transfer of CD8 T cells from IP-competent mice into both recipients ( Fig . 5D right panel ) . Of note , adoptive T cell transfer of CD8 T cells from either CVB3-infected β5i/LMP7+/+ and β5i/LMP7-/- mice into β5i/LMP7-/- recipient mice resulted in a slightly milder acute disease than in non-transfected mice ( respective myocarditis score from CVB3-infected donor mice are shown in the middle panel of Fig . 5D ) . However , these effects were detected for T cell transfer of both β5i/LMP7-/- and β5i/LMP7+/+ T cells , thus arguing against a detrimental effect of IP-deficiency on memory CD8 T cell function and being in accordance with the observation of equal virus titers and efficient viral clearance in both β5i/LMP7+/+ and β5i/LMP7-/- mice . The data above illustrated severe tissue damage in CVB3-challenged hearts in mice lacking IPs and revealed large foci of inflammatory lesions in this host . Given that IPs preserve protein homeostasis and cell viability in response to cytokine stress [24] , one may argue that viral infection induced cytokine release affects the cellular protein equilibrium in cardiomyocytes and invading inflammatory cells , which may further exacerbate heart muscle injury in IP-deficient hearts . To test this hypothesis , primary cardiomyocytes and B-cell depleted splenocytes ( which represent the major populations of invading inflammatory cells ) were isolated from β5i/LMP7+/+ and β5i/LMP7-/- mice and exposed to IFN-γ ( cell purity is depicted in Fig . S1 ) . Upon cytokine exposure , lack of IPs resulted in increased accumulation of poly-ub substrates in these cells ( Fig . 6A ) . Failure of IP expression also coincided with increased accumulation of oxidant-damaged proteins in IP-deficient cardiomyocytes and inflammatory cells in response to prolonged cytokine exposure ( Fig . 6A ) . IPs also contribute to the activation of NFκB transcription factor by accelerated turnover of IκBα , which is crucial for multiple processes in inflammation and apoptosis [24] , [33] . Impaired activation of NFκB as shown here by reduced levels of NFκB p50 subunits in IP-deficient cardiomyocytes and inflammatory cells ( Fig . 6B ) reflected reduced proteasomal degradation of NFκB p105 precursor proteins , which is in concordance with impaired proteolysis in IP-deficient cells as shown in Fig . 6A . These data suggested a role of IPs in regulating proteotoxic stress also in the infected myocardium . Indeed , IP-deficient mice failed to cope with accelerated protein turnover in CVB3 infection as reflected by increased accumulation of poly-ub proteins in acute inflammatory viral heart disease ( Fig . 6C ) . The IP-deficient myocardium was not able to efficiently cope with the required protein turnover in acute CVB3 myocarditis ( Fig . 6D + Fig . S3 ) . Consequently , we observed significantly enhanced ALIS formation in the injured myocardium ( evaluation of poly-ub-aggregates at the acute stage of myocarditis: β5i/LMP7+/+ mice: 13 . 5 ALIS / 1088 µm2±1 . 0 vs . β5i/LMP7-/- mice: 20 . 0 ALIS / 1088 µm2±1 . 8; p<0 . 05; n = 5 mice ) . These poly-ubiquitin conjugates were primarily detected within inflammatory lesions in invading inflammatory cells , and within the cytoplasm and nuclei of adjacent cardiomyocytes in acute myocarditis ( Fig . 6E ) . Also , poly-ub signals were found to be increased in β5i/LMP7-/- mice in comparison to β5i/LMP7+/+ mice at d8 p . i . ( Fig . 6E + Fig . S4 ) . Since oxidant damaged proteins become substrates of the 26S IP upon tagging by poly-ub [24] , the levels of carbonyl groups reflecting oxidant protein damage were monitored . As illustrated in Fig . 6F , oxidant protein damage was increased in acutely inflamed hearts in IP-deficient mice . Since CVB3 titers were found to be within the same range in both hosts ( Fig . 4 ) , cytolytic effects of CVB3 do apparently not explain severe tissue injury as observed here in IP-deficient mice . However , oxidative-protein damage , inefficient degradation of poly-ub protein aggregates and reduced activation of NFκB transcription factor in CVB3-challenged hearts in mice lacking IPs may affect cell viability . Indeed , cytokine-induced cellular injury predominantly occurred in vitro in cardiomyocytes and macrophages that were isolated from IP-deficient mice ( data not shown ) . To study cellular injury due to apoptotic cell death in vivo , DNA strand breaks as an early sign of apoptosis were assessed in cardiac tissue sections . No apoptotic cell death was detected in hearts from β5i/LMP7+/+ and β5i/LMP7-/- mice at d4 p . i . ( Fig . S5 ) . However , in acute heart muscle injury ( d8 p . i . ) , increased levels of DNA strand breaks were detected particularly within inflammatory lesions and the surrounding tissue in CVB3-challenged β5i/LMP7-/- mice ( Fig . 7 , Fig . S5+S6 ) . TUNEL positive staining was detected throughout the injured heart in IP-deficient mice; thus , apoptotic cell death was found to be quantitatively increased in β5i/LMP7-/- mice . Despite the fact that minor inflammatory lesions were also detected in CVB3-infected β5i/LMP7+/+ mice ( d8 p . i . ) , here no significant apoptotic cell death occurred ( Fig . 7 ) . This observation is in agreement with previously published data [29] . These findings support the role of IP formation in cardiomyocytes and in inflammatory cells to protect the injured tissue from proteotoxic stress , which may exacerbate acute heart muscle injury in viral heart disease .
Proteasomes are responsible for the generation of peptides derived from pathogens or cellular proteins that are presented by MHC class I molecules on the cell surface to cytotoxic T cells ( CTLs ) [11] . Despite the fact that in vitro studies argued in favor of an impact of IP-dependent MHC class I antigen processing [15]–[17] , [19] , in vivo studies using IP-deficient mice reported conflicting data on the induction of CD8 T cell responses [17] , [21] , [23] , [34] . CD8 T cells are crucial in virus elimination in CVB3-myocarditis [29] , [32] . To study a potential contribution of IPs to the generation of CD8 T cell responses in CVB3-myocarditis , here adoptive memory CD8 T cell transfer experiments were performed since limited knowledge on immunodominant CVB3-specific CD8 T cell epitopes restrains solid quantification of CD8 T cell responses in murine enterovirus myocarditis [19] , [35] . Transfer of CVB3 memory CD8 T cells from IP-competent mice did not reveal a beneficial effect on CVB3 myocarditis in comparison to transfer of CD8 T cells from IP-deficient mice . Likewise , CVB3 titers were within the same range in IP-deficient and wildtype control mice in acute myocarditis ( Fig . 4 ) and the virus was efficiently eliminated in both hosts at d28 p . i . These findings support the induction of efficient CD8 T cell responses also in CVB3-challenged IP-deficient mice , which is in agreement with observations in other infection models: the kinetics of lymphocytic choriomeningitis virus clearance were similar in both β5i/LMP7+/+ and β5i/LMP7-/- mice [34] . This strongly supports the notion that the key innate function of IP in enteroviral heart disease lies elsewhere . Here , we have illustrated that IPs in CVB3-induced heart muscle injury preserve protein homeostasis and maintain cell viability in order to protect the inflammation-challenged myocardium from severe damage . The UPS adapts to stress induced requirements by increased substrate turnover exerted by IPs , which possess improved peptide-hydrolyzing activity [15] , [17] , [35] , and poly-Ub-substrate turnover [24] . Indeed , cardiac IP formation in CVB3-myocarditis resulted in enhanced proteasomal peptide-hydrolyzing activity [19] . One of the pivotal functions of the UPS is to limit the accumulation of potentially toxic misfolded proteins and protein aggregate formation , which as consequence of cellular stress represent a constant threat to normal cell function and cell viability [7] , [36] . Also , CVB3-infection [2] as well as cytokine stress in inflammation [1] induce oxidative stress . Likewise , the pathogenesis of severe CVB3 myocarditis has been attributed to increased oxidative stress [37]–[39] . In agreement with our previous study reporting on the activity of the 26S IP for efficient elimination of oxidatively modified , poly-ub proteins in response to cytokine stress [24] , here we demonstrated that IP function in both residing host cells and invading inflammatory cells is crucial for the efficient degradation of poly-ub proteins in acute viral heart disease . Elimination of nascent oxidant-damaged , poly-ub proteins by the 26S IP prevented the accumulation of harmful protein aggregates [7] , [36] , which may exacerbate acute heart muscle injury . Moreover , as a consequence of impaired IP function , poly-ub proteins accumulated in ALIS in CVB3-infected hearts . Such aggregates , which are at least partially comprised of CVB3 proteins [40] , are not inert metabolic end products , but may actively influence the metabolism of cells [41] . As shown here , degradation of oxidant-damaged , poly-ub proteins by cardiac 26S IPs in CVB3-challenged hearts resulted in ALIS degradation and likewise protected cardiomyocytes and invading inflammatory cells from proteotoxic stress . This first histological demonstration of severe tissue injury in virus infection in mice lacking IPs is in agreement with findings in experimental acute encephalomyelitis ( EAE ) . IPs prevented accumulation of toxic protein aggregates in EAE , coinciding with less severe disease manifestation in IP-competent mice [24] . The detection of both poly-ub proteins in concert with apoptotic cell injury within inflammatory lesions in viral heart disease ( Fig . 6+7 ) also supports the association between proteotoxic stress and cellular injury in this model [7] , [24] . In fact , cellular injury as shown here by increased apoptotic cell death of invading inflammatory cells and adjacent cardiomyocytes in IP-deficient mice may result in the release of endogenous molecules , which as damage-associated molecular patterns ( DAMPs ) signal the threat of infection and injury to the organism . High levels of DAMPs have been linked to the pathogenesis of many inflammatory diseases , drive cellular activation and immunoreactivity [42] . This may in fact exacerbate acute inflammation and also result in killing of non-infected cardiomyocytes . Thus , IPs prevent excessive proteotoxic stress and cellular injury , which in consequence may limit additional effects like DAMP-associated activation of immunopathology . Remarkably , this is the first study illustrating a detrimental effect of IP-deficiency in a viral infection-induced phenotype despite a lack of a significant effect of IPs on pathogen load . Our findings are in agreement with the observation that absence of β5i/LMP7 expression impairs the beneficial effects of IFN-β in patients suffering from multiple sclerosis [43] . In contrast , absence of IP-function either as a result of β5i/LMP7 deficiency or inhibition of β5i/LMP7 catalytic activity by PR-957 have recently been associated with attenuated experimental colitis [44] . Depending on the pathogenesis of the underlying disease , IP deficiency seems to exert either protective effects or to aggravate the consequences of inflammation in a disease or tissue-specific manner . Indeed , cytokine responses in IP-deficient mice differ considerably and are strikingly dependent on the disease entity being studied . CVB3 infection of IP-deficient mice revealed a cytokine induction in viral heart muscle injury comparable to that observed in wildtype mice ( Fig . 5A ) . In contrast , amelioration of experimental colitis has been connected to limited induction of proinflammatory cytokines and chemokines [33] , [44] . This was attributed to impaired activation of transcription factor NFκB , a central regulator of inflammation in inflammatory bowel disease [33] . In line with impaired NFκB activation in TNF-α stimulated murine embryonic fibroblasts [33] , this study demonstrated reduced NFκB activation in cytokine-stressed cardiomyocytes and inflammatory cells lacking IPs ( Fig . 6B ) . This may be attributed to the fact that IκBα , a specific inhibitor of NFκB activation , is degraded much faster in cells expressing IPs [45] . Also , IκBα has been identified to be oxidatively-modified upon cytokine stress , which supports the role of IPs in degradation of this specific substrate [24] . Whereas increased activation of NFκB is believed to exert detrimental functions in immune and non-immune cells in tissues affected by chronic inflammation , NFκB inhibition can also be harmful for the organism , and trigger the development of inflammation and disease . These findings suggested that NFκB signaling has important functions for the maintenance of physiological immune homeostasis and for the prevention of inflammatory diseases [46] . Studies with specific inhibitors of NFκB nuclear translocation and activity revealed induction of apoptosis , thus argueing in favor for anti-apoptotic effects of this prosurvival transcription factor as well [47] . In agreement with these findings , the here described impaired activation of NFκB may additionally contribute to the effects of proteotoxic stress , which resulted in cellular injury as shown in Fig . 7 . Whereas interference with two major pathways leading to NFκB activation exerts beneficial effects in experimental colitis and anti-cancer treatment , our data indicate that activation of NFκB-mediated responses protects cytokine-challenged cardiomyocytes and inflammatory cells and argues against a significant contribution of NFκB to cytokine induction in viral heart disease . In conclusion , our findings support the view of a distinct tissue specific contribution of IP function driven by the pathogenesis of the underlying inflammatory disease .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the German animal welfare act , which is based on the directive of the European parliament and of the council on the protection of animals used for scientific purposes . This study conforms to the Berlin State guidelines for animal welfare . The protocol was approved by the Committee on the Ethics of Animal Experiments of Berlin State authorities ( Permit Number: G0311/06 ) . All efforts were made to minimize suffering . CVB3 ( cardiotropic Nancy strain ) used in this study was prepared as previously described [48] . C57BL/6 mice were initially from Jackson Laboratory . C57BL/6 β5i/LMP7-/- mice originally obtained from HJ Fehling [21] , who backcrossed them with C57BL/6 mice 10 times . Breeding pairs were kindly provided by U Steinhoff ( Berlin , Germany ) . Mice were kept at the animal facilities of the Charité University Medical Center . Six week-old mice were infected i . p . with 1×105 PFU CVB3 . Hearts were perfused with PBS , weighted and quickly frozen in liquid nitrogen before storage at -80°C . For some lymphocyte transfer experiments , B6 . SJL-Ptprca Pepcb/BoyJ were purchased from Jackson Laboratory . Primary cardiomyocytes ( CM ) were isolated from fetal mouse hearts ( E13 ) . Hearts were incubated in EDTA/Trypsin at 4°C overnight , followed by 15 min incubation at 37°C . Cardiac cells were resuspended in standard medium and transferred into cell culture flasks . Purity was determined by flow cytometry using cardiomyocyte-specific troponin I antibodies ( abcam # 47003 ) . Inflammatory cells were taken from whole spleen cell suspensions , which were B cell depleted by MACS ( Miltenyi Biotec ) . All cell lines were cultivated under standard conditions in Dulbecco's MEM ( MEFs ) each containing 10% fetal calf serum ( FCS ) , 2 mM L-glutamine , 100 U/ml penicillin and 100 mg/ml streptomycin . Cells were treated with either 100 U/ml IFN-β , 100 U/ml IFN-γ ( Sigma-Aldrich ) or 30 ng/ml TNF-α ( all Sigma-Aldrich ) . Cells were infected with CVB3 ( cardiotropic Nancy strain ) MOI 0 . 1 or 0 . 5 as indicated . Plaque assay was performed as described previously [19] . In situ hybridization of genomic CVB3 RNA , histological staining with hematoxylin / eosin ( HE ) and immunohistochemistry for detection of CD3+ T lymphocytes and Mac-3+ macrophages were carried out and analysed as described [27] . Immunohistochemical stainings for ubiquitin and B cells were performed on a Ventana Benchmark stainer using the Vectostain Elite ABC Kits ( Vector Laboratories; Burlingame , CA ) . The following primary antibodies were used: anti-ubiquitin 1∶1000 ( DAKO Cytomation , Glostrup , Denmark ) and CD45R/B220 1∶100 ( BD Pharmingen , Heidelberg , Germany ) . Biotin-labeled secondary antibodies ( goat-anti-rabbit and goat-anti-rat ) were purchased from Jackson Immuno Research ( Dianova , Hamburg , Germany ) and used at a dilution of 1∶100 . All slides were counterstained with hematoxylin . Processing of cryo-sections and Hoechst staining was performed as described [24] . Briefly , cryo-sections were fixed in 4% paraformaldehyde , washed and permeabilized with PBS/1% TritonX . Staining was performed with FK1 mAb ( PW 8805 , Biomol , Germany ) at 4°C over night . Confocal images were acquired on a Leica TCS SP2 microscope ( Leica Microsystems ) . Quantification of ALIS has been based on counting cells with accumulation of poly-ub conjugates ( focused staining over background defined as ub-rich inclusions ) in defined areas ( 1088 µm2 ) at 100-fold magnification . RNA preparation and cDNA synthesis were performed as described recently [19] . TaqMan PCR was performed using primers and probes of TaqMan Gene Expression Assays ( Applied Biosystems , Germany ) . mRNA expression was normalized to the housekeeping gene HPRT by means of the ΔCt method . Cell or tissue lyses was performed in 20 mM TRIS-HCl , pH 7 . 5 , 10 mM EDTA , 100 mM NaCl , 1% NP40 , 10 µM MG132 , 5 mM NEM , Complete protease inhibitor cocktail ( Roche , Germany ) . Immunoblot analysis was performed according to standard protocols . ubiquitin: Z0458 DAKO; α6 ( pc , K379 ) , β5i/LMP7 ( pc , K63 ) , β2i/MECL-1 ( pc , K65 ) : lab stock ( all generated against peptides of the respective protein ) , β1i/LMP2: Abcam #3328 for isolated proteasomes or β1i/LMP2: lab stock for cardiomyocytes ( cross-reaction with β1 , pc , K620/21 ) ; α-actin , GAPDH: Santa Cruz . The detection of oxidatively-damaged proteins was performed indirectly by chemical derivatization: this derivatization captures the oxidative state immediately during or after homogenization of the tissue . Oxidized proteins were visualized with the OxyBlot protein oxidation detection kit ( Chemicon International ) via immunodetection of carbonyl groups . DNA strand breaks ( TUNEL assay ) were determined by in situ cell death detection kit , TMR red ( Roche , Germany ) or in situ cell death detection kit , POD ( Roche , Germany ) according to the instructions of the manufacturer . POD stained slides were counterstained by hematoxylin . After induction of IPs with IFN-γ , primary cardiomyocytes and B-cell depleted splenocytes were stimulated with 30 ng/ml TNF-α for 30 min . p50 NFκB was determined in whole tissue homogenates by ELISA according to the manufacture's instructions ( ActiveMotif , Rixensart , Belgium ) . CVB3-specific IgG antibody titers were determined with Enterovirus ELISA Kit ( Genzyme Diagnostics ) according to the manufacture's instructions [alternative secondary antibody ( POX anti-mouse IgG , Dianova ) was used] . CVB3-specific antibody titers are presented as log2 of the maximum dilution of serum showing an optical density greater than the mean optical density of sera obtained from naive mice plus threefold SD as described recently [27] . After Fc-receptor blockade cells were incubated with different combinations of fluorescently labeled Abs ( eBiosciences and BD Biosciences ) and samples were analyzed using CYAN-ADP flow cytometer ( Beckman Coulter , Germany ) or BD FACSCalibur ( Becton Dickinson ) . At day 8 p . i . , CVB3-infected mice were sacrificed and splenocytes were isolated according to standard protocols . CD8+ T cells were purified by positive selection using commercially available kits yielding a purity of at least 85% ( Miltenyi Biotec ) . 1-2×106 CD8+ T cells from one donor were injected i . v . through the tail vein into one recipient . After T cell transfer , mice were injected i . p . with 1×105 PFU CVB3 and sacrificed at the acute stage of infection at d8 p . i . Myocarditis was assessed as described above . 20S proteasomes were isolated according to standard procedure as described [19] . The mixture of tryptic peptides was separated prior to mass spectrometric analyses by reverse phase nano HPLC on a 15 cm PepMap100-column ( 3 µl , 100 Å ) using an Proxeon System ( Odense , Denmark ) at a flow rate of 1 µl/min . Separation was carried out in a linear gradient within 86 min using 0 . 05% acetic acid , 2% acetonitrile in water and 0 . 05% acetic acid in 45% acetonitrile as eluents . MS-data were generated on an LTQ-Orbitrap-MS equipped with a nanoelectrospray ion source ( PicoTip Emitter FS360-20-20-CE-20-C12 , New Objective ) . After a first survey scan ( r = 60 , 000 ) MS2 data were recorded for the five highest mass peaks in the linear ion trap at a collision induced energy of 35% . The exclusion time was set to 30 s and the minimal signal for MS2 was 1 , 000 . Peptide identification was achieved by searching the SwissProt database rel . 57 . 1 restricted to mouse entries using SEQUEST search engine ( SageN Research ) and further processed by PeptideTeller and ProteinTeller [49] within the Elucidator system ( Rosetta Biosoftware , Seattle , WA , U . S . A . ) . ProteinTeller results were further used for annotation , with a predicted error rate of<5% . Quantitative analysis of label-free MS data was achieved with the Elucidator system using peptide intensities as proxies for label-free peptide abundance measurements . The following criteria for frame and feature annotation were used: retention time minimum cut-off 9 min , retention time maximum cut-off 80 min , m/z minimum cut-off 300 , instrument mass accuracy 5 ppm , alignment search distance 10 min . For quantitative analysis , the data were normalized and further grouped ( two biological and two technical replicates ) . Results of continuous variables are expressed as mean±standard error of mean ( SEM ) if not indicated otherwise . Two group comparisons of non-parametric data were performed using the Mann-Whitney test . Statistical significance between multiple groups was determined using two-way ANOVA and post hoc analysis with a Bonferroni test . Significance was assessed at the p<0 . 05 level ( * indicates significant differences ) . TNF-α ( Q0X0E6 , P06804 ) , IFN-β ( P01575 ) , IL-6 ( P08505 ) , IFN-γ ( P01580 ) , OASL-2 ( Q9Z2F2 ) , ISG15 ( Q64339 ) , Mx ( P09922 ) , PKR ( Q03963 ) , NFκB p105 ( P25799 ) , IκBα ( Q9Z1E3 ) , TLR7 ( P58681 ) , TLR8 ( P58682 ) , LMP2 ( P28076 ) , MECL-1 ( O35955 ) , LMP7 ( P28063 ) , CD8 ( P01731 ) , CD68 ( P31996 ) , CD3 ( P22646 ) , B220 ( P06800 ) , CD4 ( P06332 ) , CD19 ( P25918 ) | The proteasome recognizes and degrades protein substrates tagged with poly-ubiquitin chains . Immune cells and cells activated by inflammatory cytokines/interferons express immunoproteasomes ( IPs ) that are characterized by unique catalytic subunits with increased substrate turnover . In infectious disease , the function of IPs is still a matter of controversial debate . Here , we report on a novel innate function of IPs in viral infection . We studied the murine model of acute enterovirus myocarditis , which represents one of the most common viral disease entities among young people . We found that IPs protect the pathogen-challenged tissue from severe injury , which was reflected in severe myocardial destruction and large inflammatory foci in mice lacking IPs . We show data that this prevention of excessive inflammatory tissue damage in viral heart disease is primarily attributed to preservation of protein homeostasis due to accelerated substrate turnover by IPs . Thus , a major innate function of IPs in viral infection is to stabilize cell viability in inflammatory tissue injury . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
]
| [
"immunology",
"microbiology",
"histology",
"adaptive",
"immunity",
"animal",
"models",
"of",
"infection",
"inflammation",
"major",
"histocompatibility",
"complex",
"proteins",
"biology",
"immune",
"response",
"biochemistry",
"protein",
"chemistry",
"antigen",
"processing",
"and",
"recognition",
"immunity",
"virology",
"innate",
"immunity"
]
| 2011 | Impairment of Immunoproteasome Function by β5i/LMP7 Subunit Deficiency Results in Severe Enterovirus Myocarditis |
Circadian clocks are ubiquitous in eukaryotic organisms where they are used to anticipate regularly occurring diurnal and seasonal environmental changes . Nevertheless , little is known regarding pathways connecting the core clock to its output pathways . Here , we report that the HAD family phosphatase CSP-6 is required for overt circadian clock output but not for the core oscillation . The loss of function Δcsp-6 deletion mutant is overtly arrhythmic on race tubes under free running conditions; however , reporter assays confirm that the FREQUENCY-WHITE COLLAR COMPLEX core circadian oscillator is functional , indicating a discrete block between oscillator and output . CSP-6 physically interacts with WHI-2 , Δwhi-2 mutant phenotypes resemble Δcsp-6 , and the CSP-6/WHI-2 complex physically interacts with WC-1 , all suggesting that WC-1 is a direct target for CSP-6/WHI-2-mediated dephosphorylation and consistent with observed WC-1 hyperphosphorylation in Δcsp-6 . To identify the source of the block to output , known clock-controlled transcription factors were screened for rhythmicity in Δcsp-6 , identifying loss of circadian control of ADV-1 , a direct target of WC-1 , as responsible for the loss of overt rhythmicity . The CSP-6/WHI-2 complex thus participates in the clock output pathway by regulating WC-1 phosphorylation to promote proper transcriptional/translational activation of adv-1/ADV-1; these data establish an unexpected essential role for post-translational modification parallel to circadian transcriptional regulation in the early steps of circadian output .
Circadian clocks are endogenous timekeepers that control a wide variety of daily biochemical , physiological , molecular and behavioral rhythms in mammals , plants , insects , fungi and cyanobacteria . The circadian system consists of three essential parts , input , a central oscillator and output [1–4] . In fungi and animals , the backbone of the oscillator mechanism is a transcriptional and translational autoregulatory feedback loop driven by positive and negative elements . The positive element , a heterodimeric transcription factor in which the proteins interact via PAS domains , drives expression of the negative element , a complex of proteins that physically interacts with the positive element to reduce its activity . In the case of Neurospora crassa , the positive elements are the PAS domain containing transcription factors White Collar-1 ( WC-1 ) and White Collar-2 ( WC-2 ) that form the White Collar Complex ( WCC ) . WCC in turn activates transcription of the negative element gene frequency ( frq ) ; FRQ nucleates formation of a complex including FRQ Interacting RNA Helicase ( FRH ) and Casein Kinase 1 ( the FRQ/FRH Complex or FFC ) that feeds back to physically interact with WCC and suppress frq transcription [5–7] . FRQ is progressively phosphorylated over time , modifications that provide the long time constant for the cycle and that ultimately reduce the affinity of the FFC for the WCC , releasing it to initiate the next cycle of transcription . Eventually hyperphosphorylated FRQ is turned over via a ubiquitin-mediated pathway , but in a normal circadian cycle the kinetics of this turnover is not believed to influence the period length of the clock [8 , 9] . Both WC-1 and WC-2 are phosphorylated in vivo under circadian conditions and become hyperphosphorylated after a short light exposures [10 , 11] . In the current model of the circadian feedback loop , the FRQ-FRH complex ( FCC ) closes the loop by inhibiting WCC activity via the promotion of phosphorylation of WCC , primarily through kinases CK-1a and CKII [12–14] . The importance of WCC phosphorylation for circadian oscillation has been argued based on short period , low amplitude , phase shifted and arrhythmic phenotypes resulting from mutations of phosphorylation sites on WCC [15–17] . In the current model , hyperphosphorylated WCC is believed to be inactive but stable whereas hypophosphorylation WCC is active and supports transcriptional activation of frq and other genes [12–14] . PP2A ( protein phosphatase 2A ) is believed to dephosphorylate WC-1 in vivo and this is correlated with an increase frq RNA levels [2 , 18] . In addition to its clock functions , WC-1 and WC-2 ( WCC ) comprise the blue light photoreceptor that initiates the organism’s principal photoresponse . Upon illumination the WCC undergoes a rapid conformation change , binding to light-responsive elements ( LREs ) via WC-2 and functioning as a TF to bind to and regulate the expression of hundreds of light-responsive genes [10 , 19–22] . Similar to WCC functioning in the dark , hyperphosphorylated WC-1 is believed to be transcriptionally inactive and hypophosphorylated WC-1 transcriptionally active . Consistent with this are reports that hyperphosphorylated WC-1 binds less strongly to target promoters while dephosphorylation of WC-1 increases promoter binding [23 , 24] . VVD ( VIVID ) , a small PAS/LOV protein and another blue light receptor , acts as a repressor of the light response through its physical interaction with the WCC [11] , and recent studies have shown the photocycle length of VVD plays a dominant role in determining the utility of the photoreceptor [25] . Though VVD is not required for clock rhythmicity , it modulates various WCC-mediated circadian clock properties such as gating of light input of clock and phasing light resetting responses . Loss of function vvd mutants exhibit a 4-hour delay of clock-controlled conidiation [26 , 27] . Time of day information generated by the circadian clock is transduced to clock control genes ( ccgs ) whose time-of-day specific expression yields products , the output pathway , that generate overt rhythmicity in the cell [28 , 29] . The best-characterized and most easily monitored output of the FFC/WCC Oscillator is the conidiation rhythm [30 , 31] . Though enormous advances have been seen in understanding core oscillators of Neurospora in the past two decades , how circadian oscillators signal through output pathways to control rhythmic activity of those ccgs remains only partially understood at molecular level [9 , 30] . Most recently , clock-controlled genes showing consistent rhythms and comprising as much as 40% of the genome have been identified in N . crassa by RNA-seq [32] . While these will provide a great resource for studying rhythmic behavior in the future they have afforded limited insight into the connection between the core clock and individual outputs . Here we characterize functions of a HAD-domain family phosphatase protein , CSP-6 , revealing its essential role in regulating circadian output pathways including conidiation rhythms and phase resetting . CSP-6 physically interacts with a partner , WHI-2 , and this phosphatase complex , which interacts with WC-1 , is important for maintaining WC-1 protein and phosphorylation levels . Loss of function of csp-6 results in constitutively hyperphosphorylated WC-1 but does not ablate core oscillator function , although the clock shows a 3 . 5-h phase delay due to reduced amounts and hyperphosphorylation of WC-1 . This indicates that the regulation of genes and proteins acting downstream of WC-1 to control circadian developmental processes has been disrupted in Δcsp-6 , and our results are consistent with a model in which ADV-1 , a direct target of WC-1 , plays this direct role in regulating overt clock output [33] . Prior to this finding the early steps of circadian output have been viewed primarily as a transcriptional network of activators and repressors . CSP-6 demonstrates an essential role for post-translational modifications in the early steps of output .
The csp-6 gene ( NCU08380 ) encodes a member of the haloacid dehalogenase ( HAD ) superfamily containing a conserved dullard-like phosphatase domain , ( S1A Fig , HAD domain 222–388 aa ) . It was first reported in a screen of putative phosphatases as having a conidial separation phenotype [34] and was later reported ( under the name psr-1 ) to be more broadly involved in female sexual development , cell-cell-fusion and autophagy [35] . Because the conidial separation defect was similar to that seen in mutants of csp-1 and csp-2 , two transcription factors important for conidiation and the circadian clock , we asked whether Δcsp-6 has similar phenotypes . We crossed the Δcsp-6 deletion mutant with the ras-1bd mutant and performed race tube assays ( Fig 1A ) revealing that , unlike the ras-1bd strain , Δcsp-6 , ras-1bd show reduced hyphal growth and arrhythmic conidiation . Use of frq-luc to report clock core oscillator function [36] in this background ( Δcsp-6 , ras-1bd , csr::c-box-luc ) , however , revealed rhythmic frq expression with a wild type period length around 22 h , albeit having a somewhat reduced amplitude and a 3 . 5-hour phase delay ( Fig 1B and 1C ) . While luciferase reporter data as seen in Fig 1B is an excellent indicator of rhythmicity because of the density of time points , the amount of bioluminescence produced is a function of growth and development as well as the clock; Δcsp-6-induced changes in growth characteristics could thus influence bioluminescent output , so we also examined frq and FRQ biochemically to more directly view the effect of Δcsp-6 on the oscillator components . Analysis of mRNA by RT-qPCR and protein by Western blotting confirmed core clock rhythmicity , showing that FRQ oscillated with a small loss of amplitude in Δcsp-6 mutants and displayed a delay of around 4 hours ( Fig 1D ) , and frq mRNA levels rhythmic but again phase delayed about 4 hours ( Fig 1E ) . We also followed FRQ and frq levels from DD24h to DD48h during the second day after moving from LL to constant DD . The results showed FRQ protein levels were reduced but we still can observe rhythmicity in FRQ amounts along with its phosphorylation ( S1B Fig ) , consistent with c-box-luc luciferase traces in Δcsp-6 . The frq mRNA level was reduced in Δcsp-6 but the amount still oscillated with a peak time of DD36h , 4 hours delay compared to wild type ( S1C Fig ) . These data indicated that CSP-6 plays a role in maintaining robust frq/FRQ expression but is not required for the clock core oscillation . So we concluded that the arrhythmic conidiation on race tubes in the Δcsp-6 mutant was caused by disruption of a circadian output pathway . The csp-6 promoter shows very weak circadian regulation when assayed by luciferase fusion ( S1D Fig ) , but csp-6 is not a typical clock-controlled gene involved in circadian output . The weak transcriptional regulation detected by luciferase is also seen in a translational fusion reporter of CSP-6 but not confirmed by Western analysis ( S1E and S1F Fig ) , so its action appears more likely to facilitate output rather than to drive it . Saccharomyces contains both an ortholog and a paralog of csp-6 , psr1 and psr2 respectively , both of which arose together in screens and share similar mutant phenotypes , interactors , and functions ( http://www . yeastgenome . org/locus/S000004009/overview ) ; in yeast these phosphatases function to regulate the stress response [37] . Likewise , Neurospora has a csp-6 paralog ( NCU08948 provisionally named pph-11 [34] ) that contains the same conserved HAD phosphatase domain as csp-6 ( S2A Fig , underlined sequences ) . To investigate whether csp-6 and its paralog are both involved in circadian clock output , conidiation rhythms were followed on race tubes in ΔNCU08948 ( psr-2 , ras-1bd ) ( S2B Fig ) . The ΔNCU08948 , ras-1bd mutant showed normal overt rhythmicity although it grew a bit slower compared to the wild type . Assay of the core circadian oscillator via a frq-luc reporter confirmed a robust circadian rhythm ( S2C Fig ) indicating that only csp-6 but not its paralog is involved in circadian function in N . crassa , and unlike Saccharomyces , the two paralogs have some distinct functions . Because the WCC drives frq expression and we observed delayed transcriptional and translational rhythms of frq/FRQ ( Fig 1C–1E ) , we examined the expression of WC-1 . As shown in Fig 2A , the protein levels of WC-1 in DD ( constant dark ) , LL ( constant light ) or following a light pulse ( LP ) were significantly reduced in Δcsp-6 , and WC-1 was hyperphosphorylated in Δcsp-6 in all conditions we examined ( Fig 2A ) . To confirm that the mobility shift of WC-1 in Δcsp-6 is caused by phosphorylation , total protein of wild type and Δcsp-6 were treated with λ-phosphatase . The results showed that WC-1 in both wild type and Δcsp-6 collapsed to the same level in both tested time points , constant dark 24h and light pulse 15min , indicating the low mobility form of WC-1 in Δcsp-6 is a hyperphosphorylation form of WC-1 ( S3A Fig ) . In addition , WC-1 is less stable in Δcsp-6 than in wild type indicating that CSP-6 stabilized the WC-1 protein ( Fig 2B ) . Furthermore and consistent with Western analysis , wc-1 mRNA levels were significantly reduced in Δcsp-6 under all conditions as well , and , although wc-1 is still light-induced , it took longer ( LP30 min ) to reach peak in the Δcsp-6 , indicating that the light response was affected in Δcsp-6 ( Fig 2C ) . WC-1 is required for light-induced carotenogenesis and cultures of Δcsp-6 were pale pink instead of orange , so we tested expression level of three albino genes in the Δcsp-6 mutant ( S3B Fig ) . All three albino genes ( al-1 , al-2 , al-3 ) showed reduced mRNA levels in Δcsp-6 consistent with reduced carotenoid accumulation and the pale color in Δcsp-6 ( S3C Fig ) . Additionally , we examined the mRNA expression level of a few light induced genes al-1 , al-3 , sub-1 and frq , following light pulses in order to understand whether hyperphosphorylated WC-1 in Δcsp-6 would affect their light response . Interestingly , sub-1 and frq showed normal kinetics in Δcsp-6 while al-1 and al-3 showed a delayed light response compared to the wild type , and al-1 , al-3 , sub-1 all had reduced expression in Δcsp-6 ( S3D Fig ) . These data suggested that the inability to dephosphorylate WC-1 in Δcsp-6 impacted expression its downstream targets . The hyperphosphorylation of WC-1 observed in the Δcsp-6 background suggested that WC-1 could be a direct target of CSP-6 . To determine whether CSP-6 interacts with WC-1 , co-immunoprecipitation ( co-IP ) assays were performed using CSP-6 epitope tagged with V5 on its C-terminus , under conditions of DD , LL and LP60min . The data show that CSP-6 can interact with WC-1 to regulate WC-1 phosphorylation under all tested conditions ( Fig 2D ) . Additionally , although we noticed a reduced level of WC-2 in Δcsp-6 , the phosphorylation level of WC-2 was not significantly changed in the Δcsp-6 mutant indicating that CSP-6 more specifically targets WC-1 ( S3E Fig ) . Interaction was also detected between CSP-6 and WC-2 ( Fig 2D ) , although we expect this is most likely indirect , reflecting the interaction between CSP-6 and WC-1 which heterodimerizes with WC-2 . In CSP-6 , only the HAD phosphatase domain is predicted to dephosphorylate its substrates ( S1A Fig ) . To determine whether the HAD domain was sufficient to rescue defects caused by deletion of csp-6 , we transferred a construct bearing only the HAD domain ( csp-6HAD , amino acids 222–388 , under the control of 1 . 5 kb of the native csp-6 promoter ) to the csr locus . The resulting transformants failed to produce overt rhythmicity on race tubes while the full length csp-6 construct successfully complemented the banding defect though with slightly reduced growth rate ( Fig 3A ) . To exclude the possibility that the N-terminal region of csp-6 might be sufficient for circadian function , we also generated construct csp-6ΔHAD ( amino acids 1–221 ) , lacking the HAD domain , and transformed it into Δcsp-6 . The resulting transformants were as arrhythmic as Δcsp-6 on race tubes . However , one interesting phenomenon emerged when transformants were examined under conditions of glucose depletion . After a few days , transformants bearing csp-6HAD showed weak conidiation rhythms whereas no banding was detected in Δcsp-6 under the same conditions ( Fig 3B ) . To confirm core clock function in csp-6HAD we used the frq c-box-luc transcriptional reporter , showing that csp-6HAD had the delayed phase seen with Δcsp-6 but with more robust rhythmicity than that of Δcsp-6 . ( Fig 3C ) . These results indicated that the HAD domain itself could partially rescue clock defects caused by loss function of CSP-6 under tested conditions . HAD phosphatases , which have essential Asp residues in their catalytic domains , are emerging as a large family existing in plants , prokaryotes and mammals . Their conserved active sites have a consensus sequence hhhDxDx ( T/V ) ( L/V ) h , where h represents a hydrophobic residue , and x indicates any amino acid [38] , the two aspartates coordinating the essential Mg2+ in the active site . In the yeast , mutation of DXDX ( T/V ) motif cannot functionally complement the psr1/psr2 mutant . It is essential for its sodium stress response suggesting that mutation of DXDX disrupted its phosphatase activity [37] . To investigate if the DxDx motif ( DLDE ) in CSP-6 that is essential for phosphatase activity is also essential for its circadian function ( S2A Fig ) , we generated a csp-6ΔDLDE construct and transferred it into the csr locus of a Δcsp-6 mutant . Race tubes of csp-6ΔDLDE failed to show overt rhythmicity with or without glucose ( Fig 3A and 3B ) . Western blot analyses showed that compared to the wild type , FRQ protein along with phosphorylation level of csp-6ΔDLDE was rhythmic as that shown in Δcsp-6 ( Fig 3D ) . However , similar to Δcsp-6 , the FRQ level csp-6ΔDLDE was reduced significantly after moving to dark , which is consistent to the lower level WC-1 in csp-6ΔDLDE ( Fig 1D , Fig 3D and 3E ) . Additionally , we detected hyperphosphorylated WC-1 in csp-6ΔDLDE under constant dark and light ( Fig 3E , S3A Fig ) . These data confirmed that the conserved DLDE motif essential for phosphatase activity was critical for CSP-6 circadian activity . The Δcsp-6 strain displayed a four-hour phase delay , as well as a reduced level of WC-1 that is hyperphosphorylated ( Figs 1 and 2 ) , all similar to phenotypes observed in the Δvvd mutant [26] . Evidence for genetic interaction between csp-6 and vvd was seen when Δcsp-6 , Δvvd , ras-1bd , csr::c-box-luc cultures growing on minimal slants showed enhanced carotenoid accumulation ( S3B Fig ) , and this was confirmed at the protein level when Western blot analyses showed much less WC-1 in the double mutant especially in the light , indicating that deletion of csp-6 and vvd together had synergistic effects compared to the individual mutants ( Fig 4A ) . Surprisingly , even with dramatically reduced WC-1 levels , the luciferase activity driven by frq c-box in the Δcsp-6 , Δvvd double mutant appeared robust . Furthermore , a nearly 8 hr phase delay was detected in the double mutant and rhythmicity was dampened after three cycles , a plainly additive or synergistic effect as neither was observed in the single mutants Δcsp-6 or Δvvd ( Fig 4B ) . These data suggest that CSP-6 and VVD contribute to the separate but parallel pathways , based on the synergistic phenotype of double mutants . VVD governs photoadaptation and influences light responses [26] and does so by interacting physically with WC-1 [11] . Because weaker light-induction and delayed photoadaptation was detected in the Δcsp-6 mutant ( Fig 2C ) , we hypothesized that a disruption of the interaction between VVD and WC-1 could be a contributing factor . To test this , we used a vvdV5 , Δcsp-6 strain and performed a Co-IP assay between VVD and WC-1 , adjusting the input amounts to 3 times ( 6mg ) more than that used in the wild type ( 2mg ) so that equivalent WC-1 was present in all samples ( Fig 4C , S4A Fig ) . These data indicate a substantial reduction in the amount of WC-1 interacting with VVD in the Δcsp-6 mutant compared to wild type , even with comparable amounts of WC-1 input ( Fig 4C ) . This reduced interaction between VVD and WC-1 might result in the weaker light response seen in the Δcsp-6 mutant , and combined with the reduced level of WC-1 seen in Δcsp-6 could also underlie the additive phase delay . To test this hypothesis we used a strain in which wc-1 expression was driven by the inducible qa-2 promoter at the native wc-1 locus [14] in the Δcsp-6 , ras-1bd , csr::C-box-luc background . Under quinic acid ( QA ) induction ( 10-2M QA ) , high levels of WC-1 comparable to wild type can be expressed constitutively in the Δcsp-6 mutant ( S4B Fig ) and hyperphosphorylated WC-1 can be detected as well in the qawc-1 , Δcsp-6 strains . The wc-1::qa:wc-1 , Δcsp-6 strain was arrhythmic absent inducer and rhythmic with inducer for the first 3 cycles and the enhanced WC-1 expression in wc-1::qa:wc-1 , Δcsp-6 substantially rescued the phase delay phenotype , reducing the 4 hr phase delay to around two hours ( Fig 4D ) . To exclude the possibility that quinic acid caused the dampening in wc-1::qa:wc-1 , Δcsp-6 , we also tested the rhythmicity of wc-1::qa:wc-1 grown with 10-2M QA; the results showed that wc-1::qa:wc-1 displayed robust rhythmicity through tested 5 days ( S4C Fig ) . Although the QA-induced increase in WC-1 protein level in the Δcsp-6 mutant largely rescued the phase delay phenotype , race tube analysis showed that QA-induced WC-1 failed to rescue the overt conidiation rhythm in Δcsp-6 ( S4D Fig ) , indicating that downstream WCC targets or certain clock-control genes regulating circadian output were misregulated in the Δcsp-6 mutant and this most likely was caused by the hyperphosphorylated WC-1 but not the reduced WC-1 amount . To identify CSP-6-associated proteins , CSP-6 was epitope tagged with VHF ( V5 , His and FLAG tandem tag ) and used to purify CSP-6 by tandem affinity purification using FLAG agarose followed by V5 magnetic beads . Mass spectrometric analysis showed that the four bands ( labeled as 1–4 from silver staining ) clustered together around 70kDa are all CSP-6 , and the one band below was identified as a weak ortholog ( BLASTP e-11 ) of Saccharomyces WHI-2 encoded by NCU10518 ( Fig 5A ) . Two translational start sites were found in the csp-6 5’UTR suggesting that CSP-6 has two protein isoforms with a size difference of 5 . 5 kDa ( S5A Fig ) . We also performed phosphatase treatment on V5-purified CSP-6 and the results showed that the proteins can apparently be dephosphorylated indicating that , like its Saccharomyces ortholog Psr1p , post-translational modification occurs to CSP-6 after it is synthesized ( S5B Fig ) . Saccharomyces Whi2p is a general stress regulator protein known to interact physically and genetically with Psr1p and Psr2p , the ortholog and paralog of CSP-6 respectively , and is believed to activate them; knockouts of any members of the Psr ( s ) /Whi-2 complex in yeast share similar phenotypes [39 , 40] . We confirmed the interaction between CSP-6 and WHI-2 in Neurospora by performing Co-IP ( Fig 5B ) , and then used a Δwhi-2 strain obtained from the Neurospora knockout collection [41] to ask whether WHI-2 played a role in the circadian system . Δwhi-2 showed growth , morphology , and circadian phenotypes similar to Δcsp-6 including slow growth , reduced conidiation , circadian output defects on race tubes , a phase delay in the core clock ( assayed by frq c-box-luc reporter ) , and increased WC-1 phosphorylation level ( Fig 5C–5E , S6A Fig ) . However , in most cases the defects in Δwhi-2 were not as severe and were in all cases hypostatic to those seen in Δcsp-6: e . g . , the phase delay in Δwhi-2 was not as severe as Δcsp-6 ( Fig 5F ) , and WC-1 protein amounts were not reduced significantly in the Δwhi-2 mutant ( S6A Fig ) . The double mutant Δcsp-6 , Δwhi-2 had WC-1 levels similar to Δcsp-6 , and no further hyperphosphorylation of WC-1 was detected in the Δcsp-6 , Δwhi-2 double mutant ( S6B Fig ) compared to Δcsp-6 . These results suggest that in the CSP-6/WHI-2 complex , CSP-6 plays a major role in regulating circadian related phenotypes including phase and conidiation rhythmicity , while WHI-2 is more like an assistant to fully activate CSP-6 . Although driving WC-1 protein amounts to wild type levels can partially rescue the phase delay phenotype of Δcsp-6 mutant ( Fig 4D ) , it failed to rescue circadian rhythmicity on the race tubes ( S4D Fig ) . These data indicated a break in circadian output control downstream from the WCC in the absence of CSP-6 , so we sought the source of the break in control . Because in Saccharomyces Psr1p functions together with Whi2p to activate stress responses and mediate gene expression through the stress-responsive transcription factor MSN2p [39] , we asked whether the ortholog of yeast MSN2 in N . crassa , that is MSN-1 , a cutinase G-box binding protein encoded by NCU02671 , was involved in clock output . The Δmsn-1 mutant , however , displayed normal circadian banding and period length ( though with a reduced growth rate; S7A Fig ) and frq c-box-luc reporter assays showed that Δmsn-1 had a robust circadian rhythm ( S7B Fig ) , all data indicating that in Neurospora CSP-6 does not act through msn-1 to regulate circadian output . Because Ghosh et al ( 2014 ) had suggested that csp-6 might act in the same pathway as csp-1 [42] to regulate growth and conidiation , we also confirmed that transcriptional rhythmicity of csp-1 was not affected in the Δcsp-6 mutant ( S7C Fig ) . After this and following the genetic principle of epistasis , we screened transcription factors known to be targets of the WCC [21 , 22] for circadian output defects , showing regulation of fluffy ( a major regulator of conidiation ) to be still weakly rhythmic in Δcsp-6 ( S7D Fig ) before rediscovering that deletion of adv-1 ( NCU07392 ) results in defects similar to Δcsp-6 ( Fig 6A ) [22] . Though deletion of csp-6 did not affect the oscillation of csp-1 and fluffy promoter activity , we found that the transcriptional expression level of csp-1 and fluffy was reduced in Δcsp-6 compared to that in the wide type ( S7E Fig ) . ADV-1 is a transcription factor previously shown broadly to affect development [41] , to be robustly regulated by light [21] and the clock , and to be required for the overt rhythm in conidiation on race tubes [22 , 33] . Consistent with prior data [22] the frq-luc reporter remained robustly rhythmic in Δadv-1 indicating that ADV-1-regulated circadian output does not impact the core oscillation ( Fig 6B ) but importantly , by comparing adv-1 transcriptional activity in wild type and Δcsp-6 backgrounds , it is clear that the normally rhythmic transcription of adv-1 is lost in Δcsp-6 ( Fig 6C ) . Because adv-1 is a direct downstream target of WC-1 [22] , we hypothesized that the arrhythmicity of adv-1/ADV-1 in Δcsp-6 might be caused by the misregulation of WCC binding efficiency at the adv-1 promoter . ChIP assays using WC-2 antibody were performed across a time course from 4h to 24h in darkness . The results showed that rhythmic WCC-binding at adv-1 promoter sites was disrupted in the Δcsp-6 mutant as compared to wild type ( Fig 6D ) . Binding was also attenuated following light pulses ( LP15min ) ( S8A Fig ) . The loss of circadian regulation of ADV-1 in Δcsp-6 was also confirmed at the protein level ( Fig 6E ) demonstrating that CSP-6 is essential for rhythmicity of adv-1/ADV1-1 at both transcriptional and translational levels . Because of the disruption of WCC binding to the adv-1 promoter elicited by loss of csp-6 we asked whether there was a corresponding disruption of oscillator-relevant binding to the C-box within the frq promoter in Δcsp-6 ( S8B and S8C Fig ) ; overall binding was reduced roughly half in first 24h in darkness and even more significantly reduced WCC binding was detected in day two from 28-48h in darkness . Consistent with the frq-luc luciferase data in Fig 1B , deletion of csp-6 did not affect the rhythmicity of FRQ , but its robust expression . These data again suggest that the phosphorylation status of WC-1 as impacted by loss of CSP-6 has a discrete effect on circadian output , and moreover the effect may not be only on the ability of WC-1 to bind to the adv-1 promoter but also on its ability to activate expression from it . We asked separately whether CSP-6 can physically interact with ADV-1 to regulate its transcriptional and translational activity and confirmed via pull down assays using an epitope tagged adv-1V5 , csp-6FLAG strain that no direct interaction between CSP-6 and ADV-1 was detected under the conditions used ( S9A Fig ) . In addition , CSP-6 does not regulate the light response of ADV-1 though slightly reduced ADV-1 protein level was detected in Δcsp-6 ( S9B Fig ) . A dephosphorylation assay was also performed to further confirm that CSP-6 did not function on ADV-1 directly as a phosphatase ( S9C Fig ) . Taken altogether these data are most easily interpreted as showing that hyperphosphorylation of WC-1 caused by loss of CSP-6 reduced the binding efficacy of the WCC at the adv-1 promoter and that CSP-6 regulates circadian output and light-regulation via impacting transcription of adv-1 through the WCC . However , in addition to disruption of rhythmic WCC binding to the adv-1 promoter , reduced interaction between ADV-1 and WC-1 was also detected by Co-IP ( Fig 6F ) ; these experiments used three times more IP protein from the adv-1V5 , Δcsp-6 ( 6mg ) strain as compared to adv-1V5 ( 2mg ) to make up for the reduced WC-1 expression seen in adv-1V5 , Δcsp-6 . This result confirmed that loss of csp-6 substantially interrupted the interaction between WC-1 and ADV-1 . We then examined the subcellular distribution of ADV-1 in cultures grown in LL . Aliquots of total , cytosol and nuclear fractions were analyzed by Western blotting ( Fig 6G ) . Two proteins , tubulin and histone-H3 were used as cytoplasmic and nuclear protein markers , respectively , to confirm the quality of the nuclei and to control for cytoplasmic contamination in the nuclear preparation . ADV-1 was enriched in nuclei , which was consistent with its function as a transcription factor ( Fig 6G ) . Interestingly however , although slightly less total ADV-1 was detected in Δcsp-6 , more of it was enriched in nuclei , suggesting that loss of csp-6 resulted in mildly misregulated localization such that ADV-1 was more strongly partitioned to nuclei rather than to the cytoplasm . Furthermore , the Δcsp-6 mutant displayed only slightly less ADV-1 protein and no difference in ADV-1 phosphorylation as compared to WT , suggesting that CSP-6 did not directly work on ADV-1 as a phosphatase ( S9B Fig ) . Therefore , the reduced interaction between ADV-1 and WC-1 , and increased amount of nuclear ADV-1 in Δcsp-6 ( Fig 6F and 6G ) may reflect the action of CSP-6 on WC-1 rather than directly on ADV-1 . As reflected in the gene name , Saccharomyces PSR1 ( Plasma membrane Sodium Response 1 ) localizes to the plasma membrane [37 , 39] and we were interested to know whether CSP-6 has similar localization in Neurospora . Nuclei were isolated from csp-6V5 , ras-1bd and the cytoplasmic and nuclear fractions analyzed by SDS-PAGE [43] ( Fig 7A ) . CSP-6 was enriched in the nuclear fraction but was still present in the cytoplasm . Therefore , unlike yeast , Neurospora CSP-6 localized to both the cytoplasm and nucleus . Sequence analysis of CSP-6 revealed a putative nuclear localization signal ‘PKKKKG’ ( 9-14aa ) near the N-terminus that is absent from yeast Psr1p ( S10 Fig ) . To determine whether ‘PKKKKG’ is a nuclear localization signal ( NLS ) in CSP-6 , ‘PKKKKG’ was replaced by ‘AAAAAA’ ( 6A ) and the construct ( including a C-terminal V5 tag ) transformed into the csr locus of Δcsp-6 driven by its own promoter . The resulting transformant ( Δcsp-6 , csr:: csp-66A , V5 ) rescued the banding defect and growth rate of the Δcsp-6 mutant ( Fig 7B ) ; however , ‘PKKKKG’ with 6A does not abolish the nuclear localization of CSP-6 indicating that PKKKKG is not a NLS ( Fig 7A ) . In addition , we noticed that CSP-6 was reduced significantly in amount in Δcsp-6 , csr:: csp-66A , V5 suggesting PKKKKG may be essential for CSP-6 stability though not as a NLS . Cellular fractionation showed that WHI-2 displayed localization similar to that of CSP-6 ( Fig 7C ) .
csp-6 encodes a phosphatase required for Neurospora’s major overt circadian output , the daily cycle of asexual development . However , CSP-6 is constitutively expressed over the day without obvious rhythmicity and is thus not a typical clock-controlled gene [30] . Previous studies associated two protein phosphatases PP1 and PP2A , with the Neurospora core circadian oscillator , PP1 implicated in regulating stability of FRQ and PP2A impacting negative feedback and acting on WC-1 in vitro [18] . CSP-6 plainly has a major effect on WC-1 phosphorylation; however , deletion of csp-6 has relatively less effect on core clock oscillations ( a reduced amplitude and delayed phase but normal period length ) , instead only having a significant discrete effect on circadian output including phase setting and conidiation rhythms . Because of the high likelihood that WC-1 appears to be a direct target of CSP-6 in vivo , interpretation of its actions is revealing both in the context of the oscillator and of output and these are addressed sequentially . The Neurospora blue light photoreceptor and clock protein WC-1 , in association with WC-2 , regulates expression and oscillation of FRQ . FRQ then undergoes a cycle of phosphorylation that eventually impacts its ability to interact with WCC [44] . Biochemical analyses of cell extracts have suggested that dephosphorylation of WCC may enhance its DNA binding activity to the frq promoter [13 , 2 , 24]; however , these effects could be mediated by other proteins in the extracts such as FRQ whose phosphorylation status affects WCC’s activity . Similarly , hyperphosphorylated WCC observed in cell extracts of phosphatase mutants including PP1 , PP2A was reported to have a reduced binding activity to the frq promoter , although again is not possible to say whether the effect on WCC is direct or via interacting proteins . In another correlation , activation of WCC was reported to be dependent on RGB-1 , a regulatory subunit of PP2A , and was correlated with dephosphorylation of WC-1 and WC-2 , data supporting a model in which the rhythmic activity of WCC is controlled by a dynamic equilibrium of phosphorylation and dephosphorylation mediated by several kinases and phosphatases [45] . In all these studies , then , a correlation was developed between phosphorylation of WCC and reduced activity of WCC . However , direct in vivo interaction between these phosphatases and WC-1 has not been demonstrated , correlation does not establish cause and effect , and data presented here are not entirely consistent with this model for WC-1 regulation . We saw significantly hyperphosphorylated WC-1 in the strain lacking CSP-6 , a phosphatase protein that interacts with WC-1 ( Fig 2 ) . This suggests that CSP-6 dephosphorylates WC-1 to a significant degree , and yet the clock still functions with no period lengthening ( Fig 1 ) . WC-2 ChIP analysis in the Δcsp-6 deletion mutant revealed moderately ( 1 . 5–2 fold ) reduced WC-2 binding occupancy at the frq promoter C-box site as compared to wild type but no loss of rhythm ( S9 Fig ) . The reduced binding is roughly consistent with previous work correlating hyperphosphorylated WCC with reduced DNA binding activity at frq C-box promoter [10 , 13]; however , the WC-1 hyperphosphorylation plainly has little effect on the core clock feedback loop . Previous reports had suggested that transcriptionally active hypophosphorylated WCCs are unstable and that active WCC leads to very low WCC levels [13 , 16 , 24]; however , here we saw the reverse , where hyperphosphorylated WCC present in Δcsp-6 is much less stable ( Fig 2B ) but still binds to DNA and supports a robust clock at least for the first few days though not perfectly as that in the wild type . Additionally , we noticed that the FRQ protein level and frq mRNA expression level were reduced after moving to constant darkness suggesting that CSP-6 affected the expression of FRQ though not enough to disrupt the clock oscillation ( S1B and S1C Fig ) . Different from other phosphatase proteins so far examined such as PP1 , PP2A , PPP-1 or PP4 , CSP-6 did not regulate the phosphorylation level of FRQ . Furthermore , we also examined WCC binding to the c-box region over two days . The results showed reduced binding at frq C-box in Δcsp-6 compared to the wild type ( S8B and S8C Fig ) , so most likely CSP-6 plays a role in maintaining robust FRQ expression through aiding WC-1 binding activity [18 , 46] . It should be noted , however , that the hyperphosphorylated WC-1 seen in Δcsp-6 still binds to the frq promoter and still drives a circadian clock . Loss of CSP-6 resulted in an around 4 hr phase delay in the rhythm , a phenotype similar to that seen strains lacking VVD , the small blue light photoreceptor protein consisting of LOV domain and an N-terminal cap that physically interacts with WC-1 to reduce its ability to activate transcription to regulate photoadaptation [26 , 27] . Indeed , an additive phase delay was observed in the double mutant Δvvd , Δcsp-6 ( Fig 4B ) , and there was plainly much less WC-1 in Δvvd , Δcsp-6 compared to either single mutant , suggesting that the low level of WC-1 seen in Δcsp-6 might be the reason for the phase delay . However , elevated expression of WC-1 in wc-1::qa-wc-1 , Δcsp-6 only partially rescued the phase defect ( Fig 4D ) , and this suggested that the reduced VVD-WC-1 interaction also observed in the Δcsp-6 mutant , independent of the low level of WC-1 , might also underlie the phase delay . This seems to be the case , because even the addition of three times more WC-1 in immunoprecipitations from Δcsp-6 ( to make up for the reduced WC-1 level in this strain ) failed to recover a full level of interaction ( Fig 4C , S4A Fig ) . Therefore we hypothesize that hyperphosphorylation of WC-1 might independently impact the interaction between VVD and WC-1 resulting in the phase delay and weaker light response phenotypes . WC-1 and WC-2 play multiple roles in the circadian system; their protein levels contribute to the robustness and stability of the clock and they are at the top of the hierarchy of transcription factors that governs circadian output . The importance of the WCC to output was demonstrated by Cheng et al . ( 2001 ) who examined the strains qa-WC-1 or qa-WC-2 in which ORFs of WC-1 or WC-2 were under the control of quinic acid-inducible promoter ( qa-2 ) ; these strains were arrhythmic on race tubes when the QA concentration was less than 1X10-7M ( corresponding to << 10% of wt WC-1 levels ) , but the conidiation rhythm became overt and robust as inducer was increased to yield even 30% of normal levels . Our initial observation of less WC-1 in the Δcsp-6 mutant suggested that this was the cause of the loss of the overt rhythm; however , we were unable to find evidence for limiting WC-1 . Westerns ( Fig 6E ) showed ADV-1 levels in Δcsp-6 comparable to wt and light-induction of WCC target genes was only slightly reduced , not severely reduced as is seen when WC-1 becomes truly limiting as in the wc-1[MK1] allele ( 56 ) . When even full induction of WC-1 in qa-wc-1 failed to rescue the conidiation rhythmicity ( S4B and S4C Fig ) a role for CSP-6 specifically in output was suggested . This was supported by finding that the core oscillator runs with a normal period length in the Δcsp-6 mutant , although the amplitude of the daily cycle in WCC binding to frq , the positive arm in the cycle was reduced compared to wide type ( S8B and S8C Fig ) . Precedents exist for target gene-specific effects of post-translational modifications of transcription factors , for instance in mutants defective in phosphorylation of Ser ( 276 ) of NF-kB subunit p65 [47 , 48] . Circadian output is a consequence of the negative feedback of the central clock that results in the daily cycle of WCC activity . Output happens when the WCC drives expression of clock-controlled genes ( ccgs ) whose products do not impact the core oscillator itself [30 , 49] . A number of ccgs have been identified as involved in circadian output regulation [50–53] including genes encoding transcription factors ( TFs ) directly controlled by WCC involved in clock and light regulation . From these data emerged the model where WCC sits atop an interconnected hierarchy of TFs that governs light and clock regulation [22] . The results of our study indicate that the transcriptional activity of some TFs within this hierarchy that are required for conidiation rhythms and are downstream of CSP-6-WC-1 were disrupted by deletion of csp-6 . Examination of candidate TF genes including csp-1 , fluffy , msn-1 , and adv-1 revealed that both transcriptional and translational rhythmicity of adv-1/ADV-1 was significantly disrupted in the Δcsp-6 mutant ( Fig 6C and 6E , S7 Fig ) . We confirmed previous results showing arrhythmicity of Δadv-1 by race tube assay though with normal core clock function [22] as well as rhythmic WCC binding at the promoter of adv-1 [32] that we here show is weakened in Δcsp-6 ( Fig 6D ) . These data lead to the surprising conclusion that the phosphorylation status of WC-1 differentially affects its two functions: Loss of CSP-6-mediated dephosphorylation has little impact on WC-1 action in the core circadian oscillator but it abrogates the ability of WCC to regulate a salient circadian output , the daily cycle of conidiation . We also noticed that though the rhythmic expression of csp-1 and fluffy promoter was not affected by deletion of csp-6 , the mRNA expression level was reduced for both genes in Δcsp-6 ( S7E Fig ) , further suggesting that CSP-6 mediated dephosphorylation of WC-1 has distinct role in output through ADV-1 , but also generally affects the expression of other downstream targets . Based on these data , a working model is summarized in Fig 8 . CSP-6 physically interacts with and dephosphorylates WC-1 in vivo so that it can interact with VVD to regulate photoadaptation and phase resetting . WHI-2 , as an assisting protein , associates with CSP-6 to adjust the WC-1 protein amount and phosphorylation level . adv-1 , as a ccg , is one of the direct targets of the WCC that regulates circadian output . The promoter of adv-1 is rhythmically recruited by WCC , and ADV-1 directly functions downstream of CSP-6/WHI-2/WC-1 to control overt rhythmic conidiation on race tubes . In contrast , hyperphosphorylated WC-1 seen in the Δcsp-6 deletion mutant shows disrupted rhythmic DNA binding activity at the adv-1 promoter and arrhythmic transcription/translation of adv-1/ADV-1 , the essential cause of the circadian output defect in the Δcsp-6 mutant . However , hyperphosphorylated WCC in Δcsp-6 still sufficiently drives rhythmic frq expression for the clock to run , though with reduced FRQ levels indicating that CSP-6 plays a role in maintaining robust frq/FRQ expression . These data present some unexpected paradoxes and compel a more nuanced view of the role of phosphorylation of WC-1 in the core clock and in output . The existing model of the circadian feedback loop posits that WC-1 is active and unstable early in the circadian cycle prior to its undergoing changes in phosphorylation that cause it to bind less tightly to DNA . The feedback loop closes when , because of these alterations in the phosphorylation state , WC-1 becomes stable and inactive . It is plausible that the differing combinations of phosphorylation states seen at times throughout the circadian cycle could differentially affect sequence-specific binding affinities of WC-1 to target genes , thereby modulating amplitudes in a hierarchical manner such that some genes are only modestly affected whereas others are more severely affected . Hence , we cannot rule out that the effects on output–specifically the loss of the overt developmental rhythm–arising from loss of CSP-6 are the result of the reduced amount of WC-1 seen in this strain . However , this seems less likely because the reduction in the absolute levels of both WC-1 ( including all of it isoforms , Figs 2A and 4A ) and ADV-1 ( Fig 6E , 6F and 6G ) , each on the order of three-fold , are not so severe . Instead it seems that it is the quality of WC-1 and not its quantity that is important . Implicit in this statement is the prediction of different classes of phosphosites in WC-1 , some involved more directly in oscillator function and some more in output so that output , more than the oscillator itself , is modulated based on the ability of CSP-6 to dephosphorylate WC-1 . CSP-6 is plainly required to dephosphorylate WC-1 because WC-1 is hyperphosphorylated in Δcsp-6 . The existing model of the circadian feedback loop posits that WC-1 is active and unstable early in the circadian cycle when it is hypophosphorylated , and the feedback loop closes when , because of hyperphosphorylation , WC-1 becomes stable and inactive . Here , paradoxically , WC-1 is hyperphosphorylated , active , and unstable , and the clock runs , indicating that the class of phosphosites normally dephosphorylated by CSP-6 is distinct from the phosphosites mediating closure of the feedback loop . Different combinatorial states of WC-1 phosphorylation affecting DNA affinity at different promoters , along with turnover , replenishment , and inactivation , are all likely to contribute to distinguishing between oscillator and output functions of WC-1 .
The ras-1bd and 74A strains were used as clock WT strains in this study . The Δcsp-6 , Δpsr-2 , Δwhi-2 , Δadv-1 , Δvvd strains were obtained from the Fungal Genetics Stock Center [41] . These KO strains were backcrossed to ras-1bd to obtain band phenotype for race tube assays . The newly created double knock-out strains were Δcsp-6 , Δpsr-2 and Δcsp-6 , Δwhi-2 in background #1497 ( mus52::natamycin ) . Neurospora transformation was done as previously described [41] . Race tube medium contained 1xVogel’s salts , 0 . 1% glucose , 0 . 17% arginine , 50ng/mL biotin and 1 . 5% ( w/v ) agar . Race tube assays were carried out as previously described [54] . Liquid cultures were grown in medium containing 1xVogel’s , 0 . 5% arginine , and 50ng/mL biotin with 2% glucose . Protein extraction , quantification , Western blot analysis and Co-IP were performed as described previously [55 , 56] . For Western blot analysis , equal amounts of total proteins ( 30 μg ) were loaded to protein gels that were transferred to PVDF membrane after electrophoresis . The V5 antibody ( Invitrogen , NY ) was used at dilution of 1:5000 . Other antibodies including antisera directed at WC-1 , WC-2 , and FRQ were generated by our own lab [57] . For Co-IP , 2mg total protein was incubated with 30 μl V5 agarose beads ( Sigma , MO ) for 2h to overnight at 4°C in PEB buffer ( 50mM HEPES , pH7 . 4 , 150mM NaCl , 10% glycerol , 0 . 4% NP-40 ) . The V5 agarose beads were washed with wash buffer ( 50mM HEPES , pH7 . 4 , 150mM NaCl , 0 . 4% NP-40 ) four times and eluted with 4xLDS buffer ( Thermo Fisher , MA ) at 95°C for 5min . In testing the interaction between WC-1 and VVD in Δcsp-6 , because of the reduced WC-1 level detected in the Δcsp-6 mutant and to make sure that similar amounts of WC-1 were available based on the quantification by western blot ( S4A Fig ) we used three fold more input of vvdV5 , Δcsp-6 than of vvdV5 , and of the wild type ( ras-1bd ) . The same treatment was performed for the interaction between WC-1 and ADV-1 with DSP crosslink as described previously [11] . For protein purified by tandem affinity tag , total protein was isolated from 10–15 g of fresh tissue and incubated with FLAG agarose beads first , followed by V5 magnetic beads . A small amount of the final V5 precipitates were separated by SDS-PAGE and the gel was silver-stained followed manufacturer’s instruction for purification quality examination ( SilverQuest , Invitrogen ) . For Mass Spectrometry , the remainder of the V5 precipitate preparation was separated by SDS-PAGE and the specific bands were excised from a Coomassie blue stained gel and sent for Mass Spectrometry as previously reported [44] . Nuclear preparation was performed as reported [58] . Total RNA was isolated by Trizol according to the manufacturer’s protocol ( 15596–026; Invitrogen ) . For quantitative RT-PCR , 2–3 μg RNA were treated with DNase at 37°C for 60min and then incubated with inactivation buffer for 5 min following instructions ( AM2239; Life Technology ) . cDNA was generated by reverse transcription reaction according to the manufacturer’s protocol ( 18080–051; Invitrogen ) . Expression levels of genes of interest were analyzed by quantitative real-time PCR with primers listed in S1 Table in the Supplemental Material . The luciferase reporter assay was performed as described previously [36] . The C-box-luc , csr::C-box-luc , ras-1bd , A or his-3::pfrq-luc , ras-1bd , A/a were used as control strains . Knock out strains were crossed with these as appropriate to place frq-luc at the csr or his-3 locus . Camera runs showed there was no difference in rhythmicity of the frq-luc reporter at csr versus the his-3 locus . Race tube medium was used for luciferase assays and 0 . 01M quinic acid ( QA ) was added for qa-2 promoter-driven strains as appropriate . All cultures were grown in LL for 2 days and then transferred to constant darkness and luminescence was recorded every hour for six days . ChIP assays were performed as described previously [22 , 59] . Briefly , the Neurospora tissues were fixed with 1% formaldehyde for 15 min and quenched by glycine at final concentration of 125mM for 5 min . Around fifty-mg of cross-linked tissue was used for each sample and were suspended in 500 μl ChIP lysis buffer . Chromatin was sheared by sonication to 100–500 bp fragments . The immunoprecipitation was performed using 5μl WC-2 antibody [57] . Immunoprecipitated DNA was quantified using real time PCR with primer sets listed in S1 Table . ChIP quantitative PCR data were normalized to a sample of input DNA as described in instructions from the Life Technology website ( https://www . lifetechnologies . com/us/en/home/life-science/epigenetics-noncoding-rna-research/chromatin-remodeling/chromatin-immunoprecipitation-chip/chip-analysis . html ) . Each experiment was independently performed at least three times . | Though molecules and components in the core circadian oscillator are well studied in Neurospora , the mechanisms through which output pathways are coupled with core components are less well understood . In this study we investigated a HAD phosphatase , CSP-6; loss-of-function Δcsp-6 strains are overtly arrhythmic but have a functional core circadian oscillation . CSP-6 in association with WHI-2 dephosphorylates the core clock component WC-1 to regulate light-responses and development . To dissect the functions of CSP-6 in core clock and output , we screened known WC-1 targets and found that loss of CSP-6 causes misregulation of transcriptional/translational activation of ADV-1 , a key regulator of output . Thus , loss of CSP-6-mediated dephosphorylation of WC-1 leads to loss of ADV-1 activation and is responsible for the complete loss of overt developmental rhythmicity in Δcsp-6 . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"phosphorylation",
"luciferase",
"enzymes",
"gene",
"regulation",
"regulatory",
"proteins",
"enzymology",
"dna-binding",
"proteins",
"light",
"electromagnetic",
"radiation",
"phosphatases",
"circadian",
"oscillators",
"transcription",
"factors",
"chronobiology",
"proteins",
"oxidoreductases",
"gene",
"expression",
"light",
"pulses",
"physics",
"biochemistry",
"circadian",
"rhythms",
"post-translational",
"modification",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences"
]
| 2018 | A HAD family phosphatase CSP-6 regulates the circadian output pathway in Neurospora crassa |
The importance of a mesoscopic description level of the brain has now been well established . Rate based models are widely used , but have limitations . Recently , several extremely efficient population-level methods have been proposed that go beyond the characterization of a population in terms of a single variable . Here , we present a method for simulating neural populations based on two dimensional ( 2D ) point spiking neuron models that defines the state of the population in terms of a density function over the neural state space . Our method differs in that we do not make the diffusion approximation , nor do we reduce the state space to a single dimension ( 1D ) . We do not hard code the neural model , but read in a grid describing its state space in the relevant simulation region . Novel models can be studied without even recompiling the code . The method is highly modular: variations of the deterministic neural dynamics and the stochastic process can be investigated independently . Currently , there is a trend to reduce complex high dimensional neuron models to 2D ones as they offer a rich dynamical repertoire that is not available in 1D , such as limit cycles . We will demonstrate that our method is ideally suited to investigate noise in such systems , replicating results obtained in the diffusion limit and generalizing them to a regime of large jumps . The joint probability density function is much more informative than 1D marginals , and we will argue that the study of 2D systems subject to noise is important complementary to 1D systems .
The population or mesoscopic level is now recognised as a very important description level for brain dynamics . Traditionally rate based models [1] have been used: models that characterize the state of a population by a single variable . There are inherent limitations to this approach , for example a poor replication of transient dynamics that is observed in simulations of spiking neurons , and various groups have proposed a population density approach . Density methods start from individual point model neurons , consider their state space , and define a density function over this space . The density function characterizes how individual neurons of a population are distributed over state space . These methods have been used successfully for one dimensional point model neurons , i . e . models characterized by a single state variable , usually membrane potential . Such models , e . g . based on leaky- ( LIF ) or quadratic-integrate-and-fire ( QIF ) , exponential-integrate-and-fire neurons , have a long-standing tradition in neuroscience [2–6] . Related approaches consider densities of quantities such as the time since last spike [7 , 8] , but here too a single variable is considered to be too coarse grained to represent the state of a population . Recently , increased computing power and more sophisticated algorithms , e . g . [5 , 9–12] , have made the numerical solution of time dependent density equations become tractable for one dimensional neural models . In parallel , dimensional reductions of the density have been developed , usually by expressing the density in terms of a limited set of basis functions . By studying the evolution as a time-dependent weighting of this basis the dimensionality is reduced , often resulting in sets of first order non-linear differential equations , which sometimes are interpreted as ‘rate based’ models [13–15] . The one dimensional density is very tractable: membrane potential distributions and firing rates have been shown to match spiking neuron simulations accurately , particularly in the limit of infinitely large populations , at much lower computational cost than direct spiking simulations: Cain et al . [16] report a speedup of two orders of magnitude compared to a direct ( Monte Carlo ) simulation . The problem of such one dimensional models is that they leave out details that may affect the population , such as synaptic dynamics and adaptation . Mathematically , the inclusion of variables other than just the membrane potential is no problem , but this increases the dimensionality of the state space , which negates most—but not all—computational advantages that density functions have over Monte Carlo simulation . This problem has led to considerable efforts to produce effective one dimensional methods that allow the inclusion of more realistic features of neural dynamics . Cain et al . have included the effects of conductances by making synaptic effects potential dependent in an otherwise standard one dimensional paradigm . Schwalger et al . [17] consider the distribution over the last spike times of neurons . Under a quasi-renewal approximation that the probability of a neuron firing is only dependent on the last spike time and recent population activity , they are able to model the evolution of the last spike time distribution and the population activity resulting in a system of one dimensional distributions . Both groups have modeled a large-scale spiking neuron model of a cortical column , achieving impressive agreement between Monte Carlo and density methods . Another attempt to reduce the dimensionality of the problem are moment-closure methods [18] , which we will not consider here . Recently , Augustin et al have presented a method to include adaptation into a one-dimensional density approach [15] . There have been a number of studies of two dimensional densities [19–21] . They have made clear that analyzing the evolution of the joint probability density provides valuable insight in population dynamics , but they are not generic: it is not explicit that the method can be extended to other neural models without recoding the algorithm . Here , we present a generic method for simulating two dimensional densities . Unlike the vast majority of studies so far , it does not start from a Fokker-Planck assumption but starts from the master equation of a point process ( usually , but not necessarily ) Poisson , and models the joint density distribution without dimensional reduction . We believe the method is important given the trend in theoretical neuroscience to reduce complex realistic biophysical models to effective two dimensional model neurons . Adaptive-exponential-integrate-and-fire ( AdExp ) , Fitzhugh-Nagumo and Izhikevich neurons are examples of two dimensional model neurons that have been introduced as realistic reductions of more complex conductance based models . It is important to study these systems when subjected to noise . The method is extremely flexible: upon the creation of a novel neural model ( 2D ) we will be able to simulate a population subjected to synaptic input without writing a single line of new code . We require the user to present a visualization of the model in the form of the streamlines of its vector field , presented in a certain file format . Since these files can be exchanged , model exchange does not require recoding . As long as this vector field behaves reasonably—the qualification of what constitutes reasonable is a main topic of this paper—the method will be able to take it as input , and can be guaranteed to deliver sensible simulation results . The method is highly visual: it starts off with a user or stock provided visualization of a neural model , and uses computational geometry to calculate the transition matrices involved in modeling synaptic input , which is represented as a stochastic process . We will argue that with a visualization in hand one can often predict how noise will drive the system , and run a simulation to confirm these predictions . We will also show that the visualization gives a good overview of possible shapes of dynamics . The method cannot compete in speed with effective one dimensional density methods , but holds up well compared to direct spiking neuron simulations . In particular memory use is at least an order of magnitude lower than for direct simulation . As we will show , this allows the simulation of large networks on a single machine equipped with a GPGPU . Since very few assumptions are used , it can be used to examine the influence of approximations made in other methods . For example , because no diffusion approximation is made , we are able to examine the influence of strong synapses , which can lead to a marked deviation from diffusion results [11 , 12] . We can also model populations that are in partial synchrony . This work captures most one dimensional population density techniques , as they are a special case of two dimensional models , in particular the method by Cain et al . , and we also replicate results obtained in the diffusion limit as numerical solutions of Fokker-Planck equations with high precision . Although we have not tried this , theory suggests that the method should work just as well when escape noise is used [7] . With the ability to exchange neural model files , without having to recode , it is easy to check how different neural models generate dynamics in similar circuits . A software implementation of this method is available at http://miind . sf . net with a mirror repository on github https://github . com/dekamps/miind . Since this is a methods paper , the Material and Methods section contains the main result , and we will present this first so that the reader may form an understanding of how the simulation results are produced . In the results section , we will show that our method works for a number of very different neural models . We will also show that strong transients , which occur in some models as a consequence of rapidly changing input , but not in others , can be understood in geometrical terms when considering the state space of the neural model .
As an example , we consider a conductance based model with first order synaptic kinetics following [20] . It is given by: τ d V d t = − g l ( V − E l ) − g e ( t ) ( V − E e ) ( 7 ) τ e d g e d t = − g e + I syn ( t ) ( 8 ) Numerical values are taken from [20] , and given in Table 1 . Isyn ( t ) represents the influence of incoming spikes on the neurons . A conventional representation of such a model is given by a vector field , see Fig 1 . Consider a two dimensional dynamical system defined by a vector field . A point in state space will be represented by a two dimensional vector v → . A grid is constructed from strips . As mentioned previously , usually one dimension is a membrane potential , and we will denote coordinates in this dimension by a small letter v . The second dimension can be used to represent parameters such as adaptation , conductance , and will be represented by w . A strip is constructed by choosing two neighbouring points in state space , e . g . v 0 → ( t = 0 ) , v 1 → ( t = 0 ) , and integrating the vector field for a time T that is assumed to be an integer multiple of a period of time Δt , which we assume to be a defining characteristic of the grid . Let T = nΔt , then two discretized neighbouring characteristics S = { v 0 → ( t = 0 ) , ⋯ , v 0 → ( t = n Δ t ) ; v 1 → ( t = 0 ) , ⋯ , v 1 → ( t = n Δ t ) } define a strip . Within a strip , the set of points C i = { v 0 → ( t = i Δ t ) , v 0 → ( t = ( i + 1 ) Δ t ) , v 1 → ( t = ( i + 1 ) Δ t ) , v 1 → ( t = i Δ t ) } defines a cell , which is quadrilateral in shape . The quadrilateral should be simple , but not necessarily convex ( Fig 2A ) . We reject cells with less than a certain area . As we will see in concrete examples , boundaries in state space are approached through areas of vanishing measure . The area cut tends to remove complex cells , and we will reject them in general . An example of a grid generated by this procedure is given in Fig 3 . Strip numbers are arbitrary , as long as they are unique , but it is convenient to number them in order of creation . In the remainder of the paper , we will assume that strip numbers created by the integration procedure start at 1 , and are consecutive , so that the numbers i∈ {1 , ⋯ , Nstrip} with Nstrip the number of strips , each identify a unique strip . Strip no . 0 is reserved for stationary points . There may be 0 or more cells in strip 0 . The number of cells in strip i is denoted by ncell ( i ) . We refer to the tuple ( i , j ) , with i the strip number and j the cell number , as the coordinates of the bin in the grid . Ncells is the total number of cells in the grid . For all strips i ( i > 0 by construction ) , cell numbers within a strip are ordered by the dynamics: neurons that are in cell number j of strip i at time t are in cell number j + 1 mod nj of strip j at time t + Δt , where nj is the number of cells in that strip . Neurons that are in a cell in strip no . 0 are assumed to be stationary and do not move through the strip . Examples of cells in this strip are reversal bins . The handling of stationary bins will be discussed below . A simulation progresses in multiple steps of Δt , so the current simulation time tsim is specified by an integer k , defined by: t sim = k Δ t , k = 0 , 1 , 2 , ⋯ The density profile can be represented in an array M of length Ncells . Each element of this array is associated with the grid as follows . Let ccell ( 0 ) ≡ 0 and for 0 < i ≤ Nstrip let ccell ( i ) ≡ ccell ( i − 1 ) + ncell ( i − 1 ) , so ccell ( i ) represents the total number of cells in all strips up to strip i . Now define the index function I: { i=0I ( i , j , k ) =ji>0I ( i , j , k ) =ccell ( i−1 ) + ( j−k ) modncell ( j ) ( 9 ) This is a time dependent mapping: its effect is a forward motion of probability mass with each forward time step . We will refer to the updating of the mapping by incrementing k as a mass rotation as probability mass that reaches the end of a strip , will reappear at the beginning of the strip at the next time step . This effect is almost always undesirable as it would effect a jump wise displacement of probability mass . In most models this can be prevented by removing the probability mass from the beginning of each strip and setting the content of this bin to 0 , and adding the removed mass to a another bin . A typical example arises in the case of integrate-and-fire models . Here , there is usually a reversal point . Such a point can be emulated by creating a small quadrilateral , and making this cell number 0 in strip number 0 . The procedure of mapping probability mass from the beginning of a strip to special bins in state space is called a reversal mapping . It consists of a list of coordinate pairs . The first coordinate labels the bin where probability will be removed , the second coordinate labels the bin where the probability will reappear . The concept of reversal mapping extends to other neural models—we will consider adaptive-exponential-integrate-and-fire ( AdExp ) , Fitzhugh-Nagumo , and quadratic-integrate-and-fire neurons . All of these models need a prescription for what happens with the probability mass after reaching the end of a strip , and we will refer to this as the reversal mapping , even if the model does not really have a reversal bin , to contrast it from the threshold mapping . Although handling a threshold is similar , interaction with synaptic input means that the mapping requires extra precautions . We will discuss this in the section below . The whole process of advancing probability through a grid by means of updating a relationship with a grid is illustrated in Fig 3 . Up to this point we have only referred to probability mass . If a density representation is desired , one can calculate the density by: ρ ( i , j ) = M ( i , j , k ) A ( i , j ) , ( 10 ) where A ( i , j ) is the area of quadrilateral ( i , j ) , and M ( i , j , k ) is the probability mass present in the quadrilateral ( i , j ) at simulation time kΔt . We note that this procedure implements a complete numerical solution for the advective part of Eq 2 . We will assume that individual neurons will receive Poisson spike trains with a known rate for a known synaptic distribution of the post synaptic population . Without loss of generality we will limit the exposition to a single fixed synaptic efficacy; continuous distributions can be sampled by generating several matrices , one for each synaptic efficacy , and adding them together . Adding the individual matrices , which are band matrices , and very sparse , results in another band matrix , still sparse , albeit with a slightly broader band . Overall run times are hardly affected unless really broad synaptic distributions are sampled . A connection between two populations will be defined by the tuple ( Ncon , h , τdelay ) . Here Ncon is the number of connections from presynaptic neurons onto a representative neuron in the receiving population , τdelay the delay in the transmission of presynaptic spikes and h the synaptic efficacy . The firing rate ν is either given , or inferred from the state of the presynaptic population , but in both cases assumed to be known . For the population these assumptions lead to a Master equation: ∫ V d v → d ρ ( v → , t ) d t = ν { ∫ V h d v ′ → ρ ( v ′ → , t ) − ∫ V d v → ρ ( v → , t ) } , ( 11 ) where V is an area of state space and Vh the same area , translated by an amount h in dimension i . It is dependent on the neuronal model in which variable the jump takes place . In AdExp the jump is in membrane potential , in conductance based models it is in the conductance variable . Here , we will discuss the problem using conductance based neurons as an example , but the methodology applies to any model . Eq ( 11 ) determines the right hand side of Eq ( 2 ) , and the stage is set for numerical solution . The left hand side of Eq ( 2 ) describes the advective part , and is purely determined by the neuron model , which ultimately determines the grid . We already have described the movement of probability mass due to advection during a time step Δt , and need to complete this by implementing a numerical solution for Eq ( 11 ) . Eq ( 11 ) describes the transfer of probability mass from one region of state space to another . We will assume that the grid we use for the model of advection is sufficiently fine , so that the density within a single bin can be considered to be constant , and choose area V in Eq ( 11 ) to coincide with our grid bins . We approximate ( 11 ) by: d M ( i , j , k ) d t = ν { ∑ ( p , q ) ∈ C h ( i , j ) α p , q M ( p , q , k ) − M ( i , j , k ) } ( 12 ) The bin ( i , j ) translated by a distance h will cover a number of other bins of the grid . Let ( p , q ) be a bin partly covered by the translated bin ( i , j ) and let αp , q be the fraction of the surface area of the translated bin that covers bin ( p , q ) . ( By construction 0 < αp , q ≤ 1 . ) The set Ch ( i , j ) is defined as the set of tuples ( p , q ) , for all such bins , i . e . those bins that are covered by translated bin ( i , j ) ( and no others ) . We will refer to Ch ( i , j ) as the displacement set . Usually , the displacement is in one dimension only , where this is not the case we will write C h → ( i , j ) . The problem of determining Ch ( i , j ) is one of computational geometry that can be solved before simulation starts . It is illustrated in Fig 2B , where the grid of the conductance based model is shown . This problem is easily stated but hard to solve efficiently . Conceptually , a Monte Carlo approach is simplest , and since the computation can be done offline—before simulation starts—this approach is preferable . It is straightforward for a given bin of the grid ( i , j ) to generate random points that are contained within its quadrilateral . Assume these points are translated by a vector h → . It is now a matter of determining in which bin a translated point falls . In order to achieve this the grid is stored as a list of cells . Each cell , being a quadrilateral , is represented by a list of four coordinates . During construction of the grid , vertices of a cell are stored in counter clockwise order . When a quadrilateral is convex , and the vertices are stored in counter clockwise order , the × operator defined by: × ( v 1 v 2 ) ≡ ( − v 2 v 1 ) results in an “inward” pointing normal n → . If the position vector of a point has a positive scalar product with the ‘inward’ normal of all four line segments that define the quadrilateral the point is inside , otherwise it is outside . These half line tests are cheap and easy to implement . If the quadrilateral is not convex , but simple , it can be split into two triangles which are convex . We perform linear search to find a grid cell that contains the translated point , or to conclude there is no such cell . Better efficiency can be obtained with k-d trees , but we have found the generation of translation matrices not to be a bottleneck in our workflow , and linear search allows straightforward brute force parallelization . At most one cell will contain the translated point . For now , we will assume that the translated point will be inside a given bin ( p , q ) . Later , for concrete neuron models we will discuss specific ways of handling transitions falling outside the grid . If bin ( p , q ) is not represented in C h → ( i , j ) , an entry for it will be added to it . The process is then repeated , in total Npoint times . For each cell ( p , q ) represented in C h → ( i , j ) a count n ( p , q ) is maintained and αp , q is estimated by: α p , q = n ( p , q ) N p o i n t ( 13 ) Eq 12 is of the form d M d t = T · M , where T is called the transition matrix . The displacement set determines the transition matrix . Here , we have described a Monte Carlo strategy that uses serial search to determine the set C h → ( i , j ) and consequently the constants αp , q for bins ( p , q ) in that set . With these constants determined , it is a straightforward matter to solve Eq 12 numerically . The main algorithm now consists of three steps: updating the index relationship Eq 9 , which constitutes the movement of probability mass through the grid during a time interval Δt; implementing the reversal mapping; solving Eq 12 during Δt . The order of these steps matters . Implementing the reversal bin after the master equation may lead to removing probability mass from the beginning of the strip that should have been mapped to a reversal bin . Many neuron models incorporate a threshold of some sort . For example , in the original conductance based model by [20] , a threshold of -55 mV is applied . This corresponds to a vertical boundary in the ( V , g ) plane ( see Fig 1 ) . Neurons that hit this threshold from lower potentials generate a spike and are taken out of the system . After a period τref , they are reintroduced at ( Vreset , g ( tspike + τref ) ) , where tspike is the time when the neuron hits the threshold , and g ( tspike ) is the conductance value the neuron had at the time of hitting the threshold . In this model , following [20] , it is assumed that the conductance variable continues to evolve according to Eq 8 , without being affected by the spike . We handle this as follows . For each strip it is determined which cells contain the threshold boundary , i . e . at least one vertex lies below the threshold potential and at least one lies on or above the threshold potential . The set of all such cells is called the threshold set . In a similar way a reset set is constructed , the set of cells that contain the reset potential . In the simplest case , for each cell in the threshold set the cell in the reset set is identified that is closest in w to that of a threshold cell . The threshold cell is then mapped to the corresponding reset cell and the set of all such mappings is called the reset mapping . Sometimes , the value of w is adapted after a neuron spikes . In the AdExp model , for example , w → w + b after a spike . In this case , we translate each cell in the reset set in direction ( 0 , w ) , and calculate its displacement set , just as we did for the transition matrix . The reset mapping is then not implemented between the threshold cell and the original reset cell , but to the displacement set of that reset cell . We do this for all threshold cells and thus arrive at a slightly more complex reset mapping . Due to the irregularity of the grid , it may happen that some transitions of the Master equation are into cells that are above the threshold potential . This will lead to stray probability above threshold , if not corrected . We correct for this during the generation of the transition matrix . If during event generation a point ends up above threshold after translation , we look for the closest threshold cell for this point . The event is then attributed to that threshold cell , and not the stray cell above threshold . In this way transitions from below or on threshold to cells above threshold are explicitly ruled out . The reset mapping must be carried out immediately after the solution of the master equation , before the next update of the index function . All grids are finite . For that reason alone the Monte Carlo procedure described above will result in translated points that cannot be attributed to any cell . Those events are lost and will lead to unbalanced transitions: mass will flow out of bins near the edge , but will not reappear anywhere else in the system and there is a possibility that mass evaporates from the system . This problem does not occur just at the edges , but also in the vicinity of stationary points . We will see that some dynamical systems display strong non linearities that will make it impossible to cover state space densely . The ability to deal with such gaps in state space is the most important technical challenge for this method . In Fig 2 we show how to handle these gaps . Fig 2B shows that a cell which is translated by 5 mV can fall across a small cleft not part of the grid . We cover this gap by a quadrilateral ( in green ) : a fiducial cell . An event that is not within the grid , but inside this quadrilateral needs to assigned to a mesh cell , otherwise the transition matrix will not conserve probability mass . It is straightforward to maintain a list of grid cells that have at least one vertex in the fiducial bin . We assign the event to the grid cell that is closest along the projection in the jump direction . Fig 2D shows the total number of events lost in the generation of transition matrix corresponding to a jump of 5 mV , thereby revealing gaps in state space . The orange quadrilaterals are the fiducial bins . After reassignments all events fall inside the grid and probability will be balanced . It is straightforward to calculate marginal distributions . Again , we use Monte Carlo simulation to generate points inside a given quadrilateral ( p , q ) . We then histogram these points in v and w . For each bin i in the v histogram , we can now estimate a matrix element α ( p , q ) , i by dividing the number of points in bin i by the total number of points that were generated . For a given distribution , one can now multiply the total mass in bin ( p , q ) by α ( p , q ) , i to find how much of this mass should be allocated to bin i . If one does this for every cell ( p , q ) in the grid , one will find the distribution of mass over the marginal histogram , and can calculate the marginal density from this .
We consider neurons with a single excitatory synapse as given by Eq ( 8 ) . In Fig 4 we present first the simulation of a jump response: a group of neurons is at rest at time t = 0 and all neurons are at ( V = −65 mV , g = 0 ) . From t = 0 onward the neurons will receive Poisson distributed input spike trains with a rate of 1000 Hz . A neuron that receives an input spike will undergo an instantaneous state transition and move up in conductance space . Until it receives a further input spike it will start to move through state space under its endogenous neural dynamics: the neuron will depolarize and simultaneously reduce its conductance . The process was described in Sec . Materials and Methods: State Space Models of Neuronal Populations . The density is represented as a heat plot: the maximum density is white , lower density areas are shown as cooler colours from white through yellow to red . The color scale is logarithmic , so red areas represent very low probability . Fig 4A ) shows the evolution of the density of a population that was at equilibrium at t = 0 at four points in time t = 1 , 5 , 15 and 28 ms by which time steady state has been reached . We see probability mass moving mainly upwards under the influence of incoming spike trains . We will see that the mass ‘rotates’ in the direction of the threshold; and finally a steady state is realized: a state where the density profile has become stationary . We also have simulated a group of 10000 neurons and modeled incoming Poisson spike trains for each one . We keep track of their position in ( V , g ) space and represent their state at a given time as points in state space . The cloud of points clearly tracks the white areas of the density . The shot noise structure is clearly visible in the band structure early in the simulation where neurons are present at multiples of the synaptic efficacy , reflecting that some neurons have sustained multiple hits by incoming spike trains . As neurons are moving through threshold , they themselves emit a spike and contribute to the response firing rate of the population , defined as the fraction of the population that spikes per time interval , divided by that time interval . We can therefore calculate the response firing rate from the amount of mass moving through threshold per unit time . We show the jump response of the population as a plot of populating firing rate as a function of time in Fig 4B . The firing rate calculated from the density matches that calculated from the Monte Carlo simulation very well . Interestingly , there is almost no overshoot in the firing rate , as also noted by Richardson ( 2004 ) , who studied this system using Fokker-Planck equations . Although we study shot noise , in the absence of a fundamental scale in the g direction , the central limit theorem ensures that the marginal distribution in g is Gaussian within a few milliseconds . It is clear that the population disperse in the g direction and drifts towards the threshold relatively slowly . The absence of a barrier allows the dispersal of the population before it hits threshold , greatly reducing any overshoot in the firing rate , which is quite unlike one dimensional neural models , as we shall see in Sec . Results: One Dimension . Let us contrast this with a simulation where we introduce a maximum conductance gmax = 0 . 8 , which for simplicity we assume to be voltage independent . This then introduces a reflecting boundary at g = gmax , and therefore introduces a scale by which an efficacy can judged to be small or large . As expected , probability mass is squashed against this boundary ( Fig 5A ) and has nowhere to go but laterally , in the direction of the threshold . Interestingly , the mass has not dispersed and clear groupings of mass huddled against the boundary can be observed . The traversal of the threshold by these groupings produces clear oscillations in the firing rate: a “ringing” effect . The firing rate jump response reflects the effect of the presence of a maximum conductance in state space . We run two simulations: one with and one without maximum conductance , but otherwise identical , and repeat this experiment for two different synaptic efficacies: J = 1 and 3 mV . Both simulations use an input rate of 3 kHz . In the case of no maximum conductance , probability mass can disperse in the g direction and mostly does so before arriving at the threshold . In Fig 5 one sees that the introduction of a maximum conductance leads to a reduced response firing rate for high inputs . This can be interpreted as the population unable to respond to an increase of input once the majority of its ion channels are already open . Fig 5 shows that the firing rates of Monte Carlo simulations and our method agree over the entire range of input . Even when the effects on the response firing rate are moderate , the transient dynamics can be radically different . For an efficacy J = 1 mV and and input rate νin = 3 kHz , the firing rates for maximum conductance , compared to no maximum come out as 175 Hz vs 195 Hz . In Fig 5C we show the response firing rate as a function of time . The result for the unrestrained conductance is given by the red line , which despite the high output firing rate still almost produces no overshoot . When we restrict the maximum conductance we see a somewhat reduced firing rate but a pronounced transient response ( “ringing” ) which persists much longer than for an unrestrained conductance . It is striking to see that the reintroduction of a barrier in state space results in pronounced transients . In both cases , the calculated firing rates agree well with Monte Carlo simulation . We attribute this ringing to a geometrical effect: the introduction of a barrier in the direction of where the stochastic process is pushing neurons . Although these model neurons are characterized by a single dimension—the membrane potential—they can be viewed as a two dimensional model that is realized in a single strip , and where transitions take place between one bin in potential space to another . This is completely equivalent—in implementation and concept—to the geometric binning method introduced independently by de Kamps [12] and Iyer et al . [11] , with one exception: the generation of transition matrices by Monte Carlo . In one dimension it is not necessary to use Monte Carlo generation: the transition matrix elements can be calculated to an arbitrary precision because in one dimension the geometrical problem outlined in Sec . Materials and Methods: Handling Synaptic Input is much simpler and can be solved by linear search . It is clear that unlike the 2D case , it is straightforward to find the exact areas covered by translated bins , and hence no Monte Carlo generation process is required . Nevertheless , it is interesting to use this method . The transition matrix generation for the 2D case is relatively expensive , and as precision scales with the square root of the number of events it is interesting to see how few we can use in practice without distorting our results . The answer is: surprisingly few . As benchmark we set up a population of LIF neurons with membrane constant τ = 50 ms , following [5] , and assume that each neuron receives Poisson distributed spike trains with a rate ν = 800 Hz . We assume delta synapses , i . e . an instantaneous jump in the postsynaptic potential by a magnitude h = 0 . 03 , with the membrane potential V∈ [−1 , 1 ) , i . e . we use a rescaled threshold potential V = 1 . The grid is generated with a time step Δt = 0 . 1 ms , and is shown in Fig 6B . The simulation results are shown in Fig 7 and replicates earlier work [5 , 12] . The use of a finite number of points in the Monte Carlo process used for the generation of transition matrices generates random fluctuations with respect to the true values . The effect of these fluctuations is clearly visible in the shape of the density profile , and only for Npoint = 10000 the profile is as smooth as in earlier results where we calculated the transition matrix analytically . How bad is this ? To put these fluctuations into perspective , we used a direct simulation of 10000 spiking neurons and histogrammed their membrane potential at a simulation time well after t = 0 . 3 s , so that they can be assumed to sample the steady state distribution . In the figure , they have been indicated by red markers . Comparing the results we see that the fluctuations for Npoint = 10 are comparable to those of a Monte Carlo simulation using a sizeable population of 10000 neurons . Moreover , in the population firing rates the finite size effects are almost invisible . This is somewhat surprising , but a consideration of the underlying process that generates the firing rate explains this . Neurons are introduced at equilibrium and will undergo several jumps before they reach threshold . The finite size effects of the Monte Carlo process induce variations in those jumps in different regions of state space , but these fluctuations are unbiased and will average out over a number of jumps . So neurons will experience variability in the time they reach threshold , but this variability does not come in the main from fluctuations in the transition matrix elements . It should be emphasized that the transition matrices are a quenched source of randomness , because transition matrices are fixed before the simulation starts . So although ultimately caused by finite size effects , their contribution is different compared to the unquenched finite size effects that can be seen in the population of 10000 neurons . It is instructive to look at some examples because it highlights strengths and weaknesses of the method in terms of familiar results . In Fig 6A , the characteristics of both neural models are given . In Fig 6B the state space of LIF ( left ) and QIF neurons ( right ) are shown , at lower resolution than used in simulation to elucidate the dynamics . Rather than with numbers which would be unreadable at this scale , we indicate the direction in which cell numbers increase , and therefore the direction in which neural mass will move , by arrows . One can see that the LIF neuron is comprised of two strips , and the QIF neuron of three , where the arrows indicate in which direction the cell numbers are increasing . In the LIF grid , there is one stationary bin , in the QIF there are two . They are represented as separate stationary cells , covering the space between the strips , indicated by the blue downward pointing arrows . In Fig 6C we consider the steady state of LIF ( left ) and QIF neurons ( right ) after being subjected to a jump response of Poisson distributed spike trains starting at t = 0 ( LIF: νin = 800 spikes/s J = 0 . 03 ( normalized w . r . t . threshold; QIF: J = 0 . 05 ) ) . The shape of the characteristics and therefore of the grid clearly reflect their influence on the steady state density distribution . The output firing rate ( Fig 6D ) shows the clear “ringing” in the transient firing rate that is mostly absent in conductance based neurons . Again , this can be interpreted geometrically: the stochastic process pushes neurons in the direction of a threshold , but they reach it without having had the opportunity to disperse . Decorrelation only happens after most neurons have gone through threshold at least once . It is also interesting to see that for comparable firing rates the ringing is much stronger for QIF than for LIF neurons . We also interpret this as a geometrical effect: the effective threshold for QIF neurons is V = 3 ( normalized units ) , not 10 , as neurons with a membrane potential above 3 will spike . It is clear from Fig 6D that compared to LIF neurons , QIF neuron bulk up close to the threshold and are constrained more than their LIF counterparts , thereby making it harder to decorrelate before passing threshold . For reference , in Fig 8 we show that the method accurately reproduces results from the diffusion limit , as well as generalizes correctly beyond it . If one uses a single Poisson spike train to emulate a Gaussian white noise input , employing the relationship: μ = ν in J τ σ 2 = ν in J 2 τ , ( 14 ) one can use our method to predict the steady state firing rates as a function of J , the synaptic efficacy and νin the rate of the Poisson process for given membrane constant τ . Organizing the results in terms of μ and σ , as given by Eq 14 , one expects a close correspondence for low σ , since Eq 14 leads to small values of J compared to threshold . One expects deviations at high σ , where J does not come out small . Fig 8 shows that this is indeed the case when firing rates are compared to analytic results obtained in the diffusion approximation . Our method produces the correct deviations from the diffusion approximation results , and agrees with Monte Carlo simulation . Elsewhere [12] , we have shown that diffusion results can be accurately modeled using two Poisson rates for high σ . In Fig 8B we replicate the gain spectrum for QIF neurons and show that the high frequency dependence falls off as 1 ω 2 as predicted by Fourcaud-Trocmé et al . [25] . These results reaffirm that our method accurately predicts results within and beyond the diffusion limit , and that a substantial body of existing literature can be seen to be a special case of our method . Fig 8C shows a population of QIF neurons that fire in synchrony at t = 0 , undergoing a slow decorrelation by low rate Poisson input spike trains . The neurons have all been prepared in the same state , and therefore are at the same position in state space . We use F ( V ) = V2 + 1 , so these neurons are bursting , as the current parameter is larger than 0 , and there are no fixed points . Neurons that receive an input spike leave the peak and travel on their own through state space . This results in a very complex density profile , where the initial density peak is still visible after 1s . Such a peak would have diffused away rapidly in a diffusion limit approximation . Monte Carlo events in red markers show that the density profile is not a numerical artefact , but reflects the complexity of the density profile . We consider the AdExp model as presented by Brette and Gerstner [26] , which describes individual neurons by the following equations: C m d V d t = − g l ( V − E l ) + g l Δ T e ( V − V T ) Δ T − w + I τ w d w d t = a ( V − E l ) − w ( 15 ) Upon spiking , the neuron is reset to potential Vreset and increases its adaptivity value: w → w + b . Here Cm is the membrane capacity and gl the passive conductance . VT is the value at which a neuron starts to spike; the spike dynamics is controlled by ΔT . The numerical values of the parameters are summarized in Table 2 and are taken from [26] . An overview of the state space is given in Fig 9A . At w = 0 the dynamics is as expected , a drive towards the equilibrium potential that suddenly reverses into a spike onset at higher values of V , essentially producing an exponential-integrate-and-fire neuron . At high w two effects conspire to make the neuron less excitable: the equilibrium potential is lower and the drive towards this equilibrium is stronger for a given value of V . At low w values , the opposite happens: the equilibrium value is higher , closer to threshold , and below equilibrium there is a stronger depolarizing trend making the neuron more excitable . Interestingly , at hyperpolarization the system does not only respond by driving the membrane potential back towards equilibrium potential , but also downwards . There are two critical points , the equilibrium point ( El , 0 ) and a saddle point in the top right . They are at the crossing of two nullclines: the w-nullcline is a straight line , whereas the V-nullcline follows a strongly curved trajectory , which is close to the stable manifold of the saddle point in a substantial part of state space . Below ( to the right ) the stable manifold neurons spike , regardless of where they are initially , while above ( to the left ) of the stable manifold neurons converge to the equilibrium , but how , and how long this takes is strongly dependent on the initial conditions . This model is the first to require a judicial treatment of the grid boundaries . Let us examine the the equilibrium point first . The exponential build-up of cells observed in one dimensional models occurs here as well , but here it is not a good idea to introduce a fiducial cut and cover the remaining part of state space with a cell . The inset of Fig 9B shows that equilibrium is reached much faster in the V direction , than in the w direction . This is a direct consequence of the adaptation time constant τw being an order of magnitude larger than the membrane time constant τ ≡ Cm/gl . For high w , mass will move downwards along the diagonal , until low values of w are reached , as is demonstrated by the left inset of Fig 9 . A long , but very narrow region separates different parts of the grid . What to do ? First , we observe that the offending region is essentially forbidden for neurons: for most neurons starting from a random position in state space it would take a long time ( of the order of 100 ms ) to approach this no man’s land . At the input firing rates we will be considering , neurons will experience an input spike well before running off the strip , so essentially only noise can place neurons there . If we forbid this , by allocating events that are translated into the cleft between the two grid parts to the cells in the grid that are closest to it along the projection of the jump , we guarantee that no probability mass will leak out of the grid . Mass that reaches the end of the strips will be placed in a reversal bin , like the one dimensional case . Mass on the left of the side of the cleft will move in the same direction as that on the right side of the cleft . By using Euclidean distance projected along the jump direction , we minimize the bias due to this procedure , although we may artificially introduce a small extra source of variability . On the right hand side , the stable manifold almost coincides with the V nullcline , resulting in a very narrow region of dynamics in the vertical direction . Immediately outside neurons rapidly move away laterally . This part of the grid is created by reversing the time direction , integrating towards the stable manifold . The grid strongly deforms here: cell area decreases rapidly and even small numerical inaccuracies will lead to cells that are degenerate . We use cell area as a stopping criterion . The last cells before breaking off are extremely elongated . The spike region is also created by reversing the time direction . Again , we conclude that the cleft is a forbidden area: a small fluctuation in the state variable will cause a neuron to move away rapidly . Our main concern , again , is neurons that are placed into this cleft by the noise process . Again , we move neurons to the closest cell next to the cleft in the jump direction . This is reasonable , since natural fluctuations would put them there soon anyway . Effectively we have broadened the separatrix a little bit , but we still capture the upwards ( for high w—past the saddle point: downwards ) movement close to the stable manifold . In Fig 9C–9E the evolution of a population in ( V , w ) space is shown at three different points in time: t = 0 . 05 , 0 . 1 and 0 . 4 s . Fig 9C shows the input spikes pushing the state towards threshold , and a small number of neurons have spiked . They re-emerge at the reset potential , but with much higher w , due to spike adaptation . This is determined by the b parameter of the AdExp model . Close to the reset potential the banded shot noise structure , due to the use of a delta-peaked synaptic efficacy , is visible . The steady state is reached after approximately 400 ms . The population stabilizes at high w values , and the bulk of the population is clearly well below threshold , due to stronger leak behavior at these values of w . In sub figure E there is a minute deformation of the density , due to the limits of the grid , and density heaps up here , but the fraction of probability mass affected is negligible . Monte Carlo events , indicated by the dots , are not restricted to the grid and some fall outside the grid . The firing rate response corresponding to the population experiencing an excitatory input ( Fig 9C–9E ) is given in Fig 10A . Again , agreement with Monte Carlo simulation is excellent , we are able to study the relative contributions of current- and spike-based adaptation to the firing rate . We can easily simulate neurons with current- but not spike-based adaptation by not incorporating the jump in w after reset; while ignoring all forms of adaptation can be done by simply using a 1D grid and ignoring values of w ≠ 0 . The vast difference between adaptive neurons and non-adaptive neurons is also reflected in the gain spectrum . Fig 10B shows the gain spectrum of a ( non-adaptive ) exponential-integrate-and-fire neuron and a neuron that has a constant rate of adaptation due to the background rate upon which the small sinusoidal modulation has been imposed . The difference between the adaptive and non-adaptive neuron is considerable . Both neurons show a 1 ω dependence in the high frequency limit , as is expected for exponential neurons [25] . ( Fig 10A shows that the shape of the spike , which is reflected in the large cells on the right of the grid is independent of w . ) It is clear that a meaningful time-independent gain function cannot be chosen , so that it is not possible to develop linear response theory . It is interesting to observe the marginal distributions—in Fig 11 we show the marginal distributions , together with the joint distribution . The distribution in V looks remarkably like that of an LIF neuron , except near the threshold , where the spike region , which is not present for LIF , flattens the density . The w distribution suggests a much stronger overlap than the joint distribution , which shows a clear separation . It is clear that , had the three density blobs been oriented more diagonally , the marginal w distribution would have shown a single cluster . Vasilaki and Giugliano have studied the formation of network motifs [24] , using both microscopic spiking neural simulations and mean-field approximation . In their mean-field simulations they considered both spike-timing dependent long-term plasticity , and frequency-dependent short-term dynamics , where they use a version of the Tsodyks-Markram synapse [27] . The short-term dynamics is of interest because it introduces something we have not considered before: the magnitude of the jump being dependent on the position of where the jump originates . Following [24] , if Gij defines the amplitude of the postsynaptic contribution from presynaptic neuron j to postsynaptic neuron i , then this is considered to be proportional to the amount of resources used for neurontransmission uijrij and to their maximal availability Aij , so G i j = A i j u i j r i j , ( 16 ) where r relates to the recovery and u to the facilitation of synapses , and the time constants τrec and τfacil are different for facilitating and depressing synapses . They describe frequency-dependent short-term synaptic dynamics by: d r i j d t = ( 1 − r i j ) / τ rec − u i j r i j ∑ k j ∞ δ ( t − t k j ) ( 17 ) d u i j d t = − u i j / τ facil + U ( 1 − u i j ) ∑ k j ∞ δ ( t − t k j ) ( 18 ) From now on , we will drop the indices ij and just refer to a single connection . In the simulation below we will use τrec = 0 . 1 s and τfacil = 0 . 9 s and study a population of facilitating synapses ( Vasilaki and Giugliano used τrec = 0 . 9s , τfacil = 0 . 1 s for depressing synapses . ) U is a fixed constant , for facilitating ( depressing ) synapses U = 0 . 1 ( 0 . 8 ) . Eq 17 expresses that an individual synapse is subject to deterministic dynamics , and that upon the arrival of a spike at time tk both u and r undergo a finite jump , whose magnitude is dependent on the current value of u and r . Eq 2 describes this situation , when the following transition probabilities are introduced: W ( r ′ , u ′ | r , u ) = ν δ ( r ′ − r + u r ) δ ( u ′ − u − U ( 1 − u ) ) − ν δ ( r ′ − r ) δ ( u ′ − u ) ( 19 ) We have to modify the process of generating our transition matrices: now for each quadrilateral cell ( p , q ) , we determine the centroid ( u ( p , q ) , r ( p , q ) ) and we determine the covering set by defining h → = ( − u ( p , q ) r ( p , q ) U ( 1 − u ( p , q ) ) ) ( 20 ) and determining the cover set as before . The jump now becomes cell dependent . It is easy to cover almost the entire state space . In Fig 12A we show the grid . In Fig 12B , we show the sample path of three synapses , assuming that the presynaptic firing rate ν = 5 Hz . In C-F we show the evolution of a population of synapses . The influence of the step size which increases in the r ( horizontal ) direction with u and r , but decreases in the u ( vertical ) direction with u . There is good agreement with Monte Carlo simulation throughout . With the joint distribution available , it is possible to use Eq 16 and calculate the distribution of Gij or its expectation value . We consider the well-known Fitzhugh-Nagumo neuron model [28] , which is given by: d V d t = V − V 3 3 − W + I d W d t = 0 . 08 ( V + 0 . 7 − 0 . 8 W ) ( 21 ) It is an attractive neuron model as it captures many properties of the biologically realistic Hodgkin-Huxley neuron , while being much more tractable—being reduced to two dimensions aids greatly in analysis and visualization . The two variables are a nondimensionalized voltage-like variable V and a recovery variable W . Also we note a variable I representing a constant external current . When I = 0 , there is a stable equilibrium point at ≈ ( −1 . 199 , −0 . 624 ) corresponding to a resting state . As I increases , the system undergoes a Hopf bifurcation to a stable limit cycle around an unstable equilibrium . ( Increasing I further leads to a stable fixed point at positive V and W termed “excitation block” . ) In this paper , we will consider an intermediate value I = 0 . 5 in order to demonstrate how our method can be used on systems with limit cycles . We simulate white noise by providing the system with both inhibitory and excitatory noisy input with a high rate and low synaptic efficacy , and successfully capture the diffusion of probability in a neighbourhood around the limit cycle ( Fig 13A–13D for t = 5 , 10 , 50 and 1000 s , J = ±0 . 02 , ν = 20 spikes/s ) . It is interesting to study a purely excitatory input with large synapses ( J = 0 . 1 , ν = 2 Hz ) . This leads to a deformed limit cycle , shifted towards higher V . This is expected as the net input now is I = 0 . 7 . The band is also broader , as one would expected as higher values of synaptic efficacy imply larger variability . Another case we consider is noisy inhibitory input ( Fig 13F ) . As we would expect , the system is effectively driven back below the bifurcation to a stable equilibrium , although we still see some variance-driven probability follow a limit cycle that differs considerable from the original limit cycle . We can understand this by converting the noisy input into zero-mean noise and a steady inhibitory current , and looking at the streamlines of the system with these parameters instead . As seen in Fig 14D , while all the the trajectories converge to the fixed point , those starting on the right side of phase space first increase w until they reach the right branch of the cubic nullcline , then follow a path close to the limit cycle to return to the fixed point . It is interesting to see that the method captures limit cycles that do not coincide with the limit cycle of the original grid . Rabinovitch and Rogachevskii [29] describe the two “vertical” sections of the path to be transient attractors ( T-attractors ) separated by a diagonal transient repeller ( T-repeller ) ( alternatively , a separatrix [30] ) close to the central branch of the cubic nullcline . Trajectories close to each other but starting on different sides of the T-repeller separate rapidly before eventually reaching the same steady state , which creates considerable problems in creating the grid ( see Fig 14E ) . The authors perform a detailed analysis of the system by extending the notion of isochrones from limit cycles to excitable systems . We note that their isochrones are similar in character to the lines in our grid perpendicular to the streamlines of the system . Next , we outline some of the numerical subtleties involved in generating the computational grid for the Fitzhugh-Nagumo model . Following the procedure from Sec . Materials and Methods , one can attempt to generate a grid by starting with a set of initial conditions , and solving the differential equations of the system forwards in time to obtain a set of trajectories ( or integral curves ) . Each pair of trajectories then has a strip between them and the individual cells are obtained by dividing the strip into equal-time bins . However , in a system with a limit cycle , if we start with initial conditions outside the limit cycle , we see that the trajectories generated from them converge onto the limit cycle . Moreover , it is impossible to obtain trajectories inside the limit cycle from outside the limit cycle , and vice versa . This means that we have to handle the limit cycle , outside , and inside , as separate sections of the plane . Since the limit cycle is a one-dimensional object with zero width , we have to artificially define a small width around it . We then choose sets of initial conditions outside and inside the limit cycle and integrate the trajectories until they reach a certain small Euclidean distance from the limit cycle , and then define our limit cycle strip as the space left . In this left over space we define quadrilaterals so as to fill up this ring . This becomes a strip in its own right , representing the limit cycle . Earlier we described the reversal mapping: mass reaching the end of a strip must be removed and deposited in a cell representing a stable point . Here , we use a similar approach: mass that arrives at the end of a strip must be removed and deposited on the limit cycle . We find the cell on the limit cycle that is closest in Euclidean distance to the limit cycle . Since the machinery to do this is already in place in the form of a reversal mapping , we will also refer here to this process as a reversal mapping . The modeler presents this reversal mapping in the same file format as used previously . Initially we had attempted to define our limit cycle cells as having a fixed width , and then obtain strips by integrating backwards in time from the corners of these cells . Indeed , the coarse schematic grid in Fig 14 has trajectories generated in this way for the interior of the limit cycle . However , for the purposes of actual computation , this method leads to degenerate cells . This is due to the fact that close to the limit cycle , trajectories move almost parallel to it , in particular along the “horizontal” segments of the limit cycle , where the fast v dynamics dominate . This leads to long , thin cells being created , which become degenerate when approaching the limit cycle—adjacent trajectories overlap to the degree of accuracy of the numerical integrator , leading to self-intersecting cells or cells with zero area . From the outside of the limit cycle , most of the state space can be covered by simply choosing points on the edge of the region of interest and integrating forwards in time until one reaches the limit cycle . However , care must be taken when trajectories converge before arriving at the limit cycle , as shown in Fig 14B . This happens particularly along the cubic nullcline . We handle this by checking for degenerate cells or cells with area close to zero . These cells are then deleted from the grid , and instead a reversal mapping is created from the previous cell onto the closest ( in Euclidean terms ) cell . The interior of the limit cycle proves to be even more challenging . Not only is there an unstable fixed point , also there exist canard trajectories , which have been the subject of considerable mathematical interest [29–32] . Loosely speaking , near the central portion of the cubic nullcline , there are slow but unstable trajectories . This leads to two types of numerical issues—first , the slow dynamics cause a build-up of exponentially many very small cells . We work around this by defining a minimum value of ‖ v → ˙ ‖—regions below this value are considered to be approximately stationary , since they will have much slower dynamics than any noisy input we consider . The region we find is shown in Fig 14C . We use cubic splines to approximate the boundary of this region , and then use points on this boundary as initial points for trajectories on the inside of the limit cycle to generate strips . Due to the instabilities in this region of the system , trajectories can be highly curved , and trajectories with initial conditions close to each other can diverge quickly , leading to cells which may intersect with each other , as shown in Fig 14E . As these areas with highly curved trajectories are still locally smooth , it may be possible to increase the resolution of the grid until non-degenerate cells are obtained , as we do here . However , it may not always be possible to do so due to computational constraints—in that case it may be more practical to delete bad cells after the creation of the grid and cover any gaps with fiducial bins . To sum up , regions where trajectories merge—such as the limit cycle and nullclines in this case—involve moving from the two dimensional plane onto one dimensional trajectories , and pose conceptual as well as computational difficulties . Regions with highly curved trajectories may be possible to handle with very fine resolutions , but may pose difficulties at coarser resolutions . In both cases it is possible to handle such regions using an automated procedure: cells are checked for being complex quadrilaterals or having too small an area . Those satisfying this condition are deleted , and renewal mappings from the cells before them to the nearest cells are generated . Any gaps in the grid due to this can be handled using the prescribed method for creating fiducial bins . In conclusion , we have successfully extended our procedure to dynamical systems with limit cycles and complex dynamics such as canards . While we have to make some compromises in the regions which pose significant analytic difficulty , these regions are those in which neurons would not spend any significant amount of time . Hence , our method would still be suitable for studying neural circuits of such populations . We solve Eq 12 by a forward Euler scheme . Since we interleave moving probability mass through the grid with a numerical solution of Eq 12 , we solve Eq 12 over a period Δt , which can be as short as 10−4 s for some neural models . This renders sophisticated adaptive size solvers relatively inefficient . The matrices in Eq 12 tend to be sparse band matrices , and one advantage of the forward Euler scheme is that it is embarrassingly parallel . For a single population in this we partition Δt into neuler time steps . For most simulations in this paper neuler = 10 is adequate , we will discuss an exception below . In a forward Euler scheme Eq 12 is discretized and a single step is given by: M ( map k ( i , j ) , l + 1 ) = M ( map k ( i , j ) , l ) + Δ M , with Δ M = ν Δ t n e u l e r { ∑ ( p , q ) ∈ C h ( i , j ) α p , q M ( map k ( p , q ) , l ) − M ( map k ( i , j ) , l ) } ( 22 ) The current simulation time is tsim kΔt , where Δt is the mesh time step . The current map , which indicates where probability mass has moved under the influence of endogenous neural dynamics is therefore labelled by mapk ( i , j ) , which maps cell ( i , j ) to a unique mass array index as per Eq 9 . Note that the mapping should not be applied to the set Ch ( i , j ) . Simulation starts at k = 0 , l = 0 . Each Euler step l is increased , until l = neuler , upon which l is reset to 0 and k is increased by 1 , until the desired end time is reached . The sets Ch ( i , j ) and coefficients αp , q remain constant throughout simulation . Despite appearances , the right-hand side of Eq 22 is of the form of a matrix vector multiplication , where the matrix is very sparse ( Fig 15A ) . The matrix elements are numerical constants , and there is no dependence between rows , meaning that each row can be evaluated independently of the others and therefore the problem is extremely parallel: each row can be calculated in a separate thread when available . Larger networks can be simulated by vectorizing the network: vectors representing the mass of individual populations are laid out in a single array representing the mass of multiple populations . Connections from population i to population j then are represented by block matrix elements , each consisting of one or more transition matrices ( one for every efficacy associated with the connection ) generated by the process described above . We have created both a C++ and a CUDA implementation and evaluated them on single populations , as well as networks of populations . When generating networks we consider networks of conductance based populations , each population connected to constant input ( ’cortical background’ ) , with the network being sparsely connected ( connectivity 5% ) , efficacy chosen fixed ( h = 0 . 05 ) , with the number of connections drawn from a uniform distribution ( n = 1 , … , 100 ) . For small networks , the running times are not particularly onerous , with or without parallelization . For larger networks , in the C++ implementation we evaluated a single block matrix per thread using OpenMP . This means that in our implementation individual matrix vector calculations are not parallel , but that several matrix vector calculations are performed simultaneously . Since OpenMP offers a relatively small number of threads , this still makes efficient use of resources . The parallelization model for CUDA is different: we write a so-called kernel to evaluate Eq 22 and launch a kernel for each block matrix . CUDA’s loop unrolling automatically performs parallelization within the kernel , and by launching kernels in different streams , inter kernel concurrency can be achieved . It is then the question whether the large number of threads compensate for the inherently slower GPU hardware . In Fig 15D we show how the GPU interacts with the C++ driver . During initialization the mass array is set up on the GPU , as well as the mapping , and the matrix elements . During a simulation run , the mass array mapping is updated , and firing rates are exchanged , but other than for visualization purposes , the mass array is not transferred , meaning lightweight communication between GPU and CPU . We find that the number of cells in a mesh determines the performance . In Fig 15 we examine the conductance based neuron example again . If we use the original mesh , without considering performance , we find that the method is slower than direct simulation as performed by NEST by a factor of three ( 6s per population for NEST—20s per population for the CUDA implementation ) . If we reduce the granularity of the original mesh , we find that we can bring the size of the mesh down from 120k cells to 25k cells , with the density and firing rate predictions unaffected ( Fig 15A vs Fig 15B for density , B being the coarser mesh which is only visible in the bigger cells on top , C for firing rate ) . For the reduced mesh , the performance of the C++ implementation is equal to that of NEST , where the CUDA implementation is a little bit slower ( measured on a Tesla P100 ) . Both direct ( NEST ) simulation and C++ implementation use parallelization with 16 threads for this comparison . The real time factors ( real time second divided by wall time of one simulated second ) are shown in Fig 15E . A striking difference is the memory use ( Fig 15F ) , which for the CUDA implementation is orders of magnitudes lower ( 300 MB for the largest network of 1000 populations , whereas a 100 population network with NEST already uses 10 GB ) . We conclude that the CUDA implementation supports the simulation of large networks on a single PC equipped with a GPGPU , whereas direct simulation requires a substantial cluster . In the method we considered so far , a relatively large number of cells emerge around stationary points due to the exponential shrinkage of state space around them . In principle it is possible to group these cells together into larger ones , and to group them into strips that would run at a lower speed compared to the basic time step of the mesh . This would reduce the number of cells in the mesh considerably , while the basic granularity of the mesh will not be affected much . Such resulting merged cells are no longer quadrilateral and the method will have to be extended to be able to handle non-convex cells , which will be one of the first priorities in further work . We have demonstrated a very general method to study noise in 2D dynamical systems and applied them to various neural models and Tsodyks-Markram synapses . The state space of the deterministic model must be represented by a grid . The requirement that a grid be made is both a strength and a weakness: the state space relevant to the simulation must be chosen judiciously before the simulation starts . But since it must be constructed beforehand , integration can be done very accurately , using time steps that are much smaller than typically used in Monte Carlo simulators . If general purpose simulators are used with a default time step , and without adaptive methods that monitor errors , they may not alert the user to problematic regions of state space . Our method requires a careful layout of state space before simulation starts . We found that the requirement of a grid forces visualization and thereby already creates an understanding of the dynamics that can be expected . When the state space cannot contain the simulation , this is clearly visible , either through loss of mass , or by the accumulation of mass at the edge of the grid . This proved useful in one instance , where a well known neural simulator produced a crash ( due to an instability of the particular neural model implementation , not the simulator as such ) . Our method is very robust and stable , once a suitable grid is available . In general , we find that grids can be taken quite coarse in state space , but that a relatively small time step must be used for completely accurate results , such as comparison to analytic results like gain curves . When numerical errors are acceptable , and only qualitative agreement is required , much coarser grids can be used that require far less simulation time . Our method is not as efficient as effective 1D methods [15–17] , but makes very few assumptions . It handles time-dependent input without any restrictions . This is useful , for example , when comparing against basis functions expansions [13 , 14 , 33] . These basis functions are typically determined for constant input , and time-dependent input must be treated as an adiabatic approximation . Our method does not require this . In short , our method may serve as benchmark for faster methods . Population density techniques are part of an emerging ecology of simulation techniques , and it is important to consider their strength and weaknesses compared to related approaches . Direct spiking neuron simulations are straightforward in small to medium-sized networks , but hard to get right in large-scale simulations , where they are resource intensive ( they can have large memory requirements , as well as being CPU intensive ) . The “missing spike” problem , and the need to keep track of spike information and its exchange between the various processors involved are just examples demonstrating that direct simulations are not straightforward . They have developed into a discipline of their own [34] . Population density techniques are conceptually simple , but unable to model pairwise correlations within a population , and the inclusion of finite-size effects is not straightforward ( see [14] for an attempt ) . In general , population density techniques are not able to describe quenched network states , although fully connected networks are amenable to such analyses [35 , 36] . Recently , a number of studies have explored path integral approaches to calculate pairwise correlations and to suggest a functional role for such correlations ( e . g . [36–38] ) . Often , these techniques use the diffusion approximation , or are restricted to remain close to stationary states . Two advantages of the technique described in this paper is that the latter restrictions do not apply . Theoretically , population density equations have been put on a rigorous mathematical footing [39] , justifying its use for weakly connected networks where the quenched state of the network is not important . These papers also add to a substantial body of observation that even for small networks population density techniques predict the firing rate correctly ( e . g . [6 , 40] and many others ) . So , when modeling firing rates is the main objective , and the network is such that the populations may be far from equilibrium , population density techniques are a good candidate . They are also valuable in repeat experiments on single cells , as they show what noise will do to otherwise identical neurons . Modeling a complex system of real neurons probably requires a hybrid approach . Mazzucato et al . [41] give an example of such an approach . They analyse the dimensionality of neural data recorded by multielectrode array . The dimensionality is estimated from pairwise correlations , something which would be impossible in a pure population density approach , as the pairwise correlations would vanish in an infinitely large population . However , they also produce a spiking network model , to validate their explanation and here they use population density techniques to establish the dynamical regime for their spiking network . The study of 2D systems subject to noise is an important topic in its own right , given that limit cycles require at least two dimensions . The current trend in neuroscience towards 2D geometrical models reinforces this point . An important prerequisite for the method to work is that the dynamical system can be represented faithfully . We found that some systems have challenging regions of state space: stationary points , whether stable or not , and limit cycles need careful handling and a full cover of state space is not possible . However , we find that we can infer motion of probability mass inside such regions from the immediate surroundings , the limit cycle of the Fitzhugh-Nagumo system as a case in point: it emerges as a region rather than as a curve from terminating the grid as it approaches the limit cycle . There are interesting parallels between our method and a recently proposed method for determining missing spikes in hybrid time-driven , event-driven spiking neuron simulations [42] . Here , the authors consider the problem of missing spikes: the possibility that a neuron is below threshold at the end of a simulation step , but has crossed the threshold during the step . They solve this problem by determining whether a neuron is inside a volume in state space between the threshold and the backpropagated threshold . They find this easier than determining the actual point of crossing , and their method is reminiscent of ours when we calculate the transition matrix . They too consider a mapping like Eq 6 which they are able to calculate explicitly for current based neurons . They conclude that apart from the threshold and the backpropagated threshold , the boundary is given by the vanishing tangent space of the map , precisely the criterion we used numerically ( area of cell—in the absence of analytic solutions ) to define boundaries of state space . It is interesting to speculate about extending the method towards even higher dimensions . At first sight , this seems unfeasible: a three dimensional grid might already require millions of bins . It is not efficient to simulate systems with a size of the order 104 particles by a larger number of bins . It would also be considerably harder to visualize the results . Nonetheless , probability tends to cluster in specific areas of state space and we find large parts of state space effectively unoccupied . A dynamical representation of the occupied part of state space would lead to a more scalable method . Our simulation results have shown that we can simulate large networks consisting of hundreds or thousands of populations . To make really large networks run more efficiently , we need smaller meshes and the best way to achieve that we now believe is to lump the large number of small cells that emerge near stationary points into larger ones , as described above . This will be our main focus for the near future . | A group of slow , noisy and unreliable cells collectively implement our mental faculties , and how they do this is still one of the big scientific questions of our time . Mechanistic explanations of our cognitive skills , be it locomotion , object handling , language comprehension or thinking in general—whatever that may be—is still far off . A few years ago the following question was posed: Imagine that aliens would provide us with a brain-sized clump of matter , with complete freedom to sculpt realistic neuronal networks with arbitrary precision . Would we be able to build a brain ? The answer appears to be no , because this technology is actually materializing , not in the form of an alien kick-start , but through steady progress in computing power , simulation methods and the emergence of databases on connectivity , neural cell types , complete with gene expression , etc . A number of groups have created brain-scale simulations , others like the Blue Brain project may not have simulated a full brain , but they included almost every single detail known about the neurons they modelled . And yet , we do not know how we reach for a glass of milk . Mechanistic , large-scale models require simulations that bridge multiple scales . Here we present a method that allows the study of two dimensional dynamical systems subject to noise , with very little restrictions on the dynamical system or the nature of the noise process . Given that high dimensional realistic models of neurons have been reduced successfully to two dimensional dynamical systems , while retaining all essential dynamical features , we expect that this method will contribute to our understanding of the dynamics of larger brain networks without requiring the level of detail that make brute force large-scale simulations so unwieldy . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results/Discussion"
]
| [
"action",
"potentials",
"medicine",
"and",
"health",
"sciences",
"statistics",
"nervous",
"system",
"membrane",
"potential",
"electrophysiology",
"neuroscience",
"simulation",
"and",
"modeling",
"systems",
"science",
"mathematics",
"population",
"biology",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"mathematical",
"and",
"statistical",
"techniques",
"monte",
"carlo",
"method",
"dynamical",
"systems",
"population",
"metrics",
"cellular",
"neuroscience",
"cell",
"biology",
"neurophysiology",
"anatomy",
"synapses",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"population",
"density",
"statistical",
"methods"
]
| 2019 | Computational geometry for modeling neural populations: From visualization to simulation |
A fundamental assumption , common to the vast majority of high-throughput transcriptome analyses , is that the expression of most genes is unchanged among samples and that total cellular RNA remains constant . As the number of analyzed experimental systems increases however , different independent studies demonstrate that this assumption is often violated . We present a calibration method using RNA spike-ins that allows for the measurement of absolute cellular abundance of RNA molecules . We apply the method to pooled RNA from cell populations of known sizes . For each transcript , we compute a nominal abundance that can be converted to absolute by dividing by a scale factor determined in separate experiments: the yield coefficient of the transcript relative to that of a reference spike-in measured with the same protocol . The method is derived by maximum likelihood theory in the context of a complete statistical model for sequencing counts contributed by cellular RNA and spike-ins . The counts are based on a sample from a fixed number of cells to which a fixed population of spike-in molecules has been added . We illustrate and evaluate the method with applications to two global expression data sets , one from the model eukaryote Saccharomyces cerevisiae , proliferating at different growth rates , and differentiating cardiopharyngeal cell lineages in the chordate Ciona robusta . We tested the method in a technical replicate dilution study , and in a k-fold validation study .
Accurate transcriptome measurements are central to understanding the fundamental mechanisms of gene expression . A main challenge presented by the RNA-seq method for digitizing information about cellular RNA content—both its composition and abundance—is correcting noise , errors , and biases introduced in the process of making the measurement . An important step in typical library preparation for sequencing is random fragmentation of the molecules to be sequenced . The actual unit that is being digitized during RNA-seq is therefore not the RNA molecules directly , but their fragments , whose number depends on transcript length and RNA abundance . However , it is the molecular abundance of a transcript that is of interest , rather than the distribution of fragments over its gene model . The introduction of Reads per Kilobase of exon model per Million mapped reads ( RPKM ) as the unit measurement by Mortazavi et al . [1] addressed this problem . The work of Mortazavi et al . [1] and of Tarazona et al . [2] both addressed the problem of varyng sample size ( sequencing depth ) . Other researchers looked more carefully at fragment biases and developed a maximum likelihood algorithm to estimate the true differences [3] . The adoption of “read-counts” overlapping each transcript [4] instead of FPKMs ( Fragments per Kilobase of exon model per Million mapped reads ) [5] has allowed a more intuitive interpretation of the data . Among the models that have been proposed for RNA-seq count data [6–9] , the intuitively appealing negative binomial model [10] has become the most popular . The negative binomial probability mass function can be thought of as a mixture distribution arising from concatenating biological noise in transcript abundance ( described by a gamma probability density function ) and sampling noise ( described by a Poisson distribution ) in the compilation of corresponding sequencing counts . Applications of the negative binomial distribution and methods of hypothesis testing have been reviewed recently in [11] . Widely used normalization , statistical modeling , and hypothesis-testing methods are implemented in widely used R packages , edgeR [12] , EDASeq [13] , and DESeq2 [14] . Methods that focus particularly on the removal of unwanted variation from unspecified extraneous , nuisance sources are implemented in the R package RUVSeq [15] . A common assumption among all these methods is that , while the proportions of some transcripts vary across conditions/treatments , most transcripts do not vary between experimental conditions , and the total abundance of cellular RNA remain more-or-less fixed . Lovén et al . [16] were the first to demonstrate an experimental system in which the central assumption of transcriptome equivalence among conditions is not satisfied . The researchers discovered that in cells overexpressing the oncogene cMyc , 90% of all transcripts are also overexpressed . To overcome the problem and allow comparison of expression levels between normal and cMyc overexpressing cells , the group incorporated external spike-ins in their samples , which were then used as a de facto invariant pool of RNAs . The spike-in approach had been previously used successfully in microarray experiments and its use in RNA-seq was facilitated by the development of external RNA spike-in mixes by the External RNA Controls Consortium ( ERCC ) [17 , 18] . The need to normalize high-throughput RNA and DNA counts , in general , by the use of spike-in standards was recently explained and validated in the wide-sweeping paper of [19] . Even more recently , external RNA spike-ins were used in single cell RNA-seq experiments [20] . The ERCC external RNA spike-in mix 1 ( Ambion ) that was used in this study , consists of 92 different synthetic RNAs at 22 different concentrations spanning six orders of magnitude ( 30 , 000–0 . 01 amol/μL ) . Instead of relying on normalization methods that aim to match the cellular and spike-in RNA read-count distributions , we take advantage of the digital nature of the RNA-seq output and we use the spike-ins as calibrators of known absolute abundance in the samples . We demonstrate that our calibration/normalization model is applicable in two different model organisms ( the unicellular eukaryote Saccharomyces cerevisiae and the multicellular chordate Ciona robusta , three experimental setups , a growth rate regulation and a dilution study in yeast , as well as an embryonic differentiation and cell lineage specification study in Ciona ) , and two library preparation protocols . We perform hypothesis testing and detect global amplification of gene expression in both organisms .
We used three distinct experimental setups , two representing different cases in which the assumption of constant transcriptome sizes is violated . We added a fixed , known amount of external RNA spike-ins to the sample of cells . In the rest of the paper , we refer to spike-in abundance per cell to mean the ratio of spike-in amount ( molecules ) added to the sample divided by the number of cells in the sample . In all libraries we avoided the use of poly-dT for reverse transcription as it has been previously shown to be incompatible with quantitation through external spike-ins for RNA-seq [28] . In all cases the filtered aligned reads were converted to counts using the function featureCounts from the package Rsubread ( R , Bioconductor ) and strand information . GR samples and samples for the dilution study were prepared essentially as described in the Borodina et al . [29] directional RNA-seq protocol . We modified the protocol by using UMI adaptors [30 , 31] that were used to eliminate PCR duplicates from the results . RNA was extracted from ten million cells after the addition of 2 μl of 1:20 dilution of ERCC spike-in Mix 1 stock [32] in the lysis buffer . All samples were distributed in two lanes and sequenced on an Illumina HiSeq 2000 , with 100 nt-long , single end reads . The RNA-seq data ( fastq files ) were first filtered for residual rRNA reads . The data were aligned to the latest version of the yeast genome ( sacCer3 ) and spike-in sequences , and filtered for mapping quality using Bowtie with optimized parameters . The aligned reads were then processed with a custom script that removes PCR duplicates based on the combination of mapping coordinates and UMI adaptor barcodes . For the samples of three different in vivo Ciona cell lineages , 800 cells were directly sorted into lysis buffer from RNAqueous-Micro Total RNA Isolation Kit ( Ambion ) containing a fixed amount of ERCC spike-in mix 1 . Total RNA extraction was performed according to the manufacturer’s instructions , followed by depletion of rRNA in the samples . The quality and quantity of total RNA in all stages were measured using Agilent RNA 6000 Pico Kit ( Agilent ) on Agilent 2100 Bioanalyzer . cDNA were synthesized using the SMART-Seq v4 Ultra Low Input RNA Kit ( Clontech ) . RNA-Seq Libraries were prepared and barcoded using Ovation Ultralow System V2 1-16 ( NuGen ) . The samples were pooled in one lane and sequenced on an Illumina HiSeq 2500 , with 50 nt-long , single end reads . The RNA-seq reads were mapped to the Ciona genome ( v . 2008 , ghost database ) using Tophat2 with default parameters . The mapped reads were assigned to Ciona KH gene models ( v . 2013 ) . All computer programming , including data analysis , Monte Carlo simulations , hypothesis testing , and figure preparation were done in the R programming language and environment . For pairwise testing for differential gene expression , we used DESeq function in the DESeq2 package [14] . In some cases hypothesis testing followed the application of an RUV normalization method in a suite of 3 R functions , RUVr , RUVs and RUVg , in the RUVseq package [15] . For maximum likelihood estimation of parameters we used the R function nlm , a general R function that minimizes a supplied objective function over its parameters . We used it to minimize minus log likelihood of the observed data in the context of a model . The first argument of the nlm function is the name of the name of the R function that computes minus log likelihood for the problem at hand . We wrote such a function for each maximum likelihood problem we considered .
We use external RNA spike-ins as a calibration tool to normalize RNA-seq counts by introducing the variable relative yield ( Fig 1 ) . We parametrize a multinomial model for sampling noise , conditioned on native RNA abundances , with library size , relative yield coefficients , and known absolute abundances of spike-in molecules . In the context of this model , we derive a maximum likelihood estimation of nominal RNA abundance , proportional to absolute abundance , for each endogenous transcript in our sample . In the remainder of this paper , we often omit the qualifier “nominal” when we mean nominal abundance , and use the phrase “absolute abundance” when we mean molecules or attomoles ( per cell or per sample ) . In S1 Appendix , we show that the maximum likelihood estimator of RNA abundance is a library dependent scaling of counts by a factor that is proportional to total spike-in counts; it also depends on the known abundance of a reference spike-in , and fraction of overall spike-in counts contributed by this reference spike-in . The expected proportion of counts for a given molecule ( native RNA or spike-in ) i represented in library j depends on the product of its abundance ( attomoles or molecules per cell ) in the original sample ni , j and its relative yield coefficient αi; ( Fig 1B ( ii ) ) . For spike-in molecules , ni , j are known . We define the relative yield coefficient of a molecule ( spike-in or RNA ) to be the ratio of its yield coefficient to that of a reference spike-in . By yield coefficient of a molecule we mean the expected number of fragments per molecule contributed by that molecule to a total RNA-seq library of fixed size and prepared according to a fixed protocol . The relative yield coefficient captures specific properties of an RNA molecule such as transcript length and GC content . By convention , we assign the index 1 to the reference spike-in; consequently α1 = 1 , by definition ( Fig 1B ( i ) ) . For the sake of generality , we do not assume that relative yield coefficient is proportional to transcript length as in [33] . The relative yield coefficient of a spike-in is related to its FPKM within an RNA-seq spike-in library as follows . The relative yield coefficient multiplied by abundance in the original sample , and divided by length is proportional to FPKM . In S1 Appendix , we express the expected proportion of counts for each molecule , indexed by i , in library j in terms of all zi , j = αini , j , in a multinomial joint distribution of counts , and then solve for the maximum likelihood estimator of zi , j . Because we do not estimate the relative yield coefficients , αi for cellular RNA molecules in the present paper , we cannot disentangle here their relative yield coefficients αi and absolute molecular abundances , ni , j ( see Discussion ) . Consequently , we refer to zi , j as a nominal abundance . The corresponding terms for spike-in molecules do not depend on library index j when the same amounts of spike-ins are used in each sample; so , for spike-ins , we can write more simply zi = αini . For sake of clarity , we mention that , by our convention , the indices for the s spike-in molecules are i = 1 , 2 , … , s , and the molecule indices for the q detected native RNA molecules are i = s + 1 , s + 2 , … , s + q . As shown in S1 Appendix , the derived maximum likelihood values of abundances , zi , j for the RNA molecule i in library j are given by z i , j = y i , j ν j for i = s + 1 , s + 2 , … , s + q , and j = 1 , 2 , … , r ( 1 ) where yi , j is the count ( sequencing reads for RNA molecule i in library j and νj is the maximum likelihood calibration constant for library j . The calibration constant νj is given by νj=deff1LjSIn1 , ( 2 ) where f1 is the proportion of spike-in counts across all libraries contributed by the reference spike-in , n1 is the attomoles or molecules per cell , depending on one’s choice of units , for the reference spike-in , and L j SI is the size ( total counts ) of spike-in library j . In S1 Appendix we extend the estimation of zi , j to a full statistical model , including biological variation , linking cellular RNA abundance to RNA-seq counts . The νj calibration constant in Eq ( 1 ) is qualitatively like the dimensionless “technical” ( library ) size factor sj of [34] , but with an explicit relationship to absolute abundance , because 1/νj is on the scale of attomoles or molecules per cell . The relationship between the two factors is discussed thoroughly in S8 Appendix . The numerator on the right-hand side of Eq ( 2 ) , according to our statistical model , is the expected number of counts from the reference spike-in , in replicate j , given the spike-in library size L j SI . As shown in S2 Fig , the “expected” counts given by the model for the reference spike-in , closely approximate the actual number . Therefore , Eq ( 1 ) says that the inferred abundance of RNA transcript i in replicate j is given by the counts for this transcript multiplied by a scale factor that is the attomoles or molecules per cell , per count of the reference spike-in . If it were known that , in fact , RNA transcript i on average yields twice as many aligned counts per amol as the reference spike-in , the abundance of RNA transcript i in the 107 cells ( from which our sample came ) would be given by zi , j/2 . The normalization in Eq ( 2 ) is reminiscent of RPM ( reads per million mapped reads ) normalization , but the denominator involves the total number of reads in the spike-in library only . It makes intuitive sense , because read depth scales spike-in counts and endogenous RNA counts the same way [17] . Therefore , dividing an endogenous RNA count by spike-in counts derived from a fixed number of molecules in the original biological sample simultaneously normalizes for read depth and provides a measure proportional to the molecular abundance of the endogenous RNA in question . [17] applied this sort of normalization to ERCC spike-in counts , but also divided by spike-in length to obtain FPKM ( fragments per kilobase per million mapped reads ) . Under the assumption that count scales with molecular length , FPKM would be proportional to spike-in abundance , in the absence of any molecular biases , e . g . , GC content ( see Fig 2 in [17] ) . As shown in S1 Appendix , the derived maximum likelihood values of abundances for the s spike-in molecules are given by z i = n 1 f 1 f i for i = 1 , … s , ( 3 ) where fi is the empirical fraction of total spike-in counts , across all libraries , that is accounted for by spike-in molecule i , and f1 is that of the reference spike-in . Because the absolute abundances of the spike-ins , ni are known , Eq ( 3 ) and the definition zi = αini , allows us to estimate the spike-in relative yield coefficients as α i = z i n i = n 1 n i f i f 1 ( 4 ) We chose as the reference spike-in the one that contributes the largest fraction of overall spike-in counts . Any of the top few spike-ins would do just as well . We compute the spike-in relative yield coefficients , αi , describe their statistical properties and model the spike-in relative yield coefficients in terms of their biophysical properties in S6 Appendix . However , we do not yet have a large enough repertoire of spike-in molecules nor accurate enough biophysical models to use the model relative yield coefficients for spike-ins to estimate those of native RNA molecules . Thus , the computed spike-in αi terms in the present paper are not participating in the estimation of native RNA abundances . In the Discussion we analyze how αi could be used in the estimation approach . Following the recommendation of [35] , we prepared diagnostic relative-log-expression ( RLE ) plots for all three of our data sets ( dilution study , yeast GR study , Ciona embryonic differentiation study ) to help in evaluating our maximum likelihood ( νj ) calibration method . We found unwanted variation in the total inferred RNA abundance within conditions for all three data sets , and the variations are similar across data sets . We ascribe this variation to technical errors in one or more steps in the preparation of samples to be sequenced: variation in RNA extraction efficiency , error in cell count , dilution and/or volume errors in preparation of the spike-ins added to the cellular RNA . We refer to these errors collectively as library preparation errors . Details of results and analyses are presented in S1 Fig and S2 Appendix . Accordingly , we derived a library-specific scale factor , δj ( S2 Appendix ) , much like the total RNA correction factor , ξj , in the single-cell RNA-seq study of [36] ( S8 Appendix ) . The corrected nominal abundance values are computed as zi , j/δj , and we performed statistical analyses and hypothesis testing on these corrected values . In S8 Appendix we discuss , and in S5 Fig . we illustrate , the similarities and differences of this noise reduction to a removal of unwanted variation ( RUV ) method RUVr in the RUVSeq R package [15] that is based on residuals in a generalized linear model . For this dilution study we prepared 3 replicate libraries with a “high” spike-in aliquot dilution , and 3 with a“low” dilution as described in Materials and Methods . In the absence of any library preparation noise , the molar amount of spike-ins are 4 . 44 times larger in samples that were added to low dilution spike-in aliquots compared to those with the high-dilution aliquots . The generalization of the νj normalization in this case is to simply add the subscript j ( indexing library ) to nref to give ν j = f 1 L j SI / n 1 , j . Perfect performance of our method , would give identical mean abundances for libraries prepared with the high- and low-volume aliquots . In Fig 3 the MA plot shows less than ideal performance in that the ordinates in the scatter plot are offset a bit ( by 0 . 28 ) from zero . This corresponds to a mean fold difference of 1 . 2 rather than 1 . This discrepancy might be due , at least in part , to sample preparation handling errors . Modeling RNA abundances within a condition as gamma-distributed random variables results in counts with the familiar negative binomial distribution , as shown in S1 Appendix . Previous applications of the negative binomial distribution for modeling RNA sequencing counts and various methods of hypothesis testing have recently been reviewed in [11] . Our model for sequencing counts is formally equivalent to that in [34] , but with an important distinction . The library-specific size factor of [34] ( based on genes ) , written according to our notation , is s j = median i y i , j ( ∏ k = 1 r y i , k ) 1 / r . ( 5 ) The size factors sj is a dimensionless constant that carries with it an implicit assumption of fixed total amount of cellular RNA . Furthermore , in practice , the median of sj across libraries is of order 1 . In contrast , although our calibration constants νj can be thought of as “size factors , ” they are proportional to the total spike-in library size L j SI in Eq ( 2 ) , and 1/νj has dimensions of attomoles , or molecules per cell , depending on the units one chooses to use for the spike-in abundances ni ( i ∈ {1 , 2 , …s} ) . Our νj calibration factor is closely related to the extension of [34] , by [36] to estimate cellular RNA abundance with the use of a “technical size factor” resembling that in Eq ( 5 ) , but computed using spike-in counts only . Hypothesis testing methods currently in the literature that are based on the negative binomial distribution of counts with library size factors sj , such as DESeq [34] , can be used following our νj normalization in the manner described in S4 Appendix .
Our focus in this paper is on deriving a nominal cellular molecular abundance that can be converted to absolute abundance by the transcript’s relative yield coefficient , which could be measured in separate experiments . In this study however , we do not attempt to measure the relative yield coefficient values , or estimate the absolute number of molecules per cell for each transcript within a condition . The current work allows us to say , that , for example , RNA transcript A has x times more molecules per cell , on average , in condition 1 compared to condition 2 , even if the corresponding RNA-seq libraries were prepared in different batteries of experiments , different studies , or even prepared in different laboratories . Such a conclusion about what might be called , an absolute ratio of abundances , can be drawn without knowing the relative yield coefficient of transcript A . In the section that follows , we discuss the links between our work and methods by which these relative yield coefficients might be measured . In this manuscript we offer RNA abundance estimates that are proportional to absolute transcript abundance . For this we assign a ( relative ) yield coefficient value of 1 to a reference spike-in , arbitrarily chosen from among those that contribute a sizable fraction of total spike-in counts . Our nominal abundance of an RNA molecule is based on the temporary assumption that this molecule has the same yield coefficient as the reference spike-in . If our calibration method is supplemented with additional data on the effect that a broad range of transcript physicochemical characteristics has on library preparation and sequencing , a more realistic relative yield coefficient could be assigned to each RNA molecule of interest . A technical statement of the outstanding problem is that our inferred nominal abundances zi , j do not disentangle true absolute molecular abundance , ni , j , and the corresponding relative yield coefficient , αi; because , by definition , zi , j = αi ni , j . However , once one measures absolute cellular abundance of transcript i in a preparation of cells from which library j was derived ( ni , j ) , the relative yield coefficient becomes known , at , least in the idealized situation ignoring various sorts of noise , because αi = zi , j/ni , j . For example , ni , j might be measured by single-cell Fluorescence In Situ Hybridization ( FISH ) methods , performed on a large population of cells from which library j was derived . Statistical methods taking into account biological noise and technical noise could be used to compute a confidence interval for αi , provided ni , j could be estimated . Likelihood methods could be used to integrate data across several libraries in the estimation of αi . In principle , once αi is estimated from one or more libraries and a population of cells from which those libraries were derived , this estimate could be used for other libraries ( prepared using the same protocol ) , past , present , and future , to allow the determination of absolute cellular molecular abundances of transcript i . Modeling , like that presented in S6 Appendix and S2 Fig , and like that of [17] could also play a vital role in estimating relative yield coefficients , especially if a wider array of synthetic spike-ins covering a large gamut of physical properties were designed and utilized . Our methods have the potential of facilitating statistical modeling of RNA counts because of the explicit relationship between our nominal abundances and absolute molecular , cellular abundances of RNA . In principle , variation in counts as a consequence of true biological variation in random attomoles , N , and variation in counts due to variation in relative yield coefficient across transcripts with nearly identical mean abundances , μN , could be disentangled . Our approach lays the groundwork for investigating , testing , and modeling how the physical properties—e . g . , length , GC content , folding energy—determine the relative yield coefficient of spike-ins and native RNA transcripts alike . Empirical measurements of relative yield coefficients , as we have defined them , and biophysical modeling could facilitate progress in making the connection between sequencing counts and the underling molecular cellular abundances of the corresponding transcripts . Our work follows up on and extends the work of [15 , 16 , 36 , 43 , 44] . Our inference method is linear and global for each library , like that of [19] , [36] and [45] . We showed that our global ( library specific ) νj calibration constants are closely related to the Anders and Huber-like “technical” size factors of [36] , which are based on spike-in counts . We called their normalization constants s j SI , and we showed that they are proportional to our νj normalization constants in the cases of 2 of our data sets with large library sizes , as predicted by theory ( S8 Appendix ) . An important difference is that the s j SI calibration constants are on a dimensionless scale , on the order of 1 , and do not allow one to infer absolute abundances of transcripts once their relative yield coefficients become known . [16] applied loess normalization to ERCC spike-in counts to derive a normalization function that they then applied to the counts corresponding to native RNA . Our analysis and rigorous testing of our theory and methods suggest that a local nonlinear transformation , such as loess normalization of the count data is not needed for our RNA-seq data . It seems likely that any local nonlinear fitting of counts to make replicate spike-in libraries as similar as possible would involve overfitting the data . Our work has some important features in common with the HTN method of [46] , particularly , the assumptions underlying their Eq ( 1 ) and our Eqs S1 Appendix ( 2 ) and ( 3 ) . These equations explicitly allow for differences in total RNA abundance across conditions . In addition , both normalization methods are global and linear . However the HTN method of [46]: relies on having de facto housekeeping genes rather than experimentally-added spike-ins; does not include a model for biological noise; assumes that relative yield is simply proportional to transcript length; is focused primarily on testing for differential gene expression; and does not provide estimates of absolute RNA abundance . Their global scale factor for a given library is determined by minimizing the sum over spike-ins of the square differences between the spike-in counts in that library and those of a library chosen to be the reference library . That scale factor is then used for the native RNA counts within the same non-reference library . It can be shown that this library-by-library normalization procedure , in the limit as library size ( native RNA and spike-ins ) approaches infinity , will give an abundance measure that is proportional to our z abundance values based on νj normalization . A quite different suite of normalization methods , called RUV ( removal of unwanted variation ) , was introduced by [15 , 36 , 43 , 44] and applied with great effect to many different data sets . The methods involve singular value decomposition ( SVD ) variant of factor analysis to compute a factor matrix W , which is used to model nuisance sources of variation that are unrelated to the experimental design . The factor matrix W is included , in addition to a design matrix , in a generalized linear model for normalized counts . One qualitative way of thinking about the W matrix is that is adds columns to the original design matrix for explanatory variables that one didn’t originally know about . Although this method is widely effective at reducing unwanted variation in RNA-seq data , it does not allow one to infer absolute cellular molecular RNA abundances , even if the factor matrix is computed based on spike-ins or an invariant gene set ( S8 Appendix ) , as the authors are well aware . The simple reason is that proportion of spike-in count is tightly correlated with the biological phenomenon of interest the change of total RNA abundance with condition . However , we showed that results of our maximum likelihood normalization method can be improved , with respect to clustering and detection of differential gene expression , by applying an an RUV method based on residual , RUVr ( RUVSeq package [15] ) after νj normalization . We obtained closely similar results by a simpler method involving a correction factor δj for each library that was based on our discovery in a dilution study with technical replicates that we seem to have some noise in the actual overall amount of spike-ins added to the cellular RNA . We tentatively ascribed these to dilution/volume errors in handling the stock spike-in mixture . This finding highlights the importance of replacing pipetting methods for handling the spike-ins with more accurate robotic methods . The continuing discovery of examples in which there are gross transcriptome differences between cellular states , has established a need for spike-in controls in RNA-seq experiments [19] . Despite some criticisms [15] , external RNA spike-ins have been adopted in several recent studies alongside methods developed to use them for RNA-seq quantitation [16 , 19 , 36 , 46 , 47] . The model presented in this work lends itself for both absolute and relative RNA quantitation , dependent on the experimental ability to accurately isolate a fixed number cells for library preparation . In both cases , we offer evidence that our approach provides reproducible results in a wide variety of conditions and has a strong predictive power . In conclusion , the presented model allows for improved unbiased RNA-seq quantitation in any experimental setup using external RNA spike-ins . | We present a complete statistical model for the analysis of RNA-seq data from a population of cells using external RNA spike-ins and a maximum-likelihood method for genome-wide estimation of transcripts per cell . The model includes biological variability of cellular transcript number and sampling noise . We derive an unbiased estimator of transcripts per cell for every transcript , given by simply multiplying the count by a library-dependent , but transcript-independent , scale factor . This is a nominal estimate that can be converted to an absolute estimate by dividing by the transcript’s relative yield coefficient , measured in a separate experiment . A negative binomial probability mass function with novel normalization ( size ) factors allows for parametric testing of hypotheses concerning dependence of the absolute abundance of each transcript on experimental condition . Our method integrates information from every RNA-seq experiment across all replicates and experimental conditions to determine the calibration constants . We test the method with a dilution study and a k-fold cross-validation study . We illustrate our method with applications to two independent data sets from yeast and the sea squirt that were derived by different library preparation protocols . We show that our methods detect genome-wide amplification of expression , and we compare our method to others . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"cdna",
"libraries",
"sequencing",
"techniques",
"statistics",
"rna",
"extraction",
"fungi",
"mathematics",
"forms",
"of",
"dna",
"dna",
"libraries",
"molecular",
"biology",
"techniques",
"dna",
"rna",
"sequencing",
"cell",
"enumeration",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"extraction",
"techniques",
"research",
"and",
"analysis",
"methods",
"total",
"cell",
"counting",
"gene",
"expression",
"complementary",
"dna",
"molecular",
"biology",
"ribosomes",
"yeast",
"biochemistry",
"rna",
"statistical",
"models",
"eukaryota",
"ribosomal",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"non-coding",
"rna",
"organisms"
]
| 2019 | A complete statistical model for calibration of RNA-seq counts using external spike-ins and maximum likelihood theory |
An increasing risk of Schistosoma mansoni infection has been observed around Lake Victoria , western Kenya since the 1970s . Understanding local transmission dynamics of schistosomiasis is crucial in curtailing increased risk of infection . We carried out a cross sectional study on a population of 310 children from eight primary schools . Overall , a total of 238 ( 76 . 8% ) children were infected with S . mansoni , while seven ( 2 . 3% ) had S . haematobium . The prevalence of hookworm , Trichuris trichiura and Ascaris lumbricoides were 6 . 1% , 5 . 2% and 2 . 3% , respectively . Plasmodium falciparum was the only malaria parasite detected ( 12 . 0% ) . High local population density within a 1 km radius around houses was identified as a major independent risk factor of S . mansoni infection . A spatial cluster of high infection risk was detected around the Mbita causeway following adjustment for population density and other potential risk factors . Population density was shown to be a major factor fuelling schistosome infection while individual socio-economic factors appeared not to affect the infection risk . The high-risk cluster around the Mbita causeway may be explained by the construction of an artificial pathway that may cause increased numbers of S . mansoni host snails through obstruction of the waterway . This construction may have , therefore , a significant negative impact on the health of the local population , especially school-aged children who frequently come in contact with lake water .
Schistosomiasis is a parasitic disease affecting 249 million people worldwide . It is endemic in 78 countries with over 90% of cases occurring in sub-Saharan Africa [1] . About 779 million people , more than 10% of the world's population , were estimated to have been at risk of schistosome infection in mid-2003 [2]–[5] . In Africa , schistosomiasis is due predominantly to infection with Schistosoma mansoni , which causes intestinal schistosomiasis , and Schistosoma haematobium which causes urinary schistosomiasis [6] . Small scale spatial heterogeneity is a typical epidemiological feature of schistosomiasis [7] , [8] . Such heterogeneity is closely associated with the distribution of the snail intermediate host , and with human contact with infective water [9] , [10] . Past studies showed correlation between schistosome transmission and several epidemiological and socio-economic factors such as age [11] , [12] , sex , sources of drinking water , latrine availability , sanitation , hygiene [13]–[17] , urbanization and population growth . [18]–[21] . Moreover , some works have merged both our understanding of demographic risk factors together with environmental transmission dynamics in order to create large scale ( national , regional , continental ) maps that are instrumental in designing control programmes for the disease [22]–[25] , while small scale analysis is important in contributing to the local distinct need [26]–[28] . Schistosomiasis is increasingly a major health problem among communities around Lake Victoria . Geographical patterns of S . mansoni infection have been described in this area in relation to proximity to the lake [25] , [29]–[31] and in comparison of islands versus mainland habitation , where risk is higher on the islands [32] . Identifying local risk factors of infection at multiple levels is crucial so as to understand how transmission varies within small spatial scales and how it changes over time . In addition , identifying risk factors may facilitate disease control by targeting high risk groups or by informing possible intervention strategies . The main objective of this study was to identify the risk factors associated with S . mansoni infection among schoolchildren in Mbita and the two adjacent islands ( Rusinga and Ngodhe ) of Lake Victoria , Kenya .
The study was reviewed and approved by the scientific steering committee and ethical review committee of the Kenya Medical Research Institute , Kenya ( KEMRI , SSC No . 2084 ) , and the ethical review board of Institute of Tropical Medicine , Nagasaki University , Japan ( No . 10121666 ) . Written informed consent was obtained from parents/guardians and schoolchildren prior to the study . Children infected with schistosomes were treated with 40 mg/kg praziquantel and those infected with soil transmitted helminths ( STHs ) were treated with 400 mg albendazole by a clinical officer in accordance with WHO guidelines [33] . All children positive for malaria were treated with artemether/lumefantrine ( AL ) according to national guidelines for uncomplicated malaria [34] . A study feedback meeting was held with parents or guardians of participants , as well as the head masters and health teachers of the schools . This study was conducted on the shores and islands of Lake Victoria in Nyanza province , Mbita district , western Kenya ( Figure 1 ) in an area covered by a health and demographic surveillance system ( HDSS ) [35] , [36] . The Mbita HDSS includes Rusinga east and west on the island and Gembe east and west on the mainland . Ongoing HDSS data showed that the total population in Mbita was 55 , 929 during our survey conducted in 2011 . Notably , population density on Rusinga Island was twice as high as in the Gembe region . In Mbita district , the waterway separating Rusinga Island from the mainland was filled in 1985 and a road to Rusinga Island was constructed to facilitate transportation of people , goods and services . Economic activities are high around and within a 5 km radius from the centre of Mbita causeway , referred to as an urban area , while rest of the study area was treated as rural . Mbita is dominated mainly by the fishing communities living in the immediate vicinity of the lake . The temperature in Mbita ranges from 15°C to 30°C . Rain seasonality is bimodal with a short rainy season starting from October to December , while a longer rainy season lasts from March to May . The average annual rainfall ranges between 800–1 , 200 mm in the western part of the study area in Rusinga Island while Gembe receives slightly higher rainfall of 800–1 , 900 mm . In our study , HDSS data was used for obtaining household locations and population density . A cross-sectional study was conducted between September and November 2011 . According to the education office in Mbita district , the primary school enrollment rate was 91 . 6% . The inclusion criteria of the schools were to be a full grade primary school and not to have received mass-chemotherapy for a year prior to the study . As most of the private primary schools received mass drug administration for STHs a year prior to the study , they were excluded . The schoolchildren in 4th grade were targeted in this study and the total number of them were 1 , 747 in 2011 . Of these , 888 were females and 859 were males . Among the 64 public primary schools , 39 schools met inclusion criteria and 8 schools were randomly selected as clusters ( Figure 1 ) . Parent/guardian and teacher association meetings were held in all selected schools prior to the survey for communicating the study purpose and obtaining their consent with full understanding . Ninety-eight percent of parents/guardians consented and consequently , 310 of all 4th grade children were enrolled in the study . All children were instructed to provide stool specimens in a labeled specimen cup on three consecutive days . The school health teacher or class teacher guided students on stool sample collection during container distribution , a day before the survey . A trained field worker visited the school during morning break time with a registration sheet to ensure all students provided samples . Those who did not provide samples were followed up to ensure each child provided maximum possible samples . The Kato-Katz fecal thick smear technique was used for the detection and the quantification of S . mansoni eggs and the presence of STHs . Intensity of infection was estimated as the number of eggs per gramme of feces ( epg ) [37] , . Slides were prepared and examined by two independent readers within an hour for hookworm egg detection and within 24 hours for the rest of the parasites in focus . Parasite eggs were counted and the arithmetic mean of 3 slides per child was calculated to give the intensity of infection . The extent of S . mansoni infection was categorized as light ( 1–99 epg ) , moderate ( 100–399 epg ) or heavy ( ≥400 epg ) according to WHO guidelines [33] . The intensity of infection per school was calculated as the geometric mean of egg excretion among all children testing within the school . In addition , the presence of S . haematobium and Plasmodium spp . were examined to assess their association with S . mansoni infection . Midday urine was collected for the detection of S . haematobium eggs using direct microscopy examination since S . haematobium is known not to be endemic in the study area [6] . Venous blood was collected for the microscopic examination of Plasmodium spp . by thick and thin Giemsa stained smears . Additional haematological and serological examinations were also carried out for a separate study . To identify risk factors for S . mansoni infections , trained interviewers administered a questionnaire to children during the parasitological survey at the school , while parents/guardians were interviewed in the household setting . Information about individual treatment history for schistosomiasis and the water contact behaviour of each child was collected in the school setting . For socio-economic factors , the household head or the most informed adult present during the household interview gave information on: ownership of land , household size , total number of rooms in house , mother's/female guardian's education level and the main source of drinking water . The age of each child was confirmed by cross checking with official birth certificates or church baptism cards during household visits . In addition , an observation checklist was used to collect information on house structure , latrine and electricity availability in each household . Houses were categorized into two groups: traditional houses with grass roofs and modern houses with iron sheet or cemented roofs . The number of persons per room was obtained as one of the indicators of socio-economic status by dividing the household size by total number of rooms in the house . Households with more than two persons per room were categorized as overcrowded , since the average number of persons per room was two in the study population . Participants were defined as being positive for each helminthic infection if at least 1 egg was detected in their stool for S . mansoni , STHs or in urine for S . haematobium . For P . falciparum infection , both thick and thin blood smears were examined using a light microscope at ×100 with an oil immersion objective . Positive cases were defined as those with at least one malaria parasite detected in the microscopic field of 200 white blood cells for thick film or 2 , 000 red blood cells for thin film [39] . The intensity of helminth infection was expressed as the arithmetic mean of three slides per child , while the intensity of helminth infection per school was expressed as the geometric mean . Total egg counts of S . mansoni in fecal samples were analyzed in relation with potential risk factors by using both a generalized linear model ( GLM ) and a generalized linear mixed model ( GLMM ) with school as a random factor . Since fecal egg count is over-dispersed , a negative binomial generalized linear model ( NB-GLM ) and a negative binomial generalized linear mixed model ( NB-GLMM ) were used . The mixed model was employed to account for the potential lack of independence among samples that emerges from children attending the same school [40] . As children attending the same school were clearly clustered around the school , the school effects might be interpreted as the effects of the areas where children reside . Local population density was obtained for each child using the HDSS population data . The number of people living within a radius of 1 km was counted for all participants' houses using Quantum GIS version 1 . 7 . 4 . [41] . This scale was selected because population density showed the strongest association with S . mansoni infection when the radius was set at 1 km . The HDSS population data was incomplete for Ngodhe Island and therefore we used the total population of the island ( 449 persons; according to a local health staff ) , since most houses on the island were within 1 km from another participant's houses . The shortest straight-line distance from the study participants' house to the lake shore was obtained using Quantum GIS [41] . Spearman's rank correlation was used to test associations between prevalence and intensity of S . mansoni infection and distance to the lake . All statistical analyses were carried out using R version 3 . 0 . 1[42] and P-values less than 0 . 05 were considered significant . The glmmADMB package was applied for analysis of an over-dispersed continuous variable ( infection intensity ) . To examine whether the intensity of S . mansoni infection is spatially clustered , a spatial scan statistical treatment was applied to point data on household location using SaTScan software ( version 9 . 1 . 1 . ) [43] . As models for over-dispersed count data are not available in SaTScan , we applied normal model to the log ( N+1 ) transformed egg count . A purely spatial model was applied and a scan for areas with high values was performed . The maximum size of high-risk clusters was set to 50% of the total number of subjects . To evaluate statistical significance , 999 Monte-Carlo replications were conducted . To examine whether any spatial clusters could be explained by individual risk factors , a scan was also performed with the residual values of a linear regression model of log ( N+1 ) with independent variables which were significantly associated by the egg count in the NB-GLM . The spatial clusters of the residuals can be interpreted as those adjusted for independent risk factors .
Demographic and socio-economic characteristics of the study participants are shown in Table 1 . The study involved 310 fourth grade schoolchildren from eight schools , 138 ( 44 . 5% ) of the children were male , and 172 female ( 55 . 5% ) . Their ages ranged from 9 to 19 years and the median age was 12 years for both sexes . The majority ( 81 . 0% ) of the children lived in traditional houses with an average of 2–3 rooms . Most of the children ( 95 . 3% ) lived in overcrowded houses , 95 . 8% had no electricity and 53 . 9% had no latrine . Over three quarters ( 76 . 4% ) of families owned land and 84 . 8% of mothers/female guardians completed 4th grade or further education . The majority of households ( 84 . 8% ) used the lake as the main source of drinking water . Apart from one child , the rest of the children ( 99 . 7% ) had routine lake water contact an average of 2–3 times per week mainly through bathing and domestic washing purposes . Individual treatment history of schistosomiasis was also confirmed by questionnaire and none of participants was treated at least one year before the study . The overall prevalence of schistosomes , STHs and P . falciparum in each school is summarized in Table 2 . More than three quarters ( 76 . 8% ) of the students were infected with S . mansoni , while seven ( 2 . 3% ) were infected with S . haematobium . All the children infected with S . haematobium were co-infected with S . mansoni and had previously stayed in areas endemic for S . haematobium , further inland from Lake Victoria . At least 12 . 6% of the schoolchildren were infected with one or more species of STHs . Prevalence of hookworm , Trichuris trichiura and Ascaris lumbricoides was 6 . 1% , 5 . 2% and 2 . 3% , respectively . Thirty-seven schoolchildren ( 12 . 0% ) were infected with P . falciparum . A total number of 248 ( 80% ) were infected with at least one of the examined parasites . Co-infection of S . mansoni with P . falciparum 13 . 1% ( 31/236 ) was the most common in the study area . In addition , multiple infections with more than three species were found in a few cases . There was no difference in males and females in co-infection of S . mansoni with P . falciparum ( P = 0 . 52 ) . All possible combinations for parasitic infections were found nearly in the expected numbers ( data not shown ) , indicating neither synergistic nor antagonistic effects of polyparasitism . The prevalence of S . mansoni differed significantly between schools , ranging from 31 . 7 to 98 . 3 percent , ( Pearson chi-square test P<0 . 001 ) . Age was not associated with prevalence of S . mansoni in this study . There was no significant difference between males and females for prevalence of any examined parasitic infections . Table 3 shows the intensity of S . mansoni infection in each school . Among those who were positive for S . mansoni eggs , the geometric mean number of eggs excreted per gramme of feces ( epg ) varied from 2 . 0 to 303 . 5 epg between schools . The overall mean intensity of S . mansoni infection was 207 epg with inter quartile range of 8 to 214 epg . The intensity of S . mansoni infection was categorized according to the WHO guidelines [33] , children with light , moderate and heavy infections were 110 ( 35 . 5% ) , 78 ( 25 . 2% ) and 50 ( 16 . 1% ) respectively . At the school level , the intensity of S . mansoni infections was strongly correlated with its prevalence ( Spearman's rank correlation , rho = 0 . 98 , P<0 . 001 ) . The four schools with high prevalence and intensity of S . mansoni infection ( Wasaria , Wakondo , Kamasengre , and Kombe ) were aggregated around the bay in the west side of Mbita causeway ( Figure 2 ) . The mean of log ( N+1 ) transformed egg counts were significantly different between inside and outside of clusters as 5 . 23 and 2 . 55 , respectively ( common estimate for standard deviation , 1 . 94; P = 0 . 001 ) . Table 4 shows the results of a bivariate analysis on the association between the intensity of S . mansoni infection and the potential risk factors with no consideration of school effects ( NB-GLM ) . This indicated that males were more intensely infected than females ( marginally significant ) . Several household-based factors also showed a significant association with high infection risk; houses in areas with higher population density , permanent houses and houses with latrine . Spatial scan statistics was performed for the residuals of the regression model of log ( N+1 ) transformed egg count . The variables that were found to be significant and/or marginally significant in the NB-GLM ( sex , population density , house structure and houses with latrine; orange dotted circle in Figure 2 ) were included in the calculation . A significant high-risk cluster occurred in a similar location to the unadjusted cluster ( red circle in Figure 2 ) although the size of the cluster was smaller ( radius of 4 , 006 meters; 53 children were included ) . The adjusted cluster included all the children of Wakondo and some of the children of Wasaria and Kamasengre but did not include any of the children living in Kombe . The mean of the residuals was significantly higher inside ( 1 . 50 ) than outside the cluster ( −0 . 31 ) , common estimate of standard deviation , 1 . 92; P = 0 . 001 . When the school areas were included as random factor in a NB-GLMM , the effects of sex , house structure and latrine became non-significant ( Table 5 ) . Local population density was the only statistically significant factor for S . mansoni infection in the NB-GLMM ( Table 5; P = 0 . 011 ) . The association between population density and intensity of S . mansoni infection is further depicted in the map and scatter plot as shown in Figure 2 and 3 . Evidently , population density is an independent factor influencing risk and intensity of S . mansoni infection , while evaluated socio-economic factors appeared not to affect the risk and intensity of S . mansoni infection in this study area .
This study goes some way towards elucidating the risk factors associated with S . mansoni infection among schoolchildren in Mbita district , western Kenya . The prevalence of S . mansoni was high in almost all the schools sampled and more than three quarters ( 76 . 8% ) of children were infected with S . mansoni . This result is consistent with the previous reports [32] , [44] , in which they showed an increased risk of S . mansoni infection compared to the early 1970s [45] , [46] , at which time the prevalence of S . mansoni was less than 50% among schoolchildren along the shores and islands of Lake Victoria in Mbita . In a bivariate NB-GLM analysis , children living in permanent houses with latrine were infected with larger numbers of S . mansoni eggs . However , the statistical significance of these effects did not remain in the model when considering school effects as a random factor ( NB-GLMM ) . This was considerable based on previous studies which showed higher infection risk associated with lower socioeconomic status [47] . We attribute this to the fact that families living in permanent houses with latrine tended to be found in densely populated areas where infection risk was high , since the local habit of defecation along lake shore is common even though the most of families living around town centre have latrines [28] . In the result , the residential location was closely associated with S . mansoni infection risk . This finding corroborates previous research by Booth and colleagues , which clearly indicated that environmental living circumstances were tightly connected with infection status and disease burden . In short , environmental exposure due to residential location rather than some fixed characteristics of an individual determines risk of infection [48] . Several studies have reported high risks of S . mansoni infection among people living close to a permanent water body [25] , [29]–[31] . However , the effect of the distance to the lake was not significant in the present study . This could be the result of a small range of proximity to the lake , as the schools surveyed were all located within 1 . 0 km , and the children lived within 2 km , of Lake Victoria . Population density was the single most important factor associated with S . mansoni infection risk on the shores and islands of Lake Victoria . Theoretically , the basic reproductive number ( R0 ) of schistosomiasis linearly increases with human density . This is due to the fact that the rate of infection among snails depends on the absolute number , not the prevalence of infected hosts [49] . Thus higher infection risk in densely populated areas can be explained purely by numerical dynamics of transmission . In addition , higher nutritional load via domestic waste water from densely populated areas might enhance population growth of the snails [50] . We can therefore strongly suggest that the increase in population density in recent decades may partially explain the increase in S . mansoni prevalence in this area . Notably , our study revealed the highest prevalence and intensity of S . mansoni infection was around the Mbita causeway . There was a tendency that infection risk decreased towards the eastern part of the mainland ( Figure 2 ) . Additionally , infection risk was very high in the three schools on Rusinga Island but not on Ngodhe Island . Our results suggest that a simple dichotomy like island-mainland comparison may obscure micro-geographical heterogeneity in S . mansoni transmission . This calls for additional ecological and environmental survey to understand the distribution and population dynamics of snail intermediate host which directly relates with the transmission of schistosomes . Spatial analysis indicated a high-risk cluster that includes the town center , the causeway and nearby villages in Rusinga . The high risk of infection in these areas could be partially explained by local population density . However , a significant high-risk cluster remained in a similar location even after adjustment for the effects of local population and other potential risk factors . Therefore , an aggregated risk factor that was not measured in the present study may exist in the west side of Mbita causeway . The construction of Mbita causeway in the 1980s has likely impacted the ecosystem surrounding Rusinga and the mainland , by promoting population activities , restricting water circulation and free movement of aquatic biota through blockage of the natural channels [51] . Such a change in ecological conditions may be one of the reasons why S . mansoni prevalence has drastically increased compared with 1970s , before the causeway construction [52] , [53] . To conclude , increased risk of S . mansoni infection was observed in Mbita along the shores and islands of Lake Victoria . Moreover , the infection risk of S . mansoni was associated with high population density and was concentrated around the Mbita causeway . Urgent intervention efforts should be considered in order to reduce morbidity and mortality due to S . mansoni infection , taking into consideration region-specific risk factors for disease transmission . | It is estimated that more than ten percent of the world's population is at risk of schistosome transmission , with over 90% of infections occurring in sub-Saharan Africa . In Kenya , schistosomiasis remains a major public health concern particularly around Lake Victoria . The objective of this study was to identify the risk factors associated with Schistosoma mansoni infection among schoolchildren on the shores and adjacent islands of Lake Victoria in Mbita district , western Kenya . High local population density was identified as an important risk factor for S . mansoni infection . Socio-economic factors were not found to be significantly associated with infection risk . Our study suggests that environmental changes related to causeway construction and the dense human population around Mbita town may result in favourable ecological conditions for S . mansoni transmission . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"infectious",
"diseases",
"helminth",
"infections",
"medicine",
"and",
"health",
"sciences",
"schistosomiasis",
"infectious",
"disease",
"epidemiology",
"epidemiology",
"neglected",
"tropical",
"diseases",
"tropical",
"diseases",
"parasitic",
"diseases"
]
| 2014 | Risk Factors and Spatial Distribution of Schistosoma mansoni Infection among Primary School Children in Mbita District, Western Kenya |
Drosophila Lnk is the single ancestral orthologue of a highly conserved family of structurally-related intracellular adaptor proteins , the SH2B proteins . As adaptors , they lack catalytic activity but contain several protein–protein interaction domains , thus playing a critical role in signal transduction from receptor tyrosine kinases to form protein networks . Physiological studies of SH2B function in mammals have produced conflicting data . However , a recent study in Drosophila has shown that Lnk is an important regulator of the insulin/insulin-like growth factor ( IGF ) -1 signaling ( IIS ) pathway during growth , functioning in parallel to the insulin receptor substrate , Chico . As this pathway also has an evolutionary conserved role in the determination of organism lifespan , we investigated whether Lnk is required for normal lifespan in Drosophila . Phenotypic analysis of mutants for Lnk revealed that loss of Lnk function results in increased lifespan and improved survival under conditions of oxidative stress and starvation . Starvation resistance was found to be associated with increased metabolic stores of carbohydrates and lipids indicative of impaired metabolism . Biochemical and genetic data suggest that Lnk functions in both the IIS and Ras/Mitogen activated protein Kinase ( MapK ) signaling pathways . Microarray studies support this model , showing transcriptional feedback onto genes in both pathways as well as indicating global changes in both lipid and carbohydrate metabolism . Finally , our data also suggest that Lnk itself may be a direct target of the IIS responsive transcription factor , dFoxo , and that dFoxo may repress Lnk expression . We therefore describe novel functions for a member of the SH2B protein family and provide the first evidence for potential mechanisms of SH2B regulation . Our findings suggest that IIS signaling in Drosophila may require the activity of a second intracellular adaptor , thereby yielding fundamental new insights into the functioning and role of the IIS pathway in ageing and metabolism .
SH2B proteins are a recently identified family of intracellular adaptor proteins that transduce signals downstream of a number of receptor tyrosine kinases ( RTKs ) . These include the receptors for insulin , insulin-like growth factor-1 , Janus kinase 2 ( Jak2 ) , platelet derived growth factor , fibroblast growth factor and nerve growth factor [1]–[5] . Consequently , SH2B proteins have been shown to function during multiple physiological processes including glucose homeostasis , energy metabolism , hematopoesis and reproduction [6]–[9] . Moreover , mutations in SH2B orthologues in humans are associated with metabolic disregulation and obesity . Several SH2B family members have been identified in mammals so far including SH2B1 ( of which there are four splice variants: SH2B1α , SH2B1β , SH2B1γ and SH2B1δ ) , SH2B2 ( APS ) and SH2B3 ( Lnk ) . They are characterised by a number of conserved domains including a central pleckstrin homology ( PH- ) domain , a C-terminal Src Homology 2 ( SH2- ) domain , an N-terminal proline rich region , multiple consensus sites for tyrosine and serine/threonine phosphorylation and a highly conserved C-terminal c-Cbl recognition motif [6] , [10]–[12] . These domains function as protein-protein interaction motifs and so allow SH2B proteins to integrate and transduce intracellular signals from multiple signaling networks in the absence of intrinsic catalytic activity [6] , [10]–[12] . Biochemical studies have demonstrated that SH2B proteins bind via their SH2 domains to phosphotyrosine residues within the intracellular tails of several activated RTKs thereby contributing to receptor activation [10] , [13] , [14] . Once bound , SH2B proteins have been shown to undergo RTK-stimulated tyrosine phosphorylation although they might also be serine/threonine phosphorylated in their basal state as they show anomalous migration on SDS/PAGE indicative of protein structural modifications [13] , [15] , [16] . In vitro binding assays have identified interactions between SH2B proteins and a number of other intracellular adaptor proteins including the insulin receptor substrates IRS1 and IRS2 , Grb2 , Shc and c-Cbl [2] , [17] , [18] . These interactions may or may not require tyrosine phosphorylation of SH2B depending on the isoform studied [2] , [18] . Interactions with IRS proteins promote activation of the phosphoinositol-3 kinase ( PI3K ) pathway and overexpression in cell culture has been show to enhance activation of both the PI3K and the Ras/MapK pathways [17] , [19] . Binding to the proto-oncogene product , c-Cbl , a RING-type E2-dependent ubiquitin protein ligase , may facilitate either endocytosis or degradation of the receptor through receptor ubiquitination [16] , [20] . Thus , SH2B proteins may have dual functionality in both positively and negatively regulating RTK signaling . Mammalian SH2B family members are widely expressed in a number of tissues suggesting that they may share some overlapping , redundant functions [13] , [21] , [22] . For example , mice carrying a genetic deletion for SH2B3 show a selective defect in the regulation of B cell lymphopoeisis . This is consistent with the high levels of SH2B3 expression observed in hematopoetic organs such as the bone marrow and lymph nodes [22] and suggests that SH2B3 plays a specific , non-redundant role in the development of a subset of immune cells . However , SH2B3 mRNA is also abundant in non-hematopoetic tissues such as testis , brain and muscle and so presumably the absence of phenotype in these tissues indicates redundancy with other SH2B family members [22]–[24] . Studies into the physiological functions of SH2B1 and SH2B2 have produced contradictory results . Genetic deletion of SH2B1 in mice produces neonatal growth retardation and infertility probably due to impaired responses to GH or IGF-1 [7] . It was reported that SH2B1 null mice rapidly increase their body mass and develop obesity as a result of significantly impaired hypothalamic leptin signaling resulting in hyperleptinemia and hyperphagia [8] , [25] . These mice were also shown to have attenuated insulin signaling in muscle , liver and fat resulting in insulin resistance and diabetes . More recently , a second model showed that SH2B1 null mice actually have decreased fat mass possibly caused by a reduction in adipogenesis as SH2B1 deficiency was associated with reduced expression of adipogenic genes such as peroxisome proliferator-activated receptor γ ( PPARγ ) and impaired adipocyte differentiation in cell culture [26] . In the case of SH2B2 , it was reported that SH2B2 null mice develop hypoinsulinemia and show increased insulin sensitivity at young ages [27] . However , more recent reports saw no effect of SH2B2 deletion on fasted blood glucose , insulin levels , glucose or insulin tolerance [9] . The reasons for this apparent discrepancy between studies is unclear but may be confounded by differences in genetic backgrounds , diet or housing conditions . Understanding the physiological functions of SH2B proteins in mammals has therefore been complicated by the presence of multiple SH2B isoforms and conflicting data from genetic analyses . The genome of Drosophila melanogaster encodes a single SH2B homologue ( Lnk ) that shares a similar domain structure to its mammalian counterparts , with 36% sequence identity to human SH2B proteins in its PH-domain and 74% sequence identity in its PTB domain as well as containing a highly conserved c-Cbl binding motif . Furthermore , most of the basic metabolic and signaling pathways that maintain homeostasis are conserved in the fly providing an ideal context for in vivo studies of SH2B biological function . Recent evidence has shown that Drosophila Lnk is a key regulator of cell growth and proliferation during development [28] . Loss-of-function mutations in Lnk produce phenotypes reminiscent of reduced IIS signalling such as growth reduction , developmental delay and female sterility . Genetic epistasis experiments indicated that Lnk functions downstream of the Drosophila Insulin Receptor ( dInR ) and upstream of PI3K in IIS-mediated growth control . Genetic epistasis suggested that Lnk may play a similar role as the insulin receptor substrate , Chico , in the activation of PI3K upon dInR stimulation during growth [28] . Mutations that reduce IIS activity in C . elegans , Drosophila and mouse can increase lifespan in all three organisms , demonstrating that the IIS pathway has evolutionary conserved roles in the determination of adult lifespan . In Drosophila , the effects of insulin receptor activity on lifespan determination are mediated via the Chico/PI3K/forkhead transcription factor [29]–[32] . Therefore , we investigated whether Lnk also plays a role in the determination of adult lifespan . Here , we show that Lnk mutant flies exhibit increased lifespan as well as improved survival under conditions of oxidative stress and starvation . We also show that Lnk loss-of-function results in increased stored energy reserves associated with transcriptional changes in genes involved in both lipid and carbohydrate metabolism . Biochemical and genetic data indicate that Lnk functions within both the IIS and Ras/MapK signaling cascades and is itself a direct target for transcriptional regulation by the dFoxo transcription factor .
Novel alleles of Lnk were recently isolated in a genetic screen looking for new regulators of growth in flies as Lnk loss-of-function clones were found to cause cell-autonomous growth inhibition in the developing eye [28] . We have characterised two additional mutant alleles of Lnk: Lnkd07478 containing a P-element insertion within the first intron of the Lnk locus and LnkDel29 , a small deletion generated by FLP-FRT recombination between two pBAC elements that removes the first two exons of Lnk including the predicted translational start site ( Figure 1A ) . Homozygous mutants had significantly reduced levels of Lnk transcripts as measured by quantitative RT-PCR ( Figure 1B ) . Homozygous and transheterozygous mutants under normal culture conditions were adult viable but developmentally delayed with an overall reduction in body size as a result of reduced cell size and cell number ( Figure 1C and 1D ) . No further reductions in growth were observed in hemizygous combinations over a deficiency that removes the entire Lnk locus suggesting that they represent strong loss-of-function alleles ( Figure 1C ) . In addition , we were able to fully rescue the growth defects of LnkDel29 homozygotes by introducing a genomic rescue construct containing the entire Lnk locus indicating that these growth defects are specific to Lnk ( Figure 1C ) . Both alleles were backcrossed for more than eight generations into two distinct genetic backgrounds: the inbred w1118 strain and the outbred wDahomey ( wDah ) strain . We then assayed heterozygotes and homozygotes of both alleles for longevity . After backcrossing into the w1118 genetic background , heterozygosity for either Lnkd07478 or LnkDel29 did not result in any significant differences in lifespan in either males or females ( Figure 2A , 2B , 2E , and 2F ) . In contrast , we observed significant increases in both median and maximum lifespan in both males and females homozygous mutant for either allele when compared to wild-type controls ( Figure 2A , 2B , 2E , and 2F ) . Furthermore , the longevity effects of LnkDel29 males and females were fully reproducible after backcrossing into wDah ( Figure 2C and 2G ) and the lifespan extension observed in females homozygous mutant for LnkDel29 was fully rescued by the introduction of a Lnk genomic rescue construct ( Figure 2I ) , thereby confirming a role for Lnk in lifespan determination . In addition to increased lifespan , homozygous Lnk females produced significantly fewer eggs compared to their wild-type counterparts , especially LnkDel29 homozygous females , which were practically sterile ( Figure 2D ) . Furthermore , Lnk mutant ovaries were dramatically reduced in size and contained immature oocytes that were arrested in previtellogenic stages of oogenesis ( data not shown ) and the egg laying defects observed in LnkDel29 homozygous females were fully rescued in the presence of a Lnk genomic rescue construct ( Figure 2J ) . We observed no obvious defects in the fertility of Lnk homozygous males and females mated to Lnk mutant males produced comparable numbers of eggs as females mated to w1118 males ( Figure 2H ) . Interventions that extend lifespan are often associated with enhanced resistance to various stresses [29] , [33] . We therefore tested the ability of Lnk mutant flies to survive under conditions of oxidative stress and starvation . To induce oxidative stress , flies were starved for 5 hours and then fed 5% hydrogen peroxide in a sucrose/agar media . Both males and females , homozygous mutant for Lnk , showed significantly increased median survival times when fed 5% hydrogen peroxide compared to control flies under an identical regime ( Figure 3A and 3B ) . Furthermore , this increased resistance to hydrogen peroxide was fully rescued in both sexes upon introduction of the Lnk genomic rescue construct ( Figure 3A and 3B ) . We also observed a significant increase in survival times when Lnk mutant males and females were maintained on an agar-only diet to induce starvation ( Figure 3C and 3D ) . Again , the starvation resistance observed in Lnk mutants was fully rescued in both sexes in the presence of the Lnk genomic rescue construct ( Figure 3C and 3D ) . Moreover , resistance to hydrogen peroxide and starvation were observed with both Lnk mutant alleles and in both genetic backgrounds ( Figure S1 ) . Enhanced survival under conditions of starvation is often associated with increased levels of stored energy resources indicative of a disruption to metabolic homeostasis . In flies , metabolised nutrients are primarily stored as triglycerides ( TAG ) and glycogen in the fat body , the insect equivalent of the mammalian liver and white adipose tissue . We observed significantly elevated levels of both TAG and glycogen in whole-fly extracts of both males and females when we compared Lnk mutants to wild-type controls ( Figure 3E and 3F ) . These elevated levels of TAG and glycogen were restored back down to those observed in wild-type flies in the presence of a Lnk genomic rescue construct ( Figure 3E and 3F ) Despite the observed differences in metabolic stores , we did not detect any obvious differences in the feeding behaviour of Lnk mutant flies compared to age-matched controls ( Figure S2 ) suggesting that this increase in metabolic stores is unlikely to be mediated by increased feeding but by changes in cellular metabolism . In addition to glycogen , adult insects possess a second metabolic pool of carbohydrate in the form of the disaccharide trehalose which is a major sugar in the fat body , thorax muscles and hemolymph and is rapidly consumed during certain energy-requiring activities such as flight . We found that whole-body levels of trehalose were also significantly increased in Lnk mutant males and females when compared to controls ( Figure 3G ) and again , these elevated levels of trehalose were restored to those observed in wild-type flies by the introduction of the Lnk genomic rescue construct ( Figure 3G ) . However , when we measured trehalose levels in hemolymph extracted from either third instar or adult flies we found no significant differences between Lnk mutants and controls ( data not shown ) . The total volume of hemolymph in an adult fly is extremely small ( approximately 0 . 1 µl ) and so the contribution of hemolymph trehalose to the total trehalose content can be regarded as negligible . Thus , the increase in trehalose content in whole fly extracts is almost certainly caused by increased tissue trehalose stores . Insect hemolymph also contains circulating glucose which is obtained from the diet and again , we found no significant differences in circulating glucose levels in Lnk mutants compared to controls ( data not shown ) . In order to investigate further the molecular mechanisms of Lnk function , we performed microarray studies comparing the transcriptome of homozygous Lnk mutants to controls . FlyAtlas , a microarray-based atlas of adult gene expression in multiple Drosophila tissues ( http://www . flyatlas . org; [34] ) , shows that Lnk mRNA is widely expressed in the adult fly but that transcripts are particularly enriched in the central nervous system . We therefore performed our transcriptome analysis on RNAs extracted from the heads of control and Lnk homozygous mutant females . After extraction , RNAs were labelled and hybridised to Affymetrix Drosophila 2 . 0 microarrays . All experiments were conducted in quadruplicate to facilitate statistical analysis . The raw data files were background corrected and normalised using the R programming language ( see Material and Methods ) . Using biological annotation available through the Gene Ontology ( GO ) , we analysed our dataset using Catmap analysis . Catmap assigns significance to functional categories based on their representation within a ranked list of differentially expressed genes . This generated a list of GO terms associated with genes that show altered expression in Lnk mutants compared to controls ( Table S1 ) . The majority of the downregulated GO terms are involved in the metabolism of carbohydrates , amino acids , lipids and fatty acids suggesting that cellular metabolic processes are downregulated in Lnk mutant animals . Among the most significant upregulated GO terms are many linked to signal transduction and transcription indicating that these processes are upregulated in Lnk mutants compared to controls . We then determined gene expression changes in our data set using a linear model . This study revealed that 2483 transcripts show significant differential expression ( p<0 . 05; >0 . 1-fold ) between Lnk mutants and controls with 1768 genes showing increased expression and 715 genes with decreased expression ( Table S2 ) . We compared this data set to a previously reported list of 484 transcripts that function in Drosophila metabolic pathways [35] and found that a number of genes in our differentially expressed gene list overlap with genes that regulate carbohydrate and lipid catabolism ( Table S3 ) . Downregulated genes included genes encoding several enzymes of the glycolytic pathway and the mitochondrial β-oxidation pathway while genes involved in glycogen synthesis and lipid storage showed upregulated expression . These changes in gene expression are consistent with an overall metabolic switch from catabolism to synthesis/storage and are congruent with our findings that Lnk mutants show increased levels of metabolic stores . Interestingly , a number of transcripts that function in the IIS pathway were found to be upregulated in Lnk mutants compared to controls . The mammalian SH2B proteins have been shown biochemically to function as intracellular adaptors for the mammalian insulin receptor and recent genetic data from Drosophila has shown that Lnk may play a similar role to chico during IIS-mediated growth control . We therefore compared our data set to a comprehensive list of transcripts that function in the IIS pathway in Drosophila ( Figure 4A and Table S4 ) . We found upregulation of transcripts encoding positive regulators of IIS including the insulin-like ligands dilp2 , dilp3 , dilp5 and dilp6 as well as chico , Dp110 , PDK-1 and dAkt . In contrast , we found downregulation of transcripts that encode negative regulators of the IIS pathway such as the IGFBP-like , ImpL2 , and the PI3kinase inhibitor , Susi . Several of these IIS gene expression changes were further confirmed by quantitative RT-PCR ( Figure 4B ) . Transcriptional outputs of the IIS pathway are mediated via several downstream effectors including the forkhead transcription factor , dFoxo , the Ras/MapK signaling pathway and the protein kinase complex TORC1 , all of which have been shown to regulate gene expression either directly or indirectly . Further examination of our microarray data set identified four known dFoxo target genes with upregulated expression: split-ends ( CG18497 ) , ches-1-like ( CG12690 ) , eIF-4E ( CG4035 ) and CG9009 ( Table S4 ) . In addition , we observed increased expression of two additional well-characterised dFoxo target genes , 4eBP and dInR by qRT-PCR ( Figure 4C ) . These data therefore suggest that dFoxo activity is increased in Lnk mutant flies . Using EASE analysis followed by Fisher's exact test for statistical significance and Bonferroni correction for multiple comparisons , we found that IIS pathway genes and genes classified by Flybase in the functional category of Ras signal transduction were over-represented in our data set ( p = 0 . 002 and p = 0 . 004 , respectively ) ( Table S4 ) . In contrast , genes of the canonical TOR signaling pathway were not significantly over-represented in our data set ( p = 0 . 784 ) ( Table S4 ) . Taken together , these data suggest significant transcriptional feedback onto the IIS via dFoxo and the Ras/MapK pathway but not via TOR signaling in Lnk mutant animals . The potential for transcriptional feedback by dFoxo onto upstream components of the IIS cascade suggested that expression of Lnk itself may be regulated by dFoxo activity . We therefore looked for perfect matches to the mouse Foxo1/Foxo4 consensus binding site ( RWWAACA ) within 3Kb upstream of the Lnk translational start site and identified eight putative dFoxo binding sites in the Lnk promoter . To determine if dFoxo is indeed bound at the Lnk promoter , we performed chromatin immunoprecipitation ( ChIP ) using a specific dFoxo antibody [36] . Quantitative PCR ( qPCR ) was used to compare the relative DNA binding of dFoxo at a 5′ region of the Lnk promoter to two negative control genomic regions: a region within the U6 snRNA promoter and a region 3′ to the Lnk locus just downstream of the last exon ( Figure 5A ) . We observed a significant increase in the relative DNA binding at the Lnk promoter region compared to the negative controls ( Figure 5B ) . The magnitude of this increase in relative DNA binding ( approximately 2-fold ) was comparable to that observed for Lk6 , a known dFoxo target gene ( Figure 5B ) . In addition , we observed further increases in dFoxo DNA-binding at both loci in flies that had been starved or treated with paraquat prior to chromatin extraction , conditions in which dFoxo is activated ( Figure 5B ) . Furthermore , quantitative RT-PCR analysis of Lnk expression in RNA extracts from foxo mutant flies revealed that Lnk transcript levels are significantly elevated the absence of dFoxo ( Figure 5C ) suggesting that dFoxo may normally function to repress Lnk expression . To assess the biological significance of a regulatory role for Lnk in the IIS and Ras/MapK pathways , we examined the effects of RNAi-mediated knockdown of Lnk expression on insulin-stimulated signaling in insect cells . Activation of the dInR by insulin triggers activation of both the PI3K and the Ras/MapK branches of the insulin signaling pathways resulting in the phosphorylation of various intracellular effectors , including Akt and the MapK , Erk-A [37] . RNAi-mediated knockdown of Lnk resulted in reduced levels of phosphorylated Akt and Erk-A upon insulin stimulation with no significant change in the levels of total protein ( Figure 6 ) . This reduction in phosphorylated Akt and Erk-A was comparable to that caused by RNAi-mediated knockdown of either the dInR or its intracellular substrate , chico , suggesting that Lnk expression is required for full insulin signaling transduction via both the PI3K and MAPK branches of the IIS pathway in cultured cells .
Our understanding of the physiological roles of the SH2B family of intracellular adaptors has been complicated by the presence of multiple family members in mammals . Furthermore , phenotypic analysis of genetic knockouts in mice has produced contradictory results . Recent genetic evidence has described a role for the single ancestral SH2B protein in Drosophila ( Lnk ) during IIS-mediated growth control . Here , we have characterised a critical role for Lnk in the regulation of lifespan , stress responses and cellular metabolism . Our results support a model in which Lnk functions as an intracellular adaptor for transduction of the IIS and Ras/MapK signaling cascades to mediate these physiological processes . A recent genetic study has shown that mutations in Drosophila Lnk produce phenotypes reminiscent of reduced IIS during development including impaired growth , developmental delay and female sterility . Genetic epistasis experiments placed Lnk downstream of dInR and upstream of PI3K at the same level as Chico , the single fly insulin receptor substrate . Mutations in both chico and Lnk produce similar phenotypes and display similar reductions in IIS activity . Furthermore , flies homozygous mutant for both genes are lethal suggesting that they may be functionally redundant . The precise mechanisms whereby mammalian SH2B proteins transduce intracellular signaling from the insulin receptor remain unclear although like the IRS proteins , they have been shown to bind to multiple downstream mediators such as PI3K and Grb2 [6] . However , Drosophila Lnk lacks a consensus binding site for PI3K which is present in Chico so it is unlikely that they regulate similar downstream mechanisms . The IIS pathway has an evolutionary conserved role in the determination of adult lifespan mediated by the Chico/PI3K/dFoxo branch of the IIS cascade . Previous studies have shown that flies either homozygous or heterozygous for chico1 , a strong loss-of-function allele of chico , show increased lifespan [38] . We have shown that Lnk homozygotes also show increased lifespan although no obvious effects on lifespan were observed in heterozygous animals . Interestingly , the effects of Lnk mutation on lifespan extension were similar in both males and females , which is uncommon in Drosophila , even for IIS mutants . This data therefore suggests that as during growth regulation , signaling via the activated dInR during lifespan determination may require a second intracellular adaptor in addition to the insulin receptor substrate , Chico , and provides the first evidence of a role for SH2B proteins in lifespan determination . Lifespan extension in females was associated with reduced fecundity as a result of an arrest in oogenesis . However , there were no visible effects of Lnk mutation on male fertility as measured by offspring production . As male homozygous mutants were also long-lived , this suggests that the extended lifespan of Lnk mutant females is not simply due to reduced fecundity . Genetic knockouts of SH2B1 in mice also show infertility due to impaired signal transduction from the IGF-1 receptor resulting in poor gonad development [7] . The sex-specific differences on fertility observed in Lnk mutants are probably due to sex-specific differences in Lnk transcript expression as microarray analyses of Drosophila gene expression has shown that Lnk transcripts are enriched within the female ovary but not in the male testis or accessory glands ( http://www . flyatlas . org; [34] ) . A comparison of the transcriptomes of Lnk mutant flies to controls revealed a number of gene expression changes associated with genes that encode components of the Drosophila IIS pathway . Hence , we observed upregulation of a number of factors that potentiate IIS such as the insulin-like ligands dilp2 , dilp3 , dilp5 and dilp6 , as well as the insulin receptor substrate chico , the Drosophila class I PI3K , Dp110 , phosphoinositide-dependent protein kinase PDK-1 and dAkt . In contrast , the expression of negative regulators of IIS such as the IGFBP-like ImpL2 and the PI3kinase inhibitor susi were downregulated . Several of these changes in expression were confirmed by qRT-PCR analysis and these data suggest that IIS transduction is affected by Lnk mutation , further strengthening the genetic evidence that Lnk is a component of the IIS pathway in flies . Transcriptional regulation downstream of IIS is in part mediated by the dFoxo transcription factor which is activated in response to low IIS by dAkt-mediated phosphorylation . While we did not observe any differences in dFoxo mRNA or protein levels in Lnk mutants compared to controls ( data not shown ) , a number of dFoxo target genes did show changes in expression . Thus , split-ends ( CG18497 ) , ches-1-like ( CG12690 ) , eIF-4E ( CG4035 ) and CG9009 all showed upregulated expression in our microarray data set . We also observed increased expression of two well-characterised dFoxo target genes , 4eBP and dInR , by quantitative RT-PCR . Taken together , these data suggest that dFoxo activity may be increased in Lnk mutant animals . Interestingly , we observed a marked difference in the magnitude of increased expression of both 4eBP and dInR between different body parts . Thus , for 4eBP we observed a 1 . 1-fold increase in expression in head RNA extracts compared to a 3 . 8-fold increase in RNA extracts from bodies . Similarly , for dInR we observed a 1 . 5-fold increase in expression in head RNA extracts compared to a 2 . 6-fold increase in body RNA extracts . These data suggest that different tissues may exhibit differences in the magnitude of the transcriptional response to Lnk loss of function . As our microarray experiments were performed on RNA isolated from adult heads only , this may explain why 4eBP and dInR were not identified in the microarray data set as microarray analysis of gene expression is generally regarded as less sensitive than qRT-PCR especially when changes in expression are small . The observations that upstream components of the IIS pathway show transcriptional upregulation in response to Lnk loss of function suggest that transcriptional feedback back onto multiple components of the pathway may play an important regulatory role in IIS signal transduction . Previous studies have shown that dInR is itself a direct target of dFoxo so that when IIS levels are low , activated dFoxo increases dInR expression . In this study , we have shown that dFoxo also binds to the Lnk promoter in vivo suggesting that Lnk itself may be a direct target of dFoxo . dFoxo activity may also regulate transcription of IIS genes under basal conditions . Previous studies have shown that dFoxo is required for the basal expression of the dilp3 ligand [39] . In our study , we found that in the absence of dFoxo , Lnk transcript expression increases suggesting that dFoxo activity is normally required for Lnk repression . Thus , regulation by dFoxo may involve both positive and negative effects on gene expression . Our microarray data set also contained a number of differentially expressed genes that function within the Ras/MapK signal transduction pathway . Previous studies have shown that the Ras binding domain of Drosophila PI3K is required for maximal PI3K activity during growth and female egg laying linking Ras/MapK and IIS during growth and development in Drosophila [40] . Furthermore , we have shown that RNAi-mediated knockdown of Lnk inhibits insulin-stimulated Erk phosphorylation in insect cells . We cannot exclude the possibility that Lnk may play an adaptor function for Ras signaling downstream of other RTKs in addition to the insulin receptor . However , it should be noted that Lnk RNAi knockdown has no effect on Spitz-stimulated Erk phosphorylation via activation of the Drosophila EGF receptor [41] . Despite their small body size , Lnk mutants contain elevated levels of both lipid and carbohydrate stores . Consistent with their increased metabolic stores , Lnk mutants also showed increased survival under starvation conditions . Transcriptome analysis revealed gene expression changes in a number of components of metabolic regulation in Lnk mutants compared to controls . Thus , we observed reduced expression of several enzymes that function in the glycolytic pathway and upregulation of genes that function in glycogen synthesis . In addition , several genes in the mitochondrial β-oxidation pathway were downregulated whereas genes involved in the regulation of lipid storage showed increased expression . Taken together , these changes in gene expression are consistent with an overall inhibition of catabolic processes and upregulation of pathways that regulate the synthesis and storage of carbohydrates and lipids . Studies on the metabolic defects of SH2B knockouts in mice have proved inconsistent . One group has shown that genetic deletion of SH2B1 impairs adipogenesis by downregulating adipogenic gene expression including PPARγ resulting in mice with decreased fat mass [26] . A Drosophila PPAR homolog has yet to identified but the closest Drosophila relative is the orphan receptor , E75 [42] . This gene was not among the differentially expressed gene list from our microarray data . Other studies have shown that SH2B1 null mice actually increase their body mass and develop obesity as a result of hyperphagia [8] , [9] . In mammals , feeding is regulated by hypothalmic leptin signaling . Binding of leptin to its receptor results in receptor activation which in turn interacts with the non-receptor Janus kinase ( Jak ) stimulating downstream signaling events . Leptin stimulation of Jak is strongly potentiated by SH2B1 binding and so SH2B1 deletion impairs leptin signaling via Jak [4] , [43] , [44] . We did not observe any obvious differences in the feeding behaviour of Lnk mutant flies and there is no evidence to date that a leptin-like hormone exists in Drosophila . A functional Jak has been identified encoded by the hopscotch ( hop ) gene that has a well characterised role in hematopoesis in flies . We did not observe any obvious hematopoetic defects in Lnk mutants and Lnk was not found to genetically interact with any of the core JAK/STAT pathway components ( data not shown ) . Our data therefore suggests that the increased adiposity in Lnk mutant flies is unlikely to be mediated by increased feeding or by defects in Jak signaling . In fact , our data suggest that the ancestral function of Lnk in Drosophila is to regulate carbohydrate and fat storage by regulating gene expression of several key metabolic regulatory pathways . In mammalian cells , SH2B proteins have been shown to have dual functions during insulin signaling transduction by both activating and inhibiting downstream intracellular signaling events . Phosphorylation of SH2B2 by the activated insulin receptor creates a binding site for the proto-oncogene product c-Cbl . This promotes the ubiquitination of tyrosine kinase receptors by functioning as a RING-type E2-dependent ubiquitin protein ligase facilitating either endocytosis or proteasomal degradation of the receptor [16] , [20] . The c-Cbl binding motif is conserved in Drosophila Lnk and so it will be of interest to determine whether the interaction with c-Cbl is important for Lnk function especially during lifespan regulation .
w1118 and Lnkd07478 were obtained from the Bloomington Drosophila Stock Centre . yw , dfoxo21a/TM3 and dfoxo25c/TM3 were a gift from the Hafen lab [45] . The dfoxo21a and dfoxo25c alleles were backcrossed for at least 6 generations into the yw background before use . The LnkDel29 deletion was generated using the pBAC insertions Lnke01414 and Lnkf02642 ( obtained from the Exelixis Collection at Harvard Medical School ) according to published protocols [46] . A 6 kb fragment spanning from the 3′ end of CG17370 to the beginning of the first exon of CG5913 was used for the genomic rescue construct . This was inserted by means of ΦC31 mediated integration into a landing site on the second chromosome at 51D [28] . The wild-type stock Dahomey was collected in 1970 in Dahomey ( now Benin ) and has since been maintained in large population cages with overlapping generations on a 12L∶12D cycle at 25°C . The white Dahomey ( wDah ) stock was derived by incorporation of the w1118 deletion into the outbred Dahomey background by successive backcrossing . Both w1118 and wDah stocks were negative for the endosymbiont Wolbachia as determined by PCR using primers specific to Wolbachia genomic DNA . Lnk mutants were backcrossed for at least 8 generations into both w1118 and wDah genetic backgrounds before phenotypic analyses . Stocks were maintained and all experiments were conducted at 25°C on a 12h∶12h light:dark cycle at constant humidity using standard sugar/yeast/agar ( SYA ) medium [47] . For all experiments including lifespan experiments flies were reared at standard larval density and eclosing adults were collected over a 12 hour period . Flies were mated for 48 hours before sorting into single sexes . Body weights of individual male and female 7-day old flies ( n = 10 for each genotype ) were measured using a precision balance . Wing areas , cell numbers and cell sizes were measured as previously described [48] . For female fecundity tests , female flies were housed with males for 48 hours post-eclosion and then separated into vials at a density of 5 or 10 females per vial . Eggs were collected over two 24-hour periods per week for 4 weeks . The number of eggs laid per vial at each time point was counted . For male fertility tests , individual 3 day old males were mated to 30 virgin females , 3 to 5 days of age . Matings were observed and then females were separated from the males and housed in vials at a density of 3 females per vial . Eggs were collected and counted over four consecutive 24 hour periods . For lifespan experiments , flies were maintained in vials at a density of 10 flies per vial on standard SYA medium . Flies were transferred to new vials three times per week . For all stress assays , flies were reared and housed as for lifespan experiments . For oxidative stress assays , 4-day old flies were first starved for 5 hours on 1% agar and then transferred onto 5% sucrose/agar containing 5% hydrogen peroxide . For starvation experiments , 7-day old flies were transferred to 1% agar . Feeding rates of flies were measured using a proboscis-extension assay in undisturbed conditions as previously described [49] using 7-day-old mated flies . Flies were housed at a density of 5 flies of the same sex per vial and transferred to new food on the evening before the assay . Feeding data is expressed as a proportion by experimental group ( sum of scored feeding events divided by total number of feeding opportunities , where total number of feeding opportunities = number of flies in vial×number of vials in the group×number of observations ) . For statistical analyses , comparisons between experimental groups were made on the totals of feeding events by all flies within a vial , to avoid pseudoreplication . Drosophila S2 cell culture , dsRNA treatment and insulin treatment were as described in [37] . For western blots , 40 µg of total protein were resolved on 10% Tris-Glycine-SDS . Proteins were transferred to PVDF membranes and probed for total Akt ( 1∶1000; Cell Signaling ) , phospho-Akt ( 1∶1000; Cell Signaling ) , Erk ( 1∶1000; Cell Signaling ) , phospho-Erk ( 1∶1000; Cell Signaling ) and tubulin ( 1∶5000; Sigma ) . Secondary antibodies conjugated to HRP were purchased from Biorad . Hemolymph was collected and pooled from either 5 third instar larvae or 12 3-day old adult female flies . Glucose and trehalose levels were measured using the Glucose Infinity Reagent ( ThermoScientific ) as described in [33] . Whole fly trehalose in 7-day old adult males was measured as described in [33] and normalised to body weight . Glycogen content of 7-day old adult males was measured as described in [33] and normalised to body weight . Levels of TAG in 7-day old adult males were measured using the Tryglyceride Infinity Reagent ( ThermoScientific ) and also normalised to body weight . Total RNA was extracted from 10 whole adult flies , 10 adult bodies or 25 adult heads per genotype using standard Trizol ( Invitrogen ) protocols . cDNA was prepared using oligod ( T ) primer and Superscript II reverse transcriptase according to the manufacturer's protocol ( Invitrogen ) . Quantitative RT-PCR was performed using the PRISM 7000 sequence-detection system and Power SYBR® Green PCR Master Mix ( ABI ) . Relative quantities of transcripts were determined using the relative standard curve method and normalized to actin5C . Three or four independent RNA extractions were used for each genotype . Primer sequences are available upon request . Whole organism microarray experiments are generally only useful for detecting concerted changes of expression of widely expressed genes and most tissues will be under-represented in the array signal from a whole fly . Further complications arise from whole organism arrays when there are significant structural differences between treatments . Lnk transcripts are widely expressed but are particularly enriched within the central nervous system and as Lnk mutant ovaries show significant structural differences compared to controls , we restricted our microarray expression analysis to isolated heads . Raw data ( cel files ) were processed to correct for probe-sequence biases , and R's implementation of the Affymetrix's MicroArray Suite 5 . 0 software was used to determine present target transcripts [50] . A transcript was considered present if the p-value was <0 . 111 , and absent otherwise . The data was normalized using loess normalization and a linear model was fitted to identify a set of differentially expressed genes using the R limma package [51] . All individual probes have been mapped against all known and predicted transcripts of the Drosophila melanogaster genome release version 5 . 4 . Promiscuous ( some or all probes within a probe set map to more than one gene in the genome ) and orphan ( no probes in the probe set map to any known or predicted gene in the genome ) probe sets were excluded from further analysis . FlyBase gene ids were mapped to Gene Ontology ( GO ) ids ( version 1 . 107 ) . For functional analysis using all expressed genes , we used the Wilcoxon rank sum test implemented in Catmap [52] . Ranks of genes were based on the Bayes t-statistic for differential expression and , for a given functional category , the significance of the rank sum for all genes in the category was calculated analytically based on a random gene-rank distribution . Sequence analysis was performed using Regulatory Sequence Analysis Tools [53] looking for perfect matches to the mouse Foxo1/Foxo4 binding sites [RWWAACA] [54] . Chromatin immunoprecipitations were carried out essentially as described by [55] . For starvation and paraquat treatments , flies were either starved for 24 hours or fed 20 mM paraquat for 16 hours . 1000 adult female flies were crushed to a fine powder under liquid nitrogen and suspended in 6 ml of PBS supplemented with Protease Inhibitor Cocktail ( Sigma ) . Cross-linking was performed with 0 . 5% formaldehyde for 10 minutes and quenched by the addition of 1 . 5 ml of 2 . 5M glycine . The cross-linked chromatin was recovered by centrifugation and washed twice with FA/SDS buffer ( 50 mM Hepes-KOH , 150 mM NaCl , 1 mM EDTA , 0 . 1% Na Deoxycholate , 0 . 1% SDS , 1% Triton-X100 and 1 mM PMSF ) . Samples were resuspended in FA/SDS buffer and incubated for 1 hour at 4C . Chromatin was recovered by centrifugation and sheared to an average size of 400 bp by sonication , giving on average 6 ml of chromatin in FA/SDS . For immunoprecipitations ( IPs ) , 1 µl of affinity purified rabbit anti-Foxo antibody [36] was bound to Protein-G Dynabeads ( Invitrogen ) and incubated with 450 µl of chromatin for 2 hours at room temperature . Beads were washed three times with FA/SDS , once with TE , and once with 10 mM Tris-HCl pH 8 , 250 mM LiCl , 1 mM EDTA , 1% NP40 , 0 . 5% Na Deoxycholate . DNA was recovered , treated with proteases , de-cross-linked , treated with RNase and purified using the Qiagen PCR purification kit ( Qiagen ) . For quantitative PCR , a suitable dilution of total chromatin and IP was used for quantification using the PRISM 7000 sequence-detection system and Power SYBR® Green PCR Master Mix ( ABI ) . For ChIP analysis , relative amounts of the target DNA recovered after ChIP compared to total chromatin were determined using three independent biological replicates . The relative proportion of DNA binding was calculated by dividing the proportion of DNA binding in the ChIP for a single region by the average recovered for all regions for that chromatin to normalise for plate-plate differences . Statistical analyses were performed using JMP software ( version 4 . 0 . 5; SAS Institute ) . Log rank tests were performed on lifespan and stress survival curves . Other data were tested for normality using the Shapiro-Wilk W test on studentised residuals and where appropriate log-transformed . One-way analyses of variance ( ANOVA ) and planned comparisons of means were made using Tukey-Kramer HSD test . | Many human populations are experiencing increased life expectancy , and as populations age the incidence of age-related diseases becomes more prevalent . The identification of single gene mutations that extend lifespan in invertebrate model organisms has revealed that several cellular signaling pathways , including the insulin/insulin-like growth factor ( IGF ) -1 signaling ( IIS ) pathway , play a crucial role in modulating the ageing process across multiple species . Thus , studies carried out in yeast , worms , and flies have revealed evolutionarily conserved mechanisms of ageing , which are likely to be relevant to mammals , including humans . A recent study in Drosophila identified the SH2B family adaptor protein , Lnk , as an important regulator of the IIS pathway during organismal growth . In this study , we show that Lnk is also required to determine normal lifespan in Drosophila , as mutations that disrupt Lnk activity result in increased lifespan . In addition , these mutants show improved survival under conditions of stress and metabolic disregulation . Furthermore , we show that the expression of Lnk is regulated by the IIS responsive transcription factor , dFoxo . Our data therefore provide new mechanistic insights into the role of the IIS pathway in ageing . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"developmental",
"biology/aging"
]
| 2010 | Regulation of Lifespan, Metabolism, and Stress Responses by the Drosophila SH2B Protein, Lnk |
In most eukaryotes , including the majority of fungi , expression of sterol biosynthesis genes is regulated by Sterol-Regulatory Element Binding Proteins ( SREBPs ) , which are basic helix-loop-helix transcription activators . However , in yeasts such as Saccharomyces cerevisiae and Candida albicans sterol synthesis is instead regulated by Upc2 , an unrelated transcription factor with a Gal4-type zinc finger . The SREBPs in S . cerevisiae ( Hms1 ) and C . albicans ( Cph2 ) have lost a domain , are not major regulators of sterol synthesis , and instead regulate filamentous growth . We report here that rewiring of the sterol regulon , with Upc2 taking over from SREBP , likely occurred in the common ancestor of all Saccharomycotina . Yarrowia lipolytica , a deep-branching species , is the only genome known to contain intact and full-length orthologs of both SREBP ( Sre1 ) and Upc2 . Deleting YlUPC2 , but not YlSRE1 , confers susceptibility to azole drugs . Sterol levels are significantly reduced in the YlUPC2 deletion . RNA-seq analysis shows that hypoxic regulation of sterol synthesis genes in Y . lipolytica is predominantly mediated by Upc2 . However , YlSre1 still retains a role in hypoxic regulation; growth of Y . lipolytica in hypoxic conditions is reduced in a Ylupc2 deletion and is abolished in a Ylsre1/Ylupc2 double deletion , and YlSre1 regulates sterol gene expression during hypoxia adaptation . We show that YlSRE1 , and to a lesser extent YlUPC2 , are required for switching from yeast to filamentous growth in hypoxia . Sre1 appears to have an ancestral role in the regulation of filamentation , which became decoupled from its role in sterol gene regulation by the arrival of Upc2 in the Saccharomycotina .
Changes in gene regulatory networks are an important mechanism of evolutionary adaptation . Transcriptional re-wiring can result from gene loss , gene duplication , alterations in transcription factor binding sites , or changes in protein modularity that affect the interaction of transcription factors with other regulators [1] , [2] , [3] , [4] , [5] . A hybrid ancestral state may be resolved in different ways in different lineages , such as occurred in the regulation of cell type specific genes in the Saccharomycotina yeasts [1] , [6] and the substitution of transcription factors regulating of ribosomal protein genes in Ascomycota fungi [2] . Other examples include substitution of the transcription factor Cph1 with Gal4 for regulation of galactose metabolism genes in the Saccharomyces clade [7] , and changes in telomere binding proteins [8] . Large-scale analysis of promoter motifs and transcription factor conservation suggests that re-wiring of networks may be relatively common in eukaryotes [9] . However , most reported evolutionary changes involve the connection or disconnection of a group of target genes from a particular transcription factor , while the cellular function of the factor remains the same [9] . We describe here a major transcriptional re-wiring event that occurred in the evolution of sterol synthesis ( an oxygen-dependent process ) in eukaryotes . Regulation of sterol synthesis by Sterol Regulatory-element Binding Proteins ( SREBPs ) is very well conserved between metazoa and most fungi , but this conserved system has been disrupted in the clade of yeasts that includes S . cerevisiae . In these yeasts the role of SREBPs in sterol synthesis has been replaced by Upc2 ( reviewed in [10] , [11] , [12] ) . SREBPs regulate cholesterol synthesis and uptake , and fatty-acid synthesis in mammalian cells [13] . They are a family of transcription factors with a bHLH ( basic Helix-loop-Helix ) domain with a characteristic tyrosine residue . When sterol levels are high , SREBPs interact with the sterol-sensing protein Scap ( SREBP cleavage-activating protein ) and the complex is retained at the endoplasmic reticulum ( ER ) through association with INSIG ( insulin-induced protein ) [14] . When sterol levels drop , cholesterol no longer binds to Scap , disrupting the interaction with INSIG and resulting in transport of the SREBP-Scap complex to the Golgi apparatus [15] . Here , two proteases ( site-1 protease and site-2 protease ) cleave SREBP firstly in the loop within the Golgi lumen , and secondly to release the N-terminal domain . The N-terminus of SREBP enters the nucleus where it acts to regulate gene expression ( reviewed in [10] , [16] ) . SREBPs are well conserved in many fungi and have been shown to regulate sterol synthesis in several species , particularly in response to low oxygen [10] . Not all components of the pathway are conserved . In Basidiomycetes ( such as Cryptococcus neoformans ) and in some Ascomycetes ( e . g . Schizosaccharomyces pombe ) SREBPs interact with Scap proteins , but the INSIG homolog appears to play no role in retention in the ER membrane [17] . The N-terminal region of SREBP is released by a single cleavage reaction . In C . neoformans , this cleavage is carried out by a homolog of the mammalian site-2 protease [18] , [19] . In the Ascomycetes ( Sch . pombe and Aspergillus fumigatus ) , processing does not require the site-1/site-2 proteases but instead uses the Dsc E3 ligase complex and the proteasome [20] , [21] , [22] . Some Ascomycete lineages such as Eurotiomycetes ( including A . fumigatus ) have lost Scap , and it is not clear what their sterol-sensing mechanism is [10] , [11] . At least some fungal SREBPs are also regulated by oxygen levels independently of sterol levels . In Sch . pombe , oxygen-dependent degradation of the N-terminus of the SREBP protein ( Sre1N ) is regulated by Ofd1 , a member of the prolyl hydroxylase family , and by Nro1 , a nuclear protein [23] , [24] . Ofd1 also regulates binding of Sre1N to sterol regulatory elements ( SRE ) [25] . Surprisingly , within the Saccharomycotina subphylum of the Ascomycetes , SREBP proteins appear to play little or no role in regulating oxygen-dependent expression of sterol biosynthesis genes . Some SREBP-like proteins , with the characteristic tyrosine in the bHLH domain , are present in these species , for example C . albicans Cph2 and S . cerevisiae Hms1 [10] . However , these proteins are often considerably shorter than their homologs in other Ascomycetes , and no role for them in sterol gene regulation has been demonstrated [26] . Instead , sterol gene expression in Saccharomycotina is controlled by Upc2 proteins , which have Gal4-type Zn2-Cys6 zinc finger domains , and are structurally unrelated to SREBPs . Members of the Upc2 family regulate expression of sterol and other hypoxic genes in S . cerevisiae , Candida glabrata , C . albicans and C . parapsilosis [27] , [28] , [29] , [30] , [31] . We observed that Yarrowia lipolytica , representing the most divergent known lineage of the Saccharomycotina , is unique among fungi in having a genome that contains readily identifiable genes for both SREBP and Upc2 . It may thus be a ‘molecular fossil’ of a transition stage during the handover of control of sterol metabolism from SREBP to Upc2 in the Saccharomycotina . We show here that in Y . lipolytica , both Upc2 and SREBP play a role in responding to hypoxic conditions . However , Upc2 is the main regulator of sterol genes . SREBP ( Sre1 ) , and to a lesser extent Upc2 , regulates filamentous growth . Our analysis suggests that re-wiring of the ergosterol synthesis module occurred in an early ancestor of the Saccharomycotina , when regulation was ceded to Upc2 .
The Saccharomycotina is a lineage within the Ascomycetes that includes the Saccharomycetaceae and CTG clades ( Figure 1A ) . The CTG clade contains Candida species in which the codon CTG is translated as serine rather than leucine [32] , [33] . Y . lipolytica , a lipid degrading yeast [34] , lies at the very base of the Saccharomycotina . Y . lipolytica has a gene ( YALI0D15334 ) that is similar to Sre1 from Sch . pombe ( maximum 30% identity ) and SrbA from A . fumigatus ( maximum 44% ) . One of the conserved regions corresponds to a bHLH domain near the N terminus , with a tyrosine rather than an arginine in the basic domain [35] ( Figure 1A , B ) . The arginine-to-tyrosine substitution changes the DNA sequence that is bound , from the standard E-box recognized by most bHLH proteins to sequence called a sterol regulatory element-1 ( SRE-1 ) [36] . The similarity to the SREBPs extends beyond the bHLH region . The Y . lipolytica protein also has a domain ( DUF2014; Figure 1A ) towards its C terminus that is shared with Sch . pombe Sre1 and A . fumigatus SrbA and other filamentous Ascomycetes . The function of this domain is unknown . It may be important for interaction with Scap and therefore for retention of SREBP in the membrane . However , SREBP proteins of Aspergillus species , which have lost Scap [10] , retain the DUF2014 domain . The DUF2014 domain appears to have been acquired by the SREBPs in the ancestor of the Ascomycota , as it is not found in Sre1 of C . neoformans and other Basidiomycetes . Most fungi have only one family of SREBP-like proteins , but Saccharomyces and Candida species have homologs of two potential SREBP-like protein families , both of which have bHLH domains with the characteristic tyrosine residue . One family ( pink region in Figure 1C ) is represented by Tye7 in S . cerevisiae and C . albicans . This family is restricted to the Saccharomycotina and is composed of short proteins ( 218–385 amino acids ) without a DUF2014 domain . Tye7 regulates expression of glycolytic genes in S . cerevisiae and C . albicans [37] , [38] , [39] , [40] . The second family ( orange region in Figure 1C ) , is represented by Cph2 in C . albicans and Hms1 in S . cerevisiae . This family contains longer proteins with higher similarity to Sch . pombe Sre1 and Sre2 , and has bHLH regions more similar to other SREBPs ( Figure 1B , C ) . However , there is little conservation in the C-terminal regions between Hms1/Cph2 and the other fungal SREBPs , and the DUF2014 domain is not generally present , although in Meyerozyma guilliermondii ( within the Candida clade ) a remnant is recognizable ( Pfam [41] E-value of 3 . 5e-4 , compared to 3 . 4e-80 for the Y . lipolytica protein ) . Hms1 orthologs in S . cerevisiae and other species in the post-Whole Genome Duplication clade [42] are much smaller , and as a result the bHLH domain is closer to the C terminus ( Figure 1A ) . The structure of the Hms1/Cph2 proteins including the apparent historical DUF2014 domain in M . guilliermondii , and their phylogenetic closeness to Sch . pombe Sre1 and the Pezizomycotina SREBPs , suggests that they ( rather than Tye7 ) are the Saccharomycotina orthologs of SREBP . However , they have undergone substantial modification , including the degeneration of the DUF2014 domain . The Y . lipolytica protein , which we have named Sre1 , is the only one within the Saccharomycotina species that retains an intact DUF2014 domain . It clearly falls within the Cph2/Hms1 clade ( Figure 1C ) . In C . neoformans , Sch . pombe and A . fumigatus , oxygen regulates the cleavage and localization of the SREBP transcription factor domain , releasing it from the ER and Golgi membrane and facilitating entry into the nucleus [18] , [19] , [20] , [21] , [22] . The C . neoformans and Sch . pombe proteins have two predicted membrane spanning domains , suggesting that both C and N termini are facing into the cytoplasm ( [43] , Figure 1A , Figure S1 ) . SrbA from A . fumigatus is predicted to contain at least two ( and possibly up to four ) transmembrane domains , indicating that the protein is localized to membrane structures , with at least the N terminus facing the cytoplasm ( Figure S1 ) . Within the Saccharomycotina however there is very little evidence that the SREBP proteins are localized to membranes . For Sre1 from Y . lipolytica and Cph2 from C . albicans , there is <10% probability of one or two transmembrane domains respectively , and there is no indication of any transmembrane domain in the S . cerevisiae Hms1 protein . In contrast to SREBP , orthologs of Upc2 are only clearly identifiable within the Saccharomycotina ( Figure 1A; Figure 2 ) . In fact , in the post-Whole Genome Duplication clade , two paralogs ( known as Upc2 and Ecm22 [30] ) have been retained in all species ( yellow clade in Figure 2 ) . The Upc2 and Ecm22 proteins contain a fungal Zn2-Cys6 binuclear cluster domain , and a domain associated with fungal-specific transcription factors . The Upc2/Ecm22 proteins from the Saccharomycotina form a monophyletic clade that is not closely related to any other Zn2-Cys6 proteins of the Saccharomycotina ( such as Lys14 ) , or Ascomycota ( Figure 2 ) . The most likely interpretation of the phylogeny in Figure 2 is that Upc2 arose in the common ancestor of the Saccharomycotina , before the split between the Y . lipolytica lineage from the rest of the clade . Upc2 was presumably created by duplication of another zinc finger protein gene , but it has diverged to the point where its orthologs in species such as Aspergillus and Komagataella , if they exist , are unrecognizable ( Figure 2 ) . To determine the role of the Sre1 ( YALI0D15334 ) and Upc2 ( YALI0B15818 ) orthologs in Y . lipolytica we knocked out both genes in the W29 background ( Figure S2 ) . YlUPC2 was replaced with URA3 , and YlSRE1 with LEU2 , in Y . lipolytica Po1d ( leu2–270 , ura3–302 [44] ) using fusion PCR [45] and previously described transformation methods [46] . The remaining markers ( LEU2 or URA3 ) were re-introduced into all strains , to reconstitute prototrophy ( Table S1 ) . YlSRE1 and YlUPC2 were also reintroduced at the endogenous locus by insertion of a cassette containing the relevant open reading frame plus 800 bp of the upstream region and a hygromycin resistance marker ( HygEx ) . Deleting YlSRE1 has a minimal effect on growth on rich media , whereas deleting YlUPC2 reduces growth further , and the double deletion has a pronounced growth defect ( Figure 3 , growth curves are shown in Figure S3 ) . Deleting YlUPC2 dramatically increases the susceptibility of Y . lipolytica to ketoconazole , whereas deleting YlSRE1 has no obvious effect ( Figure 3A ) . The level of drug required to inhibit growth is much higher when cells are grown on synthetic complete ( SC ) media than when grown on rich media ( YPD ) ; the reason for the difference is not known , but deleting YUPC2 has the same effect on both media . The susceptibility phenotype is similar to that observed when UPC2 is deleted in S . cerevisiae [47] , C . albicans [48] , [49] and C . parapsilosis [27] , and when SrbA is deleted in A . fumigatus [50] and SRE1 in C . neoformans [19] . Azole drugs target the ergosterol pathway in fungi , and in particular the product of the ERG11 gene , which encodes Lanosterol 14-alpha-demethylase . We therefore measured the level of sterols in the various genetic backgrounds ( Figure 3B ) . Cells were grown in defined minimal media , and sterols were extracted using an alcoholic KOH solution and heptane [49] . Figure 3B shows that deletingYlUPC2 reduces absorbance at wavelengths that are indicative of lower sterol content , which are restored when the YlUPC2 gene is re-introduced . Two independent deletions of YlSRE1 had no reduction in sterol levels . There appears to be a slight additional reduction in the double deletion background relative to the Ylupc2 deletion , but this is not statistically significant . Deleting YlUPC2 greatly reduces growth in hypoxic ( 1% O2 ) compared to normoxic conditions , during growth on both synthetic complete ( SC ) media containing methionine or rich ( YPD ) media ( Figure 3A ) . Deleting YlSRE1 also reduces hypoxic growth , which is more pronounced on defined media . The strain carrying deletions of both YlUPC2 and YlSRE1 grows poorly on YPD plates in normoxia , and fails to grow at all in hypoxic conditions . Adding fatty acids to SC media ( in the form of Tween 80 ) improves growth of all strains , but the effect of deleting YlUPC2 and YLSRE1 is still evident ( Figure 3 ) . Addition of ergosterol does not rescue the phenotype any further , though this is possibly because Y . lipolytica cannot import sterols either aerobically or anaerobically ( Figure S4 ) . Reintroducing YlUPC2 and YlSRE1 in the single deletion strains restores growth in hypoxia ( Figure 3 ) . In Aspergillus fumigatus , SrbA regulates expression of iron uptake genes , as well as of ergosterol synthesis [51] . We therefore tested the effect of deleting YlUPC2 and YlSRE1 on growth in low iron conditions . Figure 3C shows that when iron levels are depleted by adding the iron chelator BPS ( 4 , 7-diphenyl-1 , 10-phenanthrolinedisulfonic acid ) , the Ylupc2 deletion strain fails to grow , whereas deleting YlSRE1 has no effect . The phenotype is rescued by adding additional exogenous iron . In Y . lipolytica , UPC2 therefore regulates both iron acquisition and sterol metabolism . Y . lipolytica and C . albicans are unusual among the Saccharomycotina species in that they can switch from growth as yeast cells to fully filamentous ( hyphal ) growth in certain conditions [52] , [53] . Other species grow as yeast and pseudohyphae , or are locked in the filamentous form [54] . Y . lipolytica is truly dimorphic [34] , [55] . Hyphae are induced by altering carbon source or pH , or by growing in hypoxic conditions [55] , [56] [57] . We determined the effect of deleting YlUPC2 and YlSRE1 on hypoxia-induced filamentation of cells growing in rich ( YPD ) or minimal ( SC ) media , in both solid and liquid conditions ( Figure 4 ) . During growth in liquid YPD in normoxia , the strains are predominantly yeast-like , irrespective of the genetic background . Very few short filaments are formed . During growth in hypoxic conditions , the wild type and reconstituted strains are hyperfilamentous . The Ylupc2 deletion also produces some long filaments . However , the Ylsre1 deletion generates only very short filaments . Adding fatty acids ( Tween 80 ) recovers the hypoxia-induced filamentation phenotype in the Ylsre1 background , and improves filamentation of the Ylupc2 deletion . There is very little difference in cell morphology in cells growing in liquid SC in normoxic or hypoxic conditions . The wild type and reconstituted strains are filamentous , whereas the Ylupc2 and Ylsre1 deletions are predominantly yeast-like , with some short filaments found in the Ylsre1 background . Adding Tween 80 rescues the induction of filamentation of Ylupc2 and Ylsre1 in hypoxic conditions , and partially increases filamentation of the Ylsre1 deletion even in normoxia . It has been reported that filamentation levels are higher on solid rather than liquid media [58] . On solid YPD media , we found that the wild type and the Ylupc2 deletion strains are filamentous in both normoxic and hypoxic conditions . However , the sre1 deletion fails to filament , even in hypoxia . On solid SC media , hypoxia induces filamentation of the wild type and the Ylupc2 deletion , but not the Ylsre1 deletion . Overall , our results show that YlSre1 is required for hypoxic-induced filamentation in all conditions and media tested , and YlUpc2 is required in most conditions . The double deletion remains in the yeast morphology , and fails to grow at all in hypoxia , suggesting that the two transcription factors act synergistically . To determine the roles of YlUpc2 and YlSre1 in regulating the hypoxic response , we first characterized the transcriptional profile of Y . lipolytica during growth in low oxygen . Y . lipolytica can tolerate oxygen levels as low as 1% ( Figure 3 ) , but it is incapable of anaerobic growth [52] . We used strand-specific RNA-seq to compare the transcriptional profile of cells grown in YPD in atmospheric oxygen levels and at 1% O2 . Differentially expressed genes were identified using DESeq [59] . Approximately 1 , 900 genes are differentially expressed in low oxygen , corresponding to 30% of the genome ( Table S2 ) . This corresponds well with the response of S . cerevisiae , where >2 , 000 genes have altered expression in anaerobic conditions [60] . We used DAVID ( Database for Annotation , Visualization and Integrated Discovery ) to identify enrichment of specific categories among the differentially expressed genes [61] . DAVID applies several categorization tools , incorporating annotation categories that include Gene Ontology assignments , KEGG metabolic pathways , and InterPro database of protein families and domains . Genes upregulated in hypoxia are enriched for categories associated with membrane structure , ion binding and oxidoreductase activity ( Table 1 ) . Notably , transcription factors are also significantly over-represented in genes upregulated in hypoxia ( Table 1 , Table S6 ) . Expression of 78 genes with potential transcription factor activity is upregulated , including both YlUPC2 and YlSRE1 ( Table S6 ) . Expression of HOY1 , a homeobox gene required for hyphal development in Y . lipolytica [58] , [62] , is among the genes with the highest fold induction . Interestingly , the transcription factor with the greatest increase in expression in hypoxia ( YALI0C03564g ) encodes a protein of unknown function , with a bHLH domain ( Table S3 ) . However , this domain does not contain the atypical Tyr residue , and probably binds to an E-box sequence rather than an SRE-1 element . Downregulated genes are enriched in processes including ribosome biogenesis , rRNA processing , translation and microtubule function ( Table 1 , Table S6 ) . These changes reflect the fact that the strains are growing slowly in hypoxic conditions ( Figure 3 , [63] ) . To compare the effects of deleting YlUPC2 and YLSRE1 on the transcriptional profile of Y . lipolytica we analyzed the RNA-seq data using gene enrichment analysis implemented in DAVID [61] , and by hierarchical clustering ( Table 1 , Figure 5 , Figure S7 ) . The overall transcriptional response of the deletion strains to hypoxia is very similar to the response of the wild type ( Figure 5A , B ) . More than 1200 genes are differentially expressed in hypoxia in all three backgrounds ( Figure 5B ) . This suggests that there are many other transcription factors apart from UPC2 and SRE1 that regulate the hypoxic response , supporting our analysis of hypoxic induction in wild type cells ( Table 1 ) . Gene enrichment analysis shows that deleting Ylupc2 results in lowered expression of steroid metabolism genes , even when the strains are grown in normoxic conditions ( Table 1 , Table S6 ) . The effect of YlUPC2 on sterol metabolism during hypoxic growth is also obvious from the hierarchical cluster analysis ( Figure 5A ) . Expression of one cluster of 14 genes is notably reduced in the Ylupc2 deletion , while remaining upregulated in the Ylsre1 strain grown in hypoxia ( Figure 5A ) . This group includes four genes required for ergosterol biosynthesis , all of which function in the oxygen-dependent part of the pathway . Two are paralogs of ERG2 ( C-8 sterol isomerase ) , one of which we have designated as ERG2-2 . Expression of both is greatly reduced in the Ylupc2 deletion , but not in the Ylsre1 background . Most of the remaining genes in the cluster have roles in redox reactions , such as formate dehydrogenase , superoxide dismutase and glutathione-S-transferase . Although not all of the ergosterol metabolism genes fall in the same cluster shown in Figure 5A , many are highly expressed in hypoxic conditions in wild type cells , and expression is greatly reduced ( or abolished ) in a Ylupc2 deletion ( Figure 5C ) . To determine if YlUpc2 is likely to be a direct regulator of ERG genes , we looked for evidence of enrichment of potential binding sites in the upstream promoters . We found that the Upc2 motif defined in S . cerevisiae is enriched in the promoter regions of ergosterol genes in all Saccharomycotina species , including Y . lipolytica ( Table 2 , genes shown in Figure 5C ) . In contrast , there is no enrichment in the equivalent promoters of Sch . pombe or A . fumigatus , species in which ergosterol genes are regulated by SREBPs [10] . Potential binding motifs were identified in 16 of the 21 promoters tested . ERG2 and ERG2-2 , the two genes with the strongest reduction in expression in the Ylupc2 deletion ( Figure 5A , Figure 6 ) , have four potential binding sites each . Genes with reduced expression in the Ylupc2 deletion relative to wild type in hypoxic conditions are also enriched for processes associated with cell redox , antioxidant activity and glutathione-S-transferase ( Table 1 , Table S6 ) . YlUpc2 may therefore be involved in protection from oxidative stress . Many of the other downregulated genes are enriched for processes associated with translation , such as ribosome biogenesis and rRNA processes ( Table 1 , Table S6 ) . This most likely reflects the fact that deleting Ylupc2 further reduces growth in hypoxia ( Figure 3 ) . Sre1 does not play a major role in regulating expression of sterol genes during long term growth in hypoxia; expression of the ERG genes in the Ylsre1 deletion is very similar to that of the wild type cells grown in hypoxic conditions ( Figure 5A , B ) , and sterol levels are not reduced in a Ylsre1 deletion ( Figure 3C ) . Deleting YlSRE1 results in increased expression of lipid metabolism and fatty acid metabolism genes ( Table 1 , Table S6 ) . However , unlike in mammalian cells , Drosophila , and Sch . pombe , the potential targets are mostly associated with lipid degradation , rather than biosynthesis ( Table S6 , [13] , [64] , [65] ) . The most highly enriched processes among downregulated genes compared to wild type are associated with ribosomal biogenesis and rRNA binding , which correlates with poor growth . Genes upregulated in the Ylsre1 deletion relative to wild type are mostly associated with proteasome-dependent proteolysis , which may also result from slow growth ( Table 2 ) . In the Ylupc2/Ylsre1 double deletion , growth is greatly diminished and most downregulated genes are associated with translation ( Table S6 ) . Enrichment categories of upregulated genes are very similar to the categories upregulated during hypoxic growth of wild type cells ( Table S6 ) . It was not possible to determine the effect of deleting both YlUPC2 and YlSRE1 on the transcriptional response to hypoxia , because the double deletion strain fails to grow in low oxygen conditions ( Figure 3 ) . For the RNA-seq experiments , cells were grown in rich media to minimize the effects of reduced cell growth , which is more pronounced in synthetic complete media ( Figure 3 ) . The strains were also grown for prolonged periods in hypoxic conditions . It is possible that the gene expression patterns would be different during growth in defined media , and during earlier stages of adaptation to low oxygen . We therefore used quantitative PCR to measure the expression of five genes in the ergosterol pathway in cells grown in synthetic complete media in high oxygen to mid log phase , which were shifted to hypoxic conditions for 2 hours . Figure 6 shows that YlUPC2 is required for maximum hypoxic induction of all five genes . In particular , expression of ERG2 and its paralog ERG2-2 is completely dependent on YlUPC2 , even in normoxic conditions . This pattern is similar to that observed in the RNA-seq experiments . However we find that YlSRE1 is required for maximal hypoxic induction of at least four genes ( ERG2 , ERG3 , ERG11 and ERG25 ) . It is therefore likely that YlSre1 plays a role in regulating expression of ergosterol genes at early stages of hypoxic adaptation .
Sterols are essential for maintaining membrane structure and function , and synthesis in fungi and other eukaryotes is very carefully regulated at several levels [66] . Our results indicate that Upc2 is the major regulator of expression of sterol synthesis genes in Y . lipolytica . Expression of many of the sterol genes is reduced in the Ylupc2 deletion , particularly during hypoxic growth , and the level of sterols in the cell is also reduced ( Figure 3 , Figure 5 ) . Promoter analysis also indicates that the Upc2 binding sites are enriched in the promoters of sterol synthesis genes in Y . lipolytica and other species of the Saccharomycotina , but not in the equivalent promoters in A . fumigatus or Sch . pombe ( Table 2 ) . It is therefore likely that Upc2 homologs are the main regulators of sterol synthesis in all Saccharomycotina species . This has been shown experimentally for C . albicans , C . parapsilosis , S . cerevisiae and C . glabrata [27] , [28] , [29] , [30] , [31] , [67] , and now for Y . lipolytica . Although the role of Upc2 is generally conserved , there are also substantial species-specific variations . Deleting Ylupc2 results in a growth defect , which has not been reported in other Saccharomycotina species [30] , [49] . YlUpc2 also regulates expression of ERG genes in normoxia ( in particular , expression of ERG2 and ERG2-2 , Figure 6 ) . In C . albicans , the role of Upc2 is generally only evident when ergosterol gene expression is induced with ketoconazole or growth in hypoxia , or with gain-of-function alleles of UPC2 [29] , [68] . In S . cerevisiae , Upc2/Ecm22 regulates expression of sterol synthesis and sterol uptake genes [30] , [69] . The Upc2 paralogs control sterol import in C . glabrata [70] , but not in C . albicans [31] . Y . lipolytica does not have an obvious ortholog of the AUS1/PDR11 sterol transporters from S . cerevisiae [69] , [71] nor of the regulator of sterol import , SUT1 [72] . Y . lipolytica also apparently does not import cholesterol ( and therefore probably ergosterol ) in aerobic or hypoxic conditions ( Figure S4 ) . The Upc2 proteins in the Saccharomycotina are under considerable evolutionary constraint ( indicated by short branch lengths in Figure 2 ) , and are relatively distant from even their closest Pezizomycotina counterparts . This supports our hypothesis that Upc2 appeared in , or was substantially modified in , the ancestor of the Saccharomycotina ( Figure 7 ) , and the function has been generally conserved since . In Y . lipolytica , Upc2 also regulates iron uptake , which is controlled by SREBPs in A . fumigatus [51] . It is currently unknown how Upc2 proteins sense oxygen . However , the mechanism seems to be different from SREBP type of proteins . Our recent unpublished data indicate that Upc2 in S . cerevisiae does not undergo proteolytic cleavage , both its C- and N-terminus localize to the nucleus upon activation and its DNA binding domain seems to lose transcriptional activity without the protein's C-terminus . In addition , introducing an HA-tag at the C terminus of C . albicans Upc2 results in a gain-of-function , which may result from altered activation rather than processing [68] . SREBPs are major regulators of sterol synthesis and of the hypoxic response in Basidiomycete fungi ( e . g . C . neoformans ) and in some Ascomycetes ( e . g . Sch . pombe and A . fumigatus , Figure 7 ) [10] , [11] , [12] . The SREBPs in the Ascomycetes gained a domain ( DUF2014 ) , whose function is currently unknown ( Figure 7 ) . DUF2014 is retained in the Y . lipolytica protein , but was lost in the most likely orthologs of Sre1 in the other Saccharomycotina species , including S . cerevisiae ( Hms1 ) and C . albicans ( Cph2 ) . Hms1 and Cph2 have no known role in hypoxic regulation . However , we find that in Y . lipolytica , deleting SRE1 reduces growth in hypoxic conditions ( Figure 3 ) . The defect in long-term hypoxic growth is unlikely to be due to regulation of sterol synthesis , because expression is not reduced in rich media ( Figure 5A ) , and sterol levels are not reduced in the Ylsre1 single deletion ( Figure 3B ) . However , Sre1 contributes to induction of ERG genes during hypoxic adaptation ( Figure 6 ) . This suggests that whereas Upc2 is the major regulator of sterol genes , both Sre1 and Upc2 act synergistically at ERG promoters is some conditions ( Figure 7 ) . YlSre1 also has a role in response to long-term hypoxic growth that is separate from that of sterol biosynthesis . The overall hypoxic response of Y . lipolytica is somewhat different to that of other fungi . Expression of fatty acid biosynthesis , drug transport and membrane proteins is increased , similar to that observed in fungi like S . cerevisiae and C . albicans ( reviewed in [11] ) . Expression of sterol metabolism is also increased; although we do not find enrichment of sterol metabolism genes in the RNA-seq analysis using DAVID , it is clear from hierarchical clustering and from qRT-PCR ( Figure 5 , Figure 6 ) . Unlike in S . cerevisiae and C . albicans , we did not observe changes in central carbon metabolism , such as upregulation of glycolysis and downregulation of the TCA cycle [29] , [73] . Similar changes do occur in some obligate aerobes like A . fumigatus [74] , but expression of glycolysis is reduced in others such as Trichoderma reesii [75] . In the aerobe C . neoformans , respiration is increased in low oxygen [19] . It is possible that we would observe different patterns in Y . lipolytica if we measured expression levels during short-term adaptation to hypoxia , or during growth in minimal media . SREBP-like proteins from both S . cerevisiae and C . albicans are involved in regulating cell morphology . Overexpressing HMS1 in S . cerevisiae results in hyperfilamentous growth [76] . The SREBP Cph2 from C . albicans was first isolated as a high copy inducer of pseudohyphal growth in S . cerevisiae , and it was subsequently shown that deleting cph2 in C . albicans reduced the ability to switch to hyphal growth on Lees medium [26] . Hyphal induction in other conditions ( such as growth following the addition of serum ) is not impaired . Cph2 acts through the TEA/ATTS transcription factor TEC1 to regulate filamentous growth . Several other pathways also regulate filamentation , including the Efg1-regulated cAMP-dependent protein kinase A ( PKA ) pathway , and the Cph1-mediated mitogen-activated protein kinase ( MAPK ) pathway ( reviewed in [77] ) . The different pathways converge on some of the same target genes [78] . In Y . lipolytica , the equivalent MAPK pathway regulates filamentous ( or mycelial ) growth , whereas the cAMP-dependent PKA pathway is required for yeast-like growth [79] , [80] , [81] , [82] . Filamentation is also regulated by Tec1; however , unlike C . albicans , Tec1 appears to promote yeast rather than hyphal growth [83] . We show that YlSre1 is required for hypoxia-induced filamentation , and that the filamentation phenotype is rescued by the addition of fatty acids ( Figure 4 ) . There are some differences in phenotype in liquid versus solid media and in rich versus minimal media , suggesting that there are several different signals inducing filamentous growth . However , deleting YlSRE1 reduces filamentation in most conditions . Our results suggest that the bHLH Sre1-like proteins are ancient regulators of morphology , and indeed of filamentous growth , in the Saccharomycotina . Family members regulate cell morphology in at least three species – S . cerevisiae , C . albicans , and in the basal species Y . lipolytica . The role may be even older , as the Sre1 homolog in A . fumigatus ( SrbA ) is also required for cell polarity and hyphal branching [50] . We do not yet know how Sre1 regulates filamentation , though it may be linked to regulation of fatty acid synthesis . However , we note that one of the few known regulators of filamentation in Y . lipolytica ( Hoy1 [58] ) is induced in hypoxia ( Tables S2 , S3 ) . Upc2 also plays a role in hypoxia-induced filamentation in Y . lipolytica ( Figure 4 ) . The phenotype is generally less pronounced than deleting Ylsre1 , especially in rich media . However , the double deletion fails to filament in any condition . It is therefore likely that in Y . lipolytica , Sre1 and Upc2 both regulate filamentation genes , similar to their dual role in regulating ergosterol metabolism . In mammalian cells and in some fungi , SREBPs act together with sterol-sensing Scap proteins to regulate oxygen sensing [10] . Scaps retain SREBPs in the membrane in C . neoformans and Sch . pombe , though there is no homolog present in A . fumigatus or other Eurotiomycete species [10] , [50] . There is a Scap homolog in Y . lipolytica ( YALI0F00968p ) that contains a sterol-sensing domain and several predicted transmembrane domains ( Figure 7 , Figure S5 ) . There is also an apparent homolog in C . albicans and other species in the CTG clade , but these proteins have lost the sterol-sensing domain ( Figure S5 ) . Within the Saccharomyces clade , Scap has been lost from S . cerevisiae and its close relatives ( Figure 7 ) . However , Scap is present in other lineages , and a potential sterol-sensing domain is clearly identifiable ( Figure 7 , Figure S5 ) . It therefore appears that Scap may play a role in sensing sterols ( and therefore low oxygen ) in some of the species in the Saccharomycotina . In other species ( and in particular in the Saccharomyces and Candida clades ) it is unlikely that Scaps and SREBPs are involved in sterol sensing . We present here a significant example of transcriptional re-wiring , resulting from the substitution of SREBPs , conserved across a wide stretch of evolutionary time from Metazoa to fungi , by the Upc2 transcription factor that arose in the Saccharomycotina . The re-wiring is particularly important for fungal pathogens because expression of the sterol synthesis module confers susceptibility to azole drugs . We also describe a dramatic change in the function of SREBP homologs in the Saccharomycotina , from regulators of sterol synthesis to regulators of filamentation . Our results suggest that Y . lipolytica may represent a transitional stage , where both Upc2 and SREBPs contribute to regulation of sterol metabolism and filamentation .
Yeast strains were maintained on solid YPD agar plates ( 1% yeast extract; 2% Bacto peptone; 2% glucose; 2% agar ) at 28°C unless otherwise indicated . Hygromycin ( Sigma ) was added to YPD ( 1% yeast extra , 2% Bacto peptone , 2% glucose ) at a final concentration of 300 µg/ml . Ketoconazole was added to YPD at a final concentration of 0 . 025 µg/ml and to SC at 1 µg/ml . Hypoxic conditions ( 1% O2 , 99% N2 ) were obtained using an InVivo2 400 hypoxic chamber . Transformants were selected on synthetic complete ( SC ) agar ( 0 . 19% yeast nitrogen base ( YNB ) without amino acids; 2% glucose; 2% agar; 0 . 5% ammonium sulfate; 0 . 075% amino acid drop-out mix lacking the relevant amino acid ) . For phenotype analysis , overnight cultures from single colonies were grown in liquid and solid SC ( 0 . 67% YNB , 2% glucose , 2% agar where required ) or YPD media at 28°C and 200 rpm . 0 . 5 ml of the saturated overnight culture was washed twice with and resuspended in 1 ml PBS . Cells were diluted to 6 . 25×105 cells in 1 ml PBS , and further 5-fold dilutions in PBS were generated . 3 µl were transferred to relevant media , incubated at 28°C at either 21% O2 or 1% O2 for 2 days and then photographed . Bacterial strains were grown on LB agar without NaCl ( 1 . 5% agar; 1% tryptone; 0 . 5% yeast extract ) supplemented with kanamycin at a final concentration of 50 µg/ml . For liquid cultures , cells from a single colony on SC medium were incubated in 10 ml YPD or SC media and 1% Tween 80- EtOH ( vol∶vol = 1∶1 ) where indicated , and incubated at 28°C and 21% O2 or 1% O2 overnight . The cultures were washed , resuspended in PBS and 5 µl was mixed with 5 µl of 0 . 2 mM Calcofluor White ( dissolved in 10 mM NaOH ) and mounted on a glass slide with coverslip . Cells from colonies on solid media after 2 days growth were washed in PBS , resuspended in 100 µl PBS , and stained as above . Cells were observed under UV fluorescence and photographed using a ColorView II camera mounted on a Zeiss AxioImager M1 fluorescent microscope using axiovision software . Gene disruption cassettes were generated using fusion PCR ( Figure S2 ) . Approximately 1000 bp upstream from the start codon of YlUPC2 and YlSRE1 were amplified from Y . lipolytica Po1d using primers UPC2_p1 and UPC2_p2uraa or SRE1_p1 and SRE1_p2_leua/SRE1_p2_uraa , and from downstream of the stop codon using primers UPC2_t1_urab and UPC2_t2 or SRE1_t1_leub/SRE1_t1_urab and SRE1_t2 ( Table S4 ) . The Y . lipolytica LEU2 and URA3 genes were amplified from plasmids JMP802 and JMP803 [84] , [85] using primers LEU-A and LEU-B or URA-A and URA-B . Primers UPC2_p2uraa and SRE1_p2_leua have 20 bp complementary to URA-A and LEU-A respectively ( highlighted in bold , Table S4 ) . Primers UPC2_t1urab and SRE1_t1_leub have 20 bp complementary to URA-B and LEU-B respectively ( marked in bold , Table S4 ) . To make the complete disruption construct , the flanking regions and marker sequences were combined in a fusion PCR reaction using Ex Taq ( TaKaRa Bio Inc . ) with primers UPC2_p1 and UPC2_t2 or SRE1_p1 and SRE1_t2 . The PCR conditions were 2 min at 94° , then 5 cycles of 30 s at 94° , 30 s at 60° , and 1 . 5 min at 72° , followed by 25 cyces of 30 s at 94° , 30 s at 60° , 3 min at 72° , and a final extension of 5 min at 72° . All PCR products were purified using a Qiagen PCR purification kit following the manufacturer's protocol and were introduced into Y . lipolytica by chemical transformation [46] . UPC2 was replaced with URA3 in Y . lipolytica Po1d generating strain SMY1 . The LEU2 marker was introduced into this background to make a prototrophic upc2 deletion ( SMY2 ) . Similarly , SRE1 was replaced with LEU2 in Y . lipolytica Po1d generating strain SMY3 . The URA3 marker was introduced into this background to make a prototrophic sre1 deletion ( SMY8 ) . SRE1 was also replaced with LEU2 in Y . lipolytica JMY330 ( URA3+ ) generating a second prototrophic Ylsre1 deletion strain , SMY5 . To make the double deletion , URA3 was used to replace UPC2 in SMY3 , generating the prototrophic strain SMY4 . The LEU2 marker was introduced into JMY330 to make a prototrophic version of Y . lipolytica Po1d ( JMY2900 ) which is used as the wild type strain in this study . To reintroduce YlUPC2 and YlSRE1 , regions from approximately 800 bp upstream of the start of the gene to the stop of the gene were amplified using primers URI_xhoI_F and URI_avrII_R or SRI_claI_F and SRI_bamHI_R . The forward primers introduce an XhoI ( YlUPC2 ) or ClaI ( YlSRE1 ) site and the reverse primers introduce AvrII or BamHI sites . Digested products were cloned into plasmid JMP804 ( unpublished ) which contains a hygromycin ( Hygex ) resistance marker . Transformation was targeted to the YlUPC2 or YlSRE1 promoters by digestion with either PshaI or PpmuI . The digested plasmids were introduced into SMY2 ( Ylupc2Δ ) or SMY5 ( Ylsre1Δ ) by chemical transformation [46] generating SMY6 ( reconstituted YlUPC2 ) and SMY7 ( reconstituted YlSRE1 ) . All strains were grown in CSM complete medium ( 0 . 17% yeast nitrogen base without amino acids ( Difco ) , 0 . 5% ammonium sulfate , 2% glucose , and supplemented with CSM supplement mixture ( Sunrise Science Products ) ) for 48 h . Cultures equivalent to 100 A600 units were spun down and washed once with 10 ml of sterile water . The pellets were resuspended in 1 ml of an alcoholic KOH solution ( 12 . 5 g KOH , 17 . 5 ml H2O and filled to 50 ml with EtOH ) and incubated at 85°C for 1 hour in 2 ml microfuge tubes . Finally 0 . 5 ml of heptane was added and vortexed for 3 minutes . After separation of the phases , the upper heptane phase was transferred to a new tube . Heptane extracts were diluted with 100% EtOH in 1∶5 ratio and absorbance between 230 and 320 nm was measured [49] . Three peaks of 270 , 280 and 295 were used for quantification ( T-test , p-value<0 . 05 for the Ylupc2/wild type comparison ) . The experiment was performed using three biological replicates , and the average measurements are presented . Cells were grown at 28°C overnight and then diluted to an A600 of 0 . 2 , and grown until they reach an A600 of 1 at 28°C in YPD , in either 21% or 1% O2 . Two to five biological replicates were used per sample ( Table S5 ) . Cells were harvested from 50 ml of culture by centrifugation , and either subjected to RNA extraction or frozen at −80°C . Total RNA was extracted from fresh or frozen cell pellets using a RiboPure Yeast Kit ( Ambion ) . RNA concentrations were determined using a NanoDrop 1000 ( Thermo Scientific ) , while quality and integrity was checked using a Bioanalyzer 2100 ( Agilent Technologies ) . mRNA was prepared from total RNA using oligo dT Dynabeads ( Invitrogen ) . 18 strand-specific libraries ( Table S5 ) were generated by incorporation of dUTP as described in Guida et al [27] , except that several samples were combined in one lane by multiplexing . One of 3 index adaptors ( i6 , i10 or i11 , Table S4 ) was ligated to the samples to allow multiplexing . Adapters were ligated by mixing 25 µl of 2× Quick DNA Ligase Buffer ( NEB ) , 1 µl ( 15 µM ) of the specific adaptor mix , and 3 µl Quick T4 DNA Ligase with library samples . Ligations were carried out for 15 min at 20°C . The DNA was purified with a QIAquick PCR purification kit and MinElute column . The DNA was eluted with 10 µl EB . Sequencing was carried out in-house on an Illumina Genome Analyzer IIx according to manufacturer's instructions , resulting in read lengths of approximately 42 bases . For four samples ( Table S5 ) libraries were generated and sequenced by GATC using an Illumina HiSeq 2000 . All data has been submitted to Gene Expression Omnibus and is available at accession number GSE47433 . Gene annotations were obtained from Génolevures and manually curated using RNAseq data by the Neuvéglise group . In-house reads were processed according to version 1 . 8 of Illumina's Genome Analysis Pipeline . Multiplexed samples were separated using a Perl script and quality was tested using FASTQC ( http://www . bioinformatics . babraham . ac . uk/projects/fastqc/ ) . Each sample dataset was aligned to the 6 Y . lipolytica chromosomes using TopHat [86] . Reads mapped to two or more locations were removed from analysis . Data were visualized using the Artemis genome browser [87] . Raw counts of reads mapped to genes were calculated using HT-Seq . These were used as input for differential gene expression analysis using DESeq , with a P-value cutoff of < = 0 . 05 and LogFC cut off of > = 1 [59] . Differentially expressed gene lists were analyzed using the online DAVID functional annotation tool [61] with the Y . lipolytica gene background and default settings ( Classification Stringency: Medium ) . Gene Ontology ( GO ) FAT terms , KEGG Pathways , InterPro and Swiss-Prot databases were selected for functional annotation clustering . Hierarchical cluster analysis implemented in R was used to identify genes with shared and different expression patterns in the wild type , Ylsre1 deletion and Ylupc2 deletion when exposed to hypoxia . Genes that were differentially expressed in at least one comparison were clustered , using the log2 fold change values generated from DESeq analysis . Hierarchical cluster analysis implemented in R ( http://www . R-project . org ) was used to identify genes with shared and different expression patterns in the wild type , Ylsre1 deletion and Ylupc2 deletion when exposed to hypoxia . Genes that were differentially expressed in at least one comparison were included . Overnight cultures grown at 28°C in YPD media were washed twice with PBS and diluted to an A600 of 0 . 2 in SC media in two flasks . Cultures were grown to an A600 of 1 . 0 at 28°C in normoxic conditions ( 21% oxygen ) and one flask was moved to a hypoxic environment ( 1% oxygen ) for 2 hours . The normoxic sample was resuspended in RNAlater ( Ambion ) and frozen at −80°C . Total RNA was extracted and cDNA was prepared as described previously [88] . qRT-PCR was carried out on an Agilent Technologies Stratagene Mx2005p system using Brilliant III Ultra-Fast SYBR Green QPCR Master Mix ( 600882 ) as per the manufacturer's instructions . Two technical replicates were used for each sample . Cycling conditions consisted of 1 cycle at 95°C for 3 min followed by 40 cycles of 95°C for 10 s and 60°C for 30 s . A final cycle of 95°C for 1 min was followed by melting curve analysis performed at 55°C to 95°C ( temperature transition , 0 . 2°C/sec ) with stepwise fluorescence detection . Primers used for analysis are listed in Table S4 . Relative expression changes were identified using the ΔCT method , compared to the expression of ACT1 . The Upc2 binding site motif [89] was downloaded from the JASPAR database ( http://jaspar . genereg . net/ ) [90] and used as input for tffind ( http://globin . cse . psu . edu/dist/tffind/ ) to scan promoter regions from each genome ( Table 2 ) using a cut off of 0 . 95 ( 95% confidence ) . A one-tailed Fisher exact test was performed to compare the enrichment among ergosterol pathway genes relative to the background group ( rest of promoter regions in genome containing a motif ) . The number of binding sites per promoter was not considered . Fisher exact tests were calculated using the R Statistics package ( http://www . r-project . org ) . SREBP-like proteins were retrieved from the NCBI protein database using BLASTP with human SREBPF1 , Schizosaccharomyces pombe Sre1 , and Cryptococcus neoformans CNJ02310 as queries . Only proteins containing the atypical Tyr residue were retained . Sequences were imported into SeaView [91] for downstream analyses . Sequences from orthologous clades were aligned using MUSCLE [92] , after which these clades were profile-aligned with each other using ClustalW2 [93] . Phylogenetic trees were constructed from the bHLH region of the alignment with PhyML , using the LG substitution model with four rate classes . Similar methods were used to construct phylogenetic trees for Upc2 and Scap proteins . Transmembrane helices were predicted using the TMHMM server ( http://www . cbs . dtu . dk/services/TMHMM/ ) [94] and protein domains were predicted using Pfam [41] . | All but a few eukaryotes die without oxygen and respond dynamically to changes in the level of oxygen available to them . One ancient oxygen-requiring biochemical pathway in eukaryotes is the pathway for the biosynthesis of sterols , leading to cholesterol in animals and ergosterol in fungi . Mutations in this pathway are a frequent cause of azole drug resistance in pathogenic fungi . The regulatory mechanism for the sterol pathway is also widely conserved between animals and fungi and is centred on a transcription activator , SREBP , which forms part of a sterol-sensing complex . However , in one group of yeasts – the Saccharomycotina , which includes the major pathogen Candida albicans – control of the sterol pathway has been taken over by an unrelated regulatory protein , Upc2 . We show here by analysis of the yeast Yarrowia lipolytica that the evolutionary switch from SREBP to Upc2 was a two-step process in which Upc2 appeared in an ancestor of Saccharomycotina , and SREBP subsequently lost its sterol-regulatory function while retaining an ancient role in filamentation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"genome",
"expression",
"analysis",
"mycology",
"fungi",
"functional",
"genomics",
"molecular",
"cell",
"biology",
"genome",
"evolution",
"fungal",
"evolution",
"comparative",
"genomics",
"biology",
"genomics",
"evolutionary",
"biology",
"microbiology"
]
| 2014 | Zinc Finger Transcription Factors Displaced SREBP Proteins as the Major Sterol Regulators during Saccharomycotina Evolution |
Because of evolutionary pressures imposed through episodic colonization by retroviruses , many mammals express factors , such as TRIM5α and APOBEC3 proteins , that directly restrict retroviral replication . TRIM5 and APOBEC restriction factors are most often studied in the context of modern primate lentiviruses , but it is likely that ancient retroviruses imposed the selective pressure that is evident in primate TRIM5 and APOBEC3 genes . Moreover , these antiretroviral factors have been shown to act against a variety of retroviruses , including gammaretroviruses . Endogenous retroviruses can provide a ‘fossil record’ of extinct retroviruses and perhaps evidence of ancient TRIM5 and APOBEC3 antiviral activity . Here , we investigate whether TRIM5 and APOBEC3 proteins restricted the replication of two groups of gammaretroviruses that were endogenized in the past few million years . These endogenous retroviruses appear quite widespread in the genomes of old world primates but failed to colonize the human germline . Our analyses suggest that TRIM5α proteins did not pose a major barrier to the cross-species transmission of these two families of gammaretroviruses , and did not contribute to their extinction . However , we uncovered extensive evidence for inactivation of ancient gammaretroviruses through the action of APOBEC3 cytidine deaminases . Interestingly , the identities of the cytidine deaminases responsible for inactivation appear to have varied in both a virus and host species–dependent manner . Overall , sequence analyses and reconstitution of ancient retroviruses from remnants that have been preserved in the genomes of modern organisms offer the opportunity to probe and potentially explain the evolutionary history of host defenses against retroviruses .
Retroviruses integrate their genomes into host-cell DNA as an essential part of their replication cycle . If a provirus is integrated into the germ line or its progenitors , then it may be inherited in a Mendelian manner as an ‘endogenous’ retrovirus . In fact , endogenous retroviruses have accumulated over time in the genomes of many organisms and are extraordinarily common in mammalian genomes , including that of humans [1] , [2] . Perhaps because of these episodic insults by retroviruses , mammals express proteins , such as TRIM5α and the APOBEC3 family of proteins , that directly inhibit retroviral replication [3] , [4] . Indeed , TRIM5α is responsible for a post-entry restriction of a variety of retroviruses in many primate species through the interaction with incoming retroviral capsids [4]–[8] . Additionally , the APOBEC3 proteins are cytidine deaminases that act primarily by infiltrating retroviral particles and thereafter catalysing cytidine deamination of single-stranded retroviral cDNA during reverse transcription , thus inhibiting viral replication [9]–[16] . Evidence for strong selection pressure on these antiviral factors comes from several observations . First , the APOBEC3 gene family has expanded from one gene in mice to seven genes in primates [17] . Second , sequence analyses of primate APOBEC3 and TRIM5 genes , reveal that they have been under strong diversifying ( positive ) pressure since the divergence of old and new world monkeys 33 million years ago [18]–[20] as well as balancing selection in certain species [21] . Third , there is a striking example of convergent evolution at the TRIM5 locus where a new hybrid TRIM5/cyclophilin gene has independently arisen via retrotransposition and has been selected in two distinct primate species [22]–[27] . Although TRIM5 and APOBEC restriction factors are most often studied in the context of modern primate lentiviruses , it seems likely that these viruses emerged too recently to explain the selective pressure that was evidently imposed on primate TRIM5 and APOBEC3 genes [18]–[20] , [28] , [29] . Rather , it is likely that APOBEC3 and TRIM5 evolved to combat ancient retroviruses , long before the appearance of primate lentiviruses . Endogenous retroviruses can provide us with a “fossil record” of extinct retroviruses and perhaps evidence of ancient TRIM5 and APOBEC3 antiviral activity . Moreover , this record can be accessed in a nearly complete form as a result of recent genome sequencing and annotation efforts . In this study , we examined two families of endogenous gammaretroviruses ( ERVs ) that are relatives of murine leukemia virus ( MLV ) and are present in endogenous form in the genomes of apes and old world monkeys; multiple copies are present in the genome sequences of chimpanzees and rhesus macaques . Curiously , these same two families of MLV-related retroviruses are absent from the human genome [1] , [30] for unknown reasons , while horizontal transfers , endogenization and extinction of these two families of viruses appears to have occurred during the past few million years in several other old world primate species . It is possible that antiretroviral factors such as TRIM5α and APOBEC3 proteins prevented cross-species transmission and perhaps even contributed to the apparent extinction of gammaretroviruses in many primate species . In this regard , it has recently been reported that the TRIM5 protein of humans is active against an ancestral form of one of the endogenous gammaretroviruses examined here ( CERV1 , otherwise referred to as ptERV1 ) [31] . By reconstituting a number of infectious chimeric retroviruses bearing capsid sequences from extinct primate gammaretroviruses , and by examining ‘fossilized’ gammaretroviral sequences , we investigated whether TRIM5 and APOBEC3 proteins restricted the replication of extinct gammaretroviruses that appear quite widespread in the genomes of old world primates but failed to colonize the human germline . Our analyses do not support the conclusion that TRIM5α proteins presented a major barrier to the cross-species transmission of two families of gammaretroviruses that were endogenized over the past few million years , or contributed to their extinction . However , we uncovered extensive evidence for inactivation of primate gammaretroviruses through the action of APOBEC3 cytidine deaminases . Interestingly , the identities of the cytidine deaminases responsible for inactivation of endogenous gammaretroviruses appear to have varied in both a virus and host species-dependent manner .
To determine whether inhibition by species-specific variants of TRIM5α proteins could have limited cross species transmission of these gammaretroviruses , or could have been responsible for their extinction in nonhuman old world primates , we constructed libraries of recombinant MLVs encoding chimeric CA proteins . Specifically , primate CA-NTDs were amplified from chimpanzee and rhesus macaque genomic DNA and introduced into an MLV Gag-Pol expression plasmid in place of the native MLV CA-NTD . Only the NTD of capsid was introduced into the recombinant MLVs because: ( i ) its structure is well conserved in diverse retroviruses and it contains all of the known determinants for restriction by TRIM5α [5]–[8] , [33] , [34] and ( ii ) confining the analysis to this smaller CA domain meant that a reasonable proportion of the amplified natural variants would be devoid of stop codons and , therefore , potentially amenable to functional analysis . For comparative purposes , similar libraries encoding enMLV CA-NTDs were generated . The libraries of recombinant retroviruses contained many hundreds of members that were subjected to the functional analyses outlined below . However , sequence analyses of these libraries revealed many discrepancies between the amplified sequences and sequences present in the genome databases . In principle , these discrepancies could arise as a result of ( i ) natural variation in endogenous CA sequences , ( ii ) sequencing errors in the genome databases , ( iii ) recombination between closely related amplicons during PCR/cloning and ( iv ) Taq polymerase errors . Thus , the ensuing analyses exclude CA-NTD sequences that did not perfectly match a variant that was present in the sequence databases . However , the conclusions described below were precisely the same when large numbers of additional CA-NTD sequences that did not perfectly match a database CA-NTD sequence , were also analyzed ( data not shown ) . To validate our approaches to test the TRIM5 sensitivity of extinct primate gammaretroviruses , we first analyzed MLV-derivatives containing CA-NTD variants from the chimpanzee endogenous gammaretroviruses ( CERV1 and CERV2 ) that precisely matched the consensus sequence for each virus ( termed CERV1-MLV and CERV2-MLV , see Dataset S1 ) . Both of these recombinant viruses were quite infectious , yielding ∼104 infectious units/ml of unconcentrated viral supernatant ( Fig . 2A ) . Notably , however , both CERV1-MLV and CERV2-MLV behaved like B-tropic MLV in that they were entirely resistant to all of the TRIM5 proteins tested . Conversely , a control virus , N-tropic MLV , was strongly inhibited by human , chimpanzee and African green monkey TRIM5α proteins ( Fig . 2A ) . These results contrast with those obtained using a CERV1-MLV chimeric virus previously described by Kaiser et al . that is significantly less infectious than the CERV1-MLV chimera described herein , but was reported to be sensitive to inhibition by human and chimpanzee TRIM5α [31] . However , Kaiser et al . introduced ancestral ( rather than consensus ) sequences and , morever , included the full-length capsid ( NTD and CTD ) as well as the flanking p12 domain from CERV1 in an MLV Gag-Pol plasmid . Our consensus CERV1 CA-NTD sequence differs at one amino acid position from the ancestral CERV1 CA-NTD sequence used by Kaiser et al . ( R35 in the consensus sequence CERV1 , is replaced by Q35 in the ancestral CERV1 CA ) . Therefore , we constructed chimeric MLV-based viruses containing a full-length CERV-1 capsid ( both NTD and CTD ) based on our own consensus CERV-1 NTD sequence , or the ancestral NTD CERV-1 sequence ( Dataset S1 ) . These recombinant viruses were highly infectious ( yielding ∼105 IU/ml ) but , importantly , neither were sensitive to any TRIM5α protein tested ( Fig . 2B ) . Thus , these data argue that CERV1 capsids are not sensitive to either human or chimpanzee TRIM5α , and indicates that our strategy in which the CA-NTDs were introduced into MLV results in capsids that accurately recapitulate the TRIM5α ( in ) sensitivity of the entire ERV capsid . A broader analysis of MLV recombinants ( CERV1-MLV , CERV2-MLV , RhERV1a-MLV , RhERV1b-MLV , RhERV2-MLV and enMLV-MLV ) containing respective consensus ERV CA-NTD sequences or naturally occurring variants ( Dataset S1 ) , revealed that almost all were completely resistant to the collection of primate TRIM5α proteins , and also to the murine restriction factors Fv1N , Fv1B and Fv1NB ( Fig . 2C , D ) . All but one enMLV CA-NTD recovered from NIH3T3 cell DNA displayed B-tropism ( sensitivity to Fv1N and Fv1NB ) ( Fig . 2B and C ) , while one clone was NB-tropic ( Fig . 2D ) . Only one recombinant virus containing a primate gammaretrovirus CA-NTD , a RhERV2-MLV clone , that contained 4 mutations relative to consensus RhERV-2 sequence , exhibited any sensitivity to a TRIM5 protein . Specifically , infection by this RhERV2-MLV variant was inhibited approximately 10-fold by chimpanzee TRIM5α and marginally by human TRIM5α . ( Fig . 2E ) . This clone had a mutation ( E91K ) that is present in ∼20% of the endogenized RhERV2 CA-NTD sequences and , using the MLV CA-NTD as a model [33] , is predicted to be exposed on the surface of the mature hexameric capsid lattice ( Fig . 2F ) . Notably , introducing the E91K mutation into the consensus RhERV2-MLV chimera recapitulated the sensitivity pattern observed with the original , naturally occurring RhERV2 CA-NTD ( Fig . 2E ) , thereby identifying a novel determinant of retroviral sensitivity to TRIM5α proteins . However , this was the sole instance of sensitivity to a TRIM5α protein that was present in the libraries of recombinant MLVs containing CA-NTDs from endogenous chimpanzee , rhesus macaque and mouse gammaretroviruses . Another major antiretroviral activity in primates that can inhibit a number of retroviruses consists of an array of APOBEC3 cytidine deaminases . To investigate whether these might have contributed to the apparent extinction of gammaretroviruses in chimpanzees and rhesus macaques , we first compared the CA-NTD sequences recovered from database searches with corresponding virus-species consensus sequences . Overall , this analysis revealed a varying , but relatively modest , degree of divergence , typically <3% for each individual CA-NTD relative to each virus-specific consensus sequence . However , the nature of the mutations that distinguished each group of CA-NTD sequences from their respective consensus varied . In particular , in several cases , there was a striking excess of changes that could potentially be attributable to cytidine deamination ( G to A and C to T changes ) ( Fig . 3A ) . Moreover , different patterns of mutation were observed in viruses from the same species . For example , in RhERV1a and RhERV1b sequences , G to A and C to T changes occurred at approximately equal frequencies , but less frequently than pooled ‘other’ changes ( Fig . 3A ) . Conversely , in RhERV2 database entries , G to A changes were clearly the most frequent mutation and comprised approximately one half of all the changes relative to the RhERV2 consensus CA-NTD sequence ( Fig . 3A ) . In general , G to A mutations in endogenous primate gammaretroviruses comprised a significantly higher fraction of the total mutations than was the case in endogenous MLVs ( Fig . 3A ) . G to A and C to T mutations are the most common mutations found in mammalian genomes . Therefore , the excessive G to A and C to T mutations observed in endogenous gammaretrovirus sequences could be the result of spontaneous , post integration mutation , or could reflect encounters with specific antiretroviral cytidine deaminases . If these endogenous retroviruses were indeed mutated by APOBEC3 enzymes , then biases in the dinucleotide contexts in which G to A mutations are observed should be evident . For example , APOBEC3G induced mutation results in the biased appearance of plus strand G to A changes in the context of GG dinucleotides ( i . e . GG to AG mutations ) . Conversely , several other APOBEC3 proteins , including primate APOBEC3F , 3B , 3H and murine APOBEC3 proteins preferentially induce GA to AA mutations [9] , [12]–[14] , [16] , [35] . Only rarely have APOBEC enzymes , specifically APOBEC1 , been reported to induce plus strand C to T mutations in retroviruses , by targeting viral genomic RNA [36] . Therefore , as a preliminary indication of whether these extinguished primate gammaretroviruses had encountered antiviral cytidine deaminases we analyzed all the recovered CA NTD sequences for specific dinucleotide context biases accompanying G to A and , as a control , C to T changes . Because the size of the CA-NTD encoding genomic region analyzed ( ∼400 nucleotides ) , and consequently the number of mutations observed in each individual CA-NTD , was rather small , we inspected the pooled CA-NTD sequence data ( Fig . 3C ) as well as the individual CA-NTD sequences ( Fig . 3B ) for each virus species . There was significant variation in the burden of G to A mutations among individual CA-NTDs , with some completely lacking G to A mutations while other contained up to 12 G to A mutations in the ∼400 nucleotide CA-NTD encoding sequence ( Fig . 3B ) . Moreover , striking and variable biases in the patterns of G to A mutations were observed . For example , in CERV1 CA-NTDs , G to A changes occurred primarily in the context of GG dinucleotides ( Fig . 3B , C ) . This bias was statistically significant , and was maintained when corrected for the frequency with which each GN dinucleotide occurred in the consensus sequence ( Fig . 3D ) . In CERV2 CA-NTDs , the occurrence of G to A changes was biased toward both GG and GC dinucleotides . However , in this case , the overall numbers of G to A changes were small ( Fig . 3B ) because only 10 CERV2 proviruses are present in chimpanzee DNA . Interestingly , the closely related rhesus macaque endogenous gammaretroviruses RhERV1a , RhERV1b and RhERV2 exhibited a different mutational bias as compared to CERV1 and CERV2 chimpanzee counterparts: G to A mutations in RhERV1a and RhERV2 occurred frequently at both GG and GA dinucleotides ( Fig . 3B , C ) , again these biases were highly statistically significant ( Fig . 3D ) . In contrast , RhERV1b CA NTDs were quite different in that they had relatively few G to A mutations and their occurrence was comparatively unbiased with respect to dinucleotide context ( Fig . 3B , C , D ) . These data suggested that some endogenous proviruses may well have encountered antiviral cytidine deaminases . Therefore , to confirm and extend these analyses , we asked if similar patterns of mutation could be observed in another region of the endogenous proviruses . In particular we analyzed Env sequences , which are obviously longer than CA-NTD sequences and permit a more robust estimate of context-specific mutation frequencies in individual proviruses . Retrieval and analysis of the endogenous gammaretroviral Env sequences revealed that their phylogenetic relationships to each other and to other gammaretroviruses were similar to those observed using CA-NTD sequences ( Fig . S1 ) . The exception to this was BaEV , which was closely related to CERV2 and RhERV2 in CA , but not in Env , presumably as a result of recombination between ancestral viruses . Because there are large numbers of CERV-1 proviruses in chimpanzee DNA , and the genome database is comparatively complete , we retrieved 10 individual Env sequences linked to CA-NTDs that had 1 or fewer G to A mutations and 10 Env sequences linked to CA-NTDs that contained 4 or more G to A mutations . Overall , there was a good correlation between the CA-NTD and Env sequences in terms of the burden and character of the G to A mutations ( Fig . 4 A , B ) . Specifically , the occurrence of GG to AG mutations in a given CERV1 CA-NTD strongly predicted the occurrence of the same type of mutation in the linked Env sequence . Indeed , there was only one exception to this finding among the 20 CERV-1 sequences analyzed , in which a CA-NTD bearing three GG to AG and one GC to AC mutation was linked to an Env sequence containing a light burden of G to A changes ( Fig . 4 A , B ) . Because there are only ten CERV2 proviruses in the chimpanzee genome ( and only four of them are complete , with others ranging in size from 2 . 6 to 7 . 7 Kb ) we took a different approach to confirm or refute the notion that they contained excessive or biased G to A mutations . Specifically , we analyzed all of the available sequence for all ten proviruses . In part because the CERV2 proviruses were intrinsically divergent , and differed in length , the absolute numbers of G to A changes as compared to the consensus sequence was quite variable among them ( Fig . 4C ) . However , when the character of the G to A mutations was examined , five of the 10 proviruses exhibited a clear , statistically significant , excess of GG to AG mutations , as compared to overall G to A changes ( Fig . 4C ) . Thus , the patterns of nucleotide substitutions strongly suggested that both groups of endogenous gammaretroviruses that integrated into the chimpanzee genome in the past few million years frequently encountered the only mutator known to preferentially induce GG to AG mutations , namely APOBEC3G , prior to or during endogenization . The coverage of the rhesus macaque reference genome database is less complete than that of the chimpanzee genome , so only 23 , 10 and 22 envelope sequences that were unambiguously linked to CA-NTDs could be retrieved for RhERV1a , RhERV1b and RhERV2 respectively . Moreover , in the case of RhERV2 , the CA-NTD linked Env sequences were distributed among two distinguishable subgroups ( Fig . S1 ) and , therefore , a separate Env consensus sequence was deduced for each group . Surprisingly , in RhERV1a , the nature and burden of G to A changes in Env sequences was not well predicted by the findings in the CA-NTD sequences . Indeed , the preferential occurrence of G to A changes in Env in the context of GG and GA dinucleotides was marginal , and clearly less pronounced than in the CA-NTD sequences ( Fig . 4 , A , B ) . The reasons for this discordance are unclear and discussed below . However , for both RhERV1b and RhERV2 the analysis of Env sequences corroborated the results obtained using CA-NTDs ( Fig . 4A , B ) . Specifically , the numbers of G to A changes in RhERV1b Env , like CA-NTD sequences , were modest and unbiased with respect to dinucleotide context , while in RhERV2 , there was a clear excess of both GG to AG and GA to AA mutations in both CA-NTD and Env sequences . ( Fig . 4A , B ) . However , while there was concordance with respect to overall context bias of G to A mutations present in RERV2 CA-NTD and Env sequences , inspection of individual linked CA-NTD and Env sequences revealed that the burden and character of G to A mutations was variable . Specifically , several Env sequences bore GA to AA hypermutation while two other Env sequences bore predominantly GG to AG hypermutation . Additionally , the degree to which CA-NTDs and Env sequences were mutated in individual proviruses did not always correlate; in some cases hypermutated RhERV2 Env sequences were linked to CA-NTD sequences that bore relatively few G to A changes ( Fig . 4A , B ) . These results , and the discordant findings with respect to G to A mutation in RhERV1a CA-NTD versus Env sequences suggest that different regions of the genome can be mutated at different frequencies ( a known characteristic of APOBE3G induced mutation ) . Alternatively it is conceivable thatrecombination between hypermutated and nonhypermutated viral genomes can occur prior to or during deposition in the germ line . Indeed , inspection of several complete or nearly complete RhERV2 proviruses ( Fig . S2 ) revealed clear variation in the extent of G to A mutation across the viral genome in some ( but not all ) proviruses . Notably , an obvious 5′ to 3′ gradient of increasing mutation intensity was present in some examples bearing excessive G to A mutations , as has previously been reported in hypermutated HIV-1 genomes [35] , [37] . The chimpanzee and rhesus gammaretroviral CA-NTD and Env sequences also contained a striking excess of plus strand C to T changes relative to their respective consensus sequences ( Fig . 3A ) . Potentially , the high rate of C to T mutations in CA-NTD sequences could be the result of spontaneous cytidine deamination after integration of the provirus into the germ line . Correspondingly , excessive plus strand G to A changes could be a reflection of spontaneous cytidine deamination events on the minus strand . In contrast to APOBEC3-mediated deamination/mutation events , whose frequency is profoundly influenced by the identity of the nucleotide in the −1 position relative to the deaminated cytidine , mutations that arise as a result of spontaneous cytidine deamination are influenced by the nucleotide in the +1 position . Specifically , the most common spontaneous cytidine deamination induced mutation occurs in the context of CG dinucleotides , ( i . e . G in the +1 position ) . This results in CG to TG or CG to CA plus strand mutations , depending on whether CG dinucleotides on the plus or minus strand are deaminated , respectively [38] , [39] . Therefore , to account for this potential source of “noise” , we counted and categorized all G to A and C to T changes for both CA-NTD and Env sequences before and after exclusion of CG dinucleotides in the minus and plus strand respectively ( Fig . S3 , S4 ) . Notably , the biases that were associated with plus strand G to A mutations ( equivalent to minus strand C to T mutations ) were not evident in a similar analysis of plus strand C to T mutations . Moreover , exclusion of C to T mutations that occurred in the context of CG dinucleotides on the plus or minus strands , reduced the overall number plus strand G to A and C to T mutations but did not affect conclusions with respect to the dinucleotide context bias associated with G to A mutations ( Fig . S3 , S4 ) . Notably , the relative extent of G to A as compared to C to T mutations varied among individual proviruses of a given virus group as well as between virus groups . This was most evident in the analysis of Env sequences ( or complete proviral sequences in the case of CERV2 ) , because the longer length of sequence analyzed permitted a more robust estimate of G to A and C to T mutation frequencies ( Fig . S4 , S5 ) . For many of the individual CERV1 , CERV2 , and RhERV2 proviruses , the frequency of G to A mutations greatly exceeded the frequency of C to T mutations ( Fig . S4A , C , Fig . S5 ) , and in proviruses where this was the case , the excessive G to A mutation was invariably in the context of GG or GA dinucleotides , and remained evident when CG dinucleotides were purged from the analysis ( Fig . S4B , D , Fig . S5 ) . Conversely , RhERV1a and RhERV1b Env sequences exhibited broadly similar levels of comparatively unbiased G to A and C to T mutation ( Fig . S4 ) suggesting that the frequency of APOBEC3-induced mutation was low compared to that from other sources . A comparision with endogenous MLVs revealed a similar situation to that in primate gammaretroviruses , and as was previously reported , two out of the three groups of endogenous MLVs ( Pmv , Mpv , but not Xmv ) exhibited higher levels of G to A than C to T mutations , and this excess in G to A mutations was associated with similar context biases [32] . Overall , clear dinucleotide context biases were observed in each situation where G to A mutations clearly outnumbered C to T mutations . Thus , the excessive and context-biased G to A mutations that are present in some of the groups of primate endogenous gammaretroviruses cannot be explained by spontaneous cytidine deamination . Rather , the analysis of both CA-NTD sequences , and linked envelope sequences , provides robust support for the hypothesis that APOBEC3G in chimpanzees , and a combination of APOBEC3G and other APOBEC3 proteins in rhesus macaques , extensively mutated gammaretroviral genomes prior to or during their endogenization . Inspection of the endogenous primate gammaretroviral CA-NTD sequences revealed that , with the exception of RhERV1b CA-NTD sequences , many harboured stop codons ( Fig . 5A ) . Strikingly , a large fraction of the stop codons were generated by G to A mutations ( Fig . 5A ) . C to T changes resulted in stop codons somewhat less frequently than G to A changes , but still provided a major source of these obviously inactivating mutations . ( Fig . 5A ) . Frameshifts and other mutations were responsible for only a small proportion of protein-truncating mutations . Many of the C to T changes occurred in CG dinucleotides and are therefore likely to be caused by spontaneous post-integration deamination events . Notably , all of the stop codons that arose through the appearance of G to A mutations were due to a change from Trp ( TGG ) to termination codons ( TAG , TGA or TAA ) , often reflecting the dinucleotide bias associated with the G to A changes observed in those species . The large number of stop codons generated at Trp codons , likely by APOBEC3-mediated cytidine deamination , would almost certainly functionally inactivate many of these endogenous retroviruses . Note that this analysis was confined to the CA-NTD sequences , and thus significantly underestimates the total number of termination codons in full-length proviruses . In addition to stop codons , G to A and C to T changes introduced missense mutations in CA NTDs whose effects cannot be determined simply by inspecting sequence data . Although it was not practical to determine the effect of each G to A and C to T mutation in isolation , a few ( fourteen ) members of the libraries of chimeric CERV1- RhERV1b- or RhERV2-MLVs bore single , naturally occurring , missense mutations that could likely be attributed to enzymatic cytidine deamination ( GG to AG changes ) or spontaneous cytidine deamination ( CG to CA or CG to TG changes ) . Chimeric MLVs bearing these single mutations were invariably less infectious than corresponding MLVs encoding the intact CERV1 , RhERV1b or RhERV2 consensus CA-NTD sequence ( Fig . 5B ) . The magnitude of decreases in infectivity were variable , and in one case a missense GG to AG mutation completely inactivated a CERV1-MLV chimera . Thus , amino acid differences from the consensus , likely attributable to enzymatic or spontaneous cytidine deamination decreased infectivity , suggesting that they were unlikely to be positively selected to provide benefit to a replicating virus . Rather , in addition to their propensity to induce the occurrence of termination codons , the abundant nonsynonymous G to A and C to T changes likely contributed significantly to a deleterious genetic load borne by these populations of endogenized proviruses .
Ancient epidemics of viral infection likely contributed to the selection of modern genes with potent antiretroviral activity , such as the TRIM5 and APOBEC3 families of antiretroviral factors . Since these selection events did not involve contemporary viruses , understanding how reciprocal selective pressures were imparted is problematic . However , the availability of nearly complete genome sequences for some modern primates , and the propensity of germline-tropic ancient retroviruses to be preserved in the genomes of modern organisms , potentially allows part or all of their sequences to be accessed and reconstituted in infectious form . In this study , we undertook an analysis of two groups of MLV-related retroviruses that are preserved in the genomes of modern old world primates , but are notably absent from human DNA . The goals of this study were to obtain evidence for or against the notion that TRIM5 and/or APOBEC3 limited the host range and/or contributed to the extinction of these retroviruses . Notably , a prototype gammaretrovirus , MLV , has been shown to be sensitive to inhibition by human APOBEC3G and one variant , N-MLV , has been shown to be sensitive to inhibition by several old world primate TRIM5α proteins . Thus , it was plausible that ancient gammaretroviruses might be relevant targets for host-range-limiting antiretroviral defense mechanisms . By reconstituting infectious viruses that encoded endogenous retroviral CA-NTDs , we obtained in vitro evidence suggesting that TRIM5α proteins did not play a prominent role in the inactivation of , or limit the cross-species transmission of these primate gammaretroviruses . Indeed , all TRIM5 proteins tested lacked restriction activity against all endogenous N-terminal domain capsids studied , except in one instance where a RhERV2 CA-NTD carrying a E91K mutation was shown to be sensitive to chimpanzee TRIM5α . Nevertheless , a close relative of RhERV2 , namely CERV2 , is present in the chimpanzee genome . It is perhaps noteworthy that none of the CERV-2 CA-NTDs encode K91 , perhaps suggesting that a limited subgroup of the CERV2/RhERV2 group of viruses had the propensity to colonize chimpanzees , or that chimpanzee TRIM5α might constrain CERV2 CA-NTD sequence variation . There are two potential caveats to the conclusion that TRIM5α proteins did not cause extinction or impact the host range of these primate gammaretroviruses . First , the inactivation of these primate gammaretroviruses , and the consequent termination of the putative selective pressure imposed on TRIM5 genes , may have occurred in the sufficiently distant past that subsequent pressure on TRIM5 genes caused loss of gammaretroviral restriction activity . In other words , the modern TRIM5 proteins tested here may lack specificity for ancient viral capsids that ancestral TRIM5 proteins possessed . However , since the replication of CERV1/2 occurred after the divergence of humans and chimpanzees six million years ago , any significant shifts in restriction factor evolution caused by the rise and fall of species-specific retroviruses should be reflected in the divergence of the human and chimpanzee genes; but despite the absence of CERV1/2 from the human genome and its presence in the chimpanzee , TRIM5α remains largely conserved , in sequence and functional terms , in these two species . In short , the replication of CERV1/2 and their orthologs appears too recent relative to TRIM5 functional divergence for a subsequent shift in TRIM5α specificity to account for the observation that modern TRIM5α proteins do not inhibit these two groups of retroviruses . Second , it is possible that there are determinants of TRIM sensitivity that are outside of the N-terminal domain of capsid . This was of particular concern given the findings of Kaiser et al . who , while this work was in progress , reported that a chimeric MLV encoding the p12 and capsid proteins of an ancestral CERV1 , was indeed restricted by human and chimpanzee TRIM5α [31] . The presence of the CERV1 p12 domain could , theoretically , affect TRIM5 sensitivity , but all known determinants of TRIM5 or Fv1 sensitivity map to the CA-NTD [5]–[8] , [33] , [34] . Moreover , cleavage from the p12 domain is required for MLV CA to interact with TRIM5α and Fv1 [40] , so it is difficult to see how p12 sequences could affect sensitivity to TRIM5α . More likely was the possibility that a CA-CTD element that affects capsid stability or TRIM5 interaction might affect TRIM5 sensitivity and would not be represented in our CERV1 CA-NTD/MLV chimera . However , when we generated a chimera that contained the full-length CERV1 CA , with either consensus or ancestral NTD linked to an ancestral CTD , the resulting virus was highly infectious ( ∼105 IU/ml ) and clearly insensitive to all TRIM5α proteins tested . Thus , this result supports the notion that CERV1 was not restricted by the TRIM5α proteins encountered in its natural target species and validates our experimental strategy for analyzing a broader range of CA-NTDs for TRIM5 sensitivity . We note that TRIM5 is not the only TRIM family member that is a potential candidate for a suppressor of endogenous retroviral activity . TRIM1 in particular , as well as other TRIM proteins , have been demonstrated to exhibit anti-MLV activity in vitro [7] , [41] . Additionally , other TRIM proteins , specifically TRIM22 can exhibit signatures of positive seletion , suggesting a potential role as an antiviral factor [42] . However , the identity or identities of viruses that were responsible for positive selection in TRIM genes remains to be determined . Lentiviruses , spumaretroviruses , and gammaretroviruses have all been shown to be intrinsically capable of being restricted by TRIM5 proteins [4]–[8] , [43] . Thus retroviruses from any one , or any combination of these ( or other ) genra , could potentially have imposed selection on TRIM genes . Moroever , it is not necessarily the case that the retroviruses that were responsible for diversifying selection of TRIM5 are represented in the genomes of modern organinsms . Indeed , most retroviruses that infected ancestral hosts probably did not become established in the germ line of their host species . While these findings do not suggest that TRIM5α proteins inhibited the replication of these two groups of primate gammaretroviruses , we did find clear evidence for encounters between the endogenized viruses and another major arm of the antiretroviral host defense , in both the chimpanzee and the rhesus macaque . This conclusion is based on the frequent occurrence of G to A changes that occurred in dinucleotide contexts ( GG to AG and GA to AA ) indicative of the action of APOBEC3 cytidine deaminases . Notably some viruses , in particular the rhesus macaque virus RhERV2 , displayed two distinct dinucleotide biases on distinct proviruses , while only the GG to AG bias was evident in CERV-1 and CERV2 . Thus , it appears that APOBE3G , the only known mutator that preferentially induces GG to AG changes , was largely responsible for hypermutation in chimpanzees while rhesus macaque gammaretroviruses were subjected to the action of two different APOPBEC3 proteins . Assays employing Vif-deficient HIV-1 indicate that rhesus macaque APOBEC3 proteins that induce GA to AA mutations , particularly APOBEC3F and APOBEC3H [14] , [44] , [45] , are more active mutators relative to their human counterparts , while human APOBEC3G is a more potent mutator than rhesus macaque APOBEC3G [14] , [45] . Thus , assuming that the properties of chimpanzee APOBEC3 proteins are similar to those of humans , the spectrum of mutations found in highly related gammaretroviruses that colonized two different primate species appears to correlate with the properties of the APOBEC3 proteins found therein . Several analyses confirmed that the excessive G to A mutations in GG or GA dinucleotides were both nonrandom and not due to amplification of a relatively small number of G to A changes from founder viruses ( data not shown ) . Moreover , by removing CG dinucleotides from our analyses , we confirmed that the GG to AG mutations observed were not due to spontaneous deamination at methylated CG dinucleotides . In some cases , the appearance of biased G to A mutations in CA-NTD was corroborated by the appearance of the same biases in linked Env or complete proviral sequences . Sometimes , however , individual CA-NTDs were not good predictors of frequent G to A mutation in other regions of the provirus . This is likely because: ( i ) the individual CA-NTDs are short ( ∼400 nt ) sequences and thus subject to stochastic variation in the numbers of G to A mutations , ( ii ) variation in the mutation frequency ( including 5′ to 3′ gradients ) across an individual provirus , ( iii ) recombination between hypermutated and intact proviruses prior to or during deposition in the germline , and/or ( iv ) position-dependent intrinsic variation in non-APOBEC3 induced sequence diversity in different viral genes that may mask low frequency APOBEC3-mediated editing . APOBEC3-mediated and spontaneous cytidine deamination events contributed a significant genetic burden to many of the individual proviruses analyzed here , through the frequent introduction of stop codons and other deleterious nonsynonomous mutations . Because of the high frequency of stop codons that were generated by GG to AG mutations , it is likely that APOBEC3 activity had an impact on the viral populations within each infected individual and may have inhibited intra- and interspecies transmissions . It is even possible that APOBEC3-induced mutation was responsible for terminating the replication of these retroviruses in their respective hosts . Such a definitive conclusion would be premature , however , because the frequency with which APOBEC3 mutated proviruses were deposited in the chimpanzee and rhesus macaque germ line may not be representative of the frequency with which APOBEC mutation actually occurred during active infection . Indeed , the population of proviruses present in the germ line is likely enriched for defective , including APOBEC3-mutated , variants , as these are less likely to be deleterious to the host . An intriguing possibility , currently under investigation , is that defective APOBEC-3 mutated proviruses may in fact be beneficial to the host as a source of defective interfering genomes or proteins . Concurrently , we and others have found that another endogenous retrovirus , HERV-K , has also been subjected to hypermutation by APOBEC3G in humans [46] , [47] . Additionally , some endogenous variants of MLV have been shown to bear similar footprints of encounters with the mouse APOBEC3 protein [32] . Thus , evidence is accumulating that APOBEC3 proteins , particularly APOBEC3G , provided a broad defense against a variety of formerly exogenous , now endogenous , retroviruses . Reconstitution of ancient retroviruses from remnants that have been preserved in the genomes of modern organisms offers the opportunity to probe and potentially explain the evolutionary history of host defenses against retroviruses .
To identify MLV-related viruses in primate genomes , the amino acid sequence of the capsid N-terminal domain of moloney murine leukemia virus first was used in a TBLASTN search of the chimpanzee genome ( http://www . ensembl . org ) . After retrieving all complete TBLASTN hits , the sequences were separated into two families , corresponding to CERV1 and CERV2 , according to sequence homology and length . After removing redundant sequences , ClustalW was used to obtain nucleotide and amino acid consensus sequences for both CERV1 and CERV2 capsid N-terminal domains . The same method was used to find orthologues of CERV1 and CERV2 CA-NTDs in Macaca mulatta ( RhERV1 and RhERV2 , respectively ) . RhERV1 CA-NTDs included two phylogenetically distinct subgroups , termed RhERV1a and RhERV1b , and two independent consensus sequences were derived . Similarly , CA-NTD sequences from enMLV , were retrieved from the C57BL/6J genome sequence database . A phylogeny was reconstructed for all capsid NTDs using ClustalX1 . 8 and phylogenetic trees were drawn using Figtree . Individual CA-NTD amino acid sequences were used in TBLASTN searches and regions positioned immediately 3′ to precise hits were inspected for Env-like sequences . The moloney MLV Env sequence was used to assist in the definition of theoretical CERV1/CERV2/RhERV1/RhERV2 Env open reading frames . To define a consensus envelope sequence for each endogenous retrovirus family , a single envelope amino acid sequence was used as a TBLASTN query . All resulting envelope sequences were aligned to derive a consensus sequence , regardless of their linkage to a previously retrieved capsid sequence . Analyses of the CA-linked envelope genes was then performed as for the CA-NTD sequences . In the case of RhERV2 , two distinct groups of envelope sequences were observed and separate alignments and consensus sequences were derived for each subgroup . Similarly , three distinct enMLV envelopes were grouped according to the previously defined tropisms of these viruses: polytropic , modified polytropic , and xenotropic [48] . In the case of CERV1 , because of the large numbers of unique CA-NTD sequences , the analysis was confined to two groups of 10 CA-NTDs that were each used to define linked Env sequences . One CA-NTD group had zero or one G to A mutations , and a second group had ≥4 G to A mutations . Because of the small number of ( 10 ) CERV2 integrations in the chimpanzee germ line , all CERV2 proviral sequences were compiled . Flanking regions 1500 nucleotides 5′ and 8000 nucleotides 3′ to the capsid NTDs were aligned . Four of the proviruses aligned for 8293 nucleotides with pairwise identity scores of at least 83% . Therefore , those portions of sequence were trimmed and defined as full length CERV2 provirus . Six additional incomplete proviruses were defined . One database derived proviral sequence was excluded because a 99 . 9% sequence identity with another locus made it unlikely to be an independent retroviral integration , and is instead likely to be either a genomic duplication or an error in the compiling of genomic sequence . A library of chimeric GagPol plasmids was generated that contained MoMLV GagPol in which the capsid NTD was replaced with that from endogenous retroviruses . Genomic DNA was isolated using standard protocols from Pan troglodytes verus skin fibroblasts AG06939 ( Coriell ) , Macaca mulatta 221 cells , and NIH 3T3 cells and used as PCR template . Primers were designed to anneal to the capsid NTD consensus sequences described above . Because of relatively high conservation at the 3′ end of the capsid NTD , the same reverse primer was used for all endogenous retrovirus families: 5′- TACYTTRGCCAAATTRGTRGG -3′ . The 5′ primers were specific to each virus family and all contained a BsmI restriction site: CERV1: 5′- CTCGCAGGCATTCCCCCTTCGGGAAATAGG -3′; CERV2: 5′- CTCGCAGGCATTCCCCCTCCGCACCGTG -3′ . To amplify capsid genes from the Rhesus macaque and mouse genomes , primers were designed to anneal to a p12 region immediately 5′ of the capsid- RERV1a: 5′- CAGCYGCCTGACTCYAYGGTGGCATTCCCCCTT -3′; RERV1b: 5′- CRACTCCCTGACTCCACYGTGGCATTCCCTCTC -3′; RERV2: 5′- CCTTCYACTTGGCAATCCTCGGCATTCCCCCTC -3′; and enMLV: 5′- GCRGAYTCCACCWCCTCYCRGGCATTCCCACTC -3′ . A small fragment was amplified from MLV GagPol using the reverse complement of the capsid NTD 3′ primer: 5′- CCYACYAATTTGGCYAARGTA -3′; and a primer annealing in the MLV capsid C-terminal domain: 5′- CTTCTAACCTCTCTAACTTTCTCC -3′ . The amplified portion of the MLV GagPol plasmid was engineered to contain an AfeI restriction site . Thereafter , an overlapping PCR product was generated that included the amplified endogenous virus capsid NTDs and the MoMLV capsid CTD fragment . This PCR product was cloned into the MLV GagPol plasmid using BsmI and AfeI restriction enzymes . The resulting chimeric GagPol plasmids were isolated and screened for functional capsid genes by cotransfection with an MLV-based vector containing GFP and VSV-G envelope using polyethylenimine ( PEI ) in 293T cells . Supernatant from these cells were used to infect hamster CHO-KI-derived cells . Two days post-infection , cells were trypsinized , fixed in 2% PFA and subjected to FACS analysis using a Guava Easycyte to determine the percentage infected ( GFP-positive ) cells . To measure TRIM5α , TRIMCyp and Fv1 sensitivity , Mus dunni tail fibroblasts ( MDTF ) or MDTF stably expressing human TRIM5α , chimpanzee TRIM5α , rhesus macaque TRIM5α , african green monkey TRIM5α , owl monkey TRIMCyp , Fv1N , Fv1B , or Fv1NB that have previously been described [5] , [49] , [50] . Fv1NB is an unnatural chimeric Fv1 protein that restricts both N-tropic and B-tropic MLV strains . The TRIM5 and Fv1-expressing cells were infected in the presence of 5 µg/mL polybrene and GFP-positive cells were quantified by FACS analysis two days post-infection . To determine whether the E91K mutation in the RhERV2 CA-NTD modulates its sensitivity to human and chimpanzee TRIM5α , this point mutation was introduced by PCR into a chimeric MLV GagPol plasmid containing a consensus RhERV2 CA-NTD sequence . Additionally , to determine whether the CA-CTD altered CERV1 sensitivity to TRIM5α , a CERV1 CTD sequence , as described in Kaiser S . M . et . al . [31] , was synthesized using a series of overlapping ∼60 nucleotide olignucleotides . The resulting product was then used in an overlapping PCR reaction with the CERV1 CA-NTD consensus sequence and this full-length capsid was cloned into the MLV GagPol plasmid . In order to match the CERV1 CA-NTD consensus to the ancestral sequence described in Kaiser et . al . [31] , the R35Q mutation was also introduced using PCR-based mutagenesis into this CERV1 full-length capsid/MLV chimera . Mutations and their dinucleotide sequence contexts were quantified using Hypermut ( http://www . hiv . lanl . gov/content/sequence/HYPERMUT/hypermut . html ) . We generated fasta alignments using MacVector to use as input for Hypermut and the consensus sequence for each virus family ( described above ) was used as a reference sequence . The Hypermut output provided the total number nucleotide changes ( each insertion/deletion was reduced to a single nucleotide change in MacVector ) as well as the number of G to A mutations relative to the consensus sequence . Hypermut also provided the number of G to A changes that occurred in each of the four possible dinucleotide contexts when the nucleotide immediately 3′ to the G ( or the +1 position ) in the consensus sequence is considered ( i . e . GG , GA , GC and GT ) . To quantify the −1 position for C to T changes , the +1 position for C to T changes , and the −1 position for G to A changes , the reverse complement , complement , and reverse of each alignment was generated respectively and used as Hypermut inputs . The latter two analyses allowed us to quantify the number of C to T and G to A changes occurring in CG dinucleotides . We repeated this analysis , after removal of G to A and C to T changes occurring in CG dinucleotides . In some charts , the percentages of G to A or C to T changes that were in each dinucleotide context were normalized according to the dinucleotide composition of a given consensus sequence . P-values were calculated for these data using a chi-square goodness of fit statistical test . | Retroviruses integrate their genomes into host-cell DNA as an essential part of their replication cycle . If a retrovirus is integrated into a cell that becomes a germ line cell such as a sperm or an egg , then it may be inherited as an ‘endogenous’ retrovirus . In fact , endogenous retroviruses are extraordinarily common in mammalian DNA , constituting about 8% of human DNA . These endogenous retroviruses are mostly derived from ancient viruses that are now extinct . In this study , we recovered parts of two groups of extinct retroviruses , many strains of which became integrated into genomes of many nonhuman primates over the past few million years , but are absent from human DNA . We were able to generate infectious retroviruses by inserting a part of the extinct viruses into a modern retrovirus found in mice , and in so doing were able to functionally analyze properties of the extinct virus . Using a combination of these functional analyses , as well as sequence analysis , we obtained evidence that some rapidly evolving host defense molecules present in modern primates were able to inhibit the replication of these extinct viruses . Therefore , particular host defenses may have limited transmission of ancient retroviruses between species and perhaps contributed to their extinction . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"virology/host",
"antiviral",
"responses"
]
| 2008 | Evidence for Restriction of Ancient Primate Gammaretroviruses by APOBEC3 but Not TRIM5α Proteins |
Neuroinvasive larvae of the worldwide occurring zoonotic roundworms Toxocara canis and T . cati may induce neurotoxocarosis ( NT ) in humans , provoking a variety of symptoms including cognitive deficits as well as neurological dysfunctions . An association with neuropsychological disorders has been discussed . Similar symptoms have been described in T . canis-infected mice , whereas data on T . cati-induced NT are rare . Therefore , it was aimed to obtain insights into the impact on neurobehaviour as well as progression of neurological symptoms and behavioural alterations during the course of NT directly comparing T . canis- and T . cati-infected mice as models for human NT . C57BL/6 mice were orally infected with 2000 embryonated T . canis or T . cati eggs , respectively , the control group received tap water . Mice were screened weekly for neurobehavioural alterations and memory function starting one day prior infection until 97 days post infection ( pi; T . canis-infection ) and day 118 pi ( T . cati-infection , uninfected control ) . Mostly motoric and neurological parameters were affected in T . canis-infected mice starting day 20 pi with severe progression accompanied by stereotypical circling . In contrast , T . cati-infected mice mostly showed reduced response to sudden sound stimulus ( indicator for excitability ) and flight behaviour starting day 6 pi . Interestingly , enhanced grooming behaviour was observed exclusively in T . cati-infected mice , indicating a possible role of neurotransmitter dysregulation . Reduced exploratory behaviour and memory impairment was observed in both infection groups with delayed onset and less severe progression in T . cati- compared to T . canis-infected mice . Results highlight the need to consider T . cati beside T . canis as causative agent of human NT . Findings provide valuable hints towards differences in key regulatory mechanisms during T . canis- and T . cati-induced NT , contributing to a comprehensive picture and consequently a broader understanding of NT , which will aid in developing strategies towards prevention in addition to novel diagnostic and therapeutic approaches .
Toxocarosis , caused by infective third stage larvae ( L3 ) of the worldwide occurring roundworms of dogs and cats , Toxocara canis and Toxocara cati , respectively , is one of the most common parasitic infections in humans . Nevertheless , the global importance of the zoonotic helminth is considered to be underestimated [1 , 2] . Consequently , toxocarosis has been listed as one of the five most neglected parasitic infections in humans targeted as priority for public health action by the Centers for Disease Control and Prevention ( CDC ) [3] . Infections are most frequently acquired by ingestion of embryonated infective eggs from the environment with poor sanitation as well as poor hygiene standards favouring the risk for infection . Thus , mostly children and people living in poverty are affected . Favorable environmental conditions for development of Toxocara eggs as well as transmission in tropical climates result in comparably high contamination as well as infection rates in these countries . Exemplarily , seroprevalences up to 92 . 8% in La Réunion , France , were reported , demonstrating frequent exposure of humans to Toxocara . Considerably lower seroprevalences are detected in temperate regions as for example 2 . 4% in Denmark [4] . Human toxocarosis is associated with several forms of disease with varying symptoms [1] . One of these forms , neurotoxocarosis ( NT ) , is induced by neuroinvasive L3 and may result in pathological presentations such as meningitis , encephalitis and myelitis . Also , cerebral lesions have been predominantly observed in the white matter in addition to occlusion of cerebral arterial vessels [5–7] . This CNS involvement may result in symptoms such as motor impairments , neuropsychological disturbances like dementia or depression as well as in behavioural alterations [7–13] . Previous studies also correlated Toxocara-seroprevalence with reduced cognitive development and function in children and young adults [14–17] . Sporadically , an implication in neurodegenerative diseases like multiple sclerosis and Alzheimer’s disease has been suggested [18 , 19] . Similar observations on clinical symptoms and neurological alterations have been made in mice which are therefore considered suitable model hosts for human NT . However , an extensive neurotropism with persistence in the brain has solely been described for T . canis larvae . Contrary , T . cati larval migration to neuronal tissue is less frequently observed and larvae mainly accumulate in skeletal muscle tissue [20] . Both species cause structural damage such as malacia and demyelination , which is more severe in T . canis- than in T . cati-infected mice [21–24] . Additionally , previous studies demonstrated partially severe neurological symptoms as well as behavioural alterations and memory impairment in T . canis-infected mice [25–27] , whereas behavioural data about T . cati-induced NT is scarce . However , previous studies indicate earlier onset accompanied by more severe neurological symptoms in T . canis- compared to T . cati-infected mice [24] . Nevertheless , dysregulations of genes associated with the biological modules “behaviour and taxis” as well as “sensory perception/neurological system process” have been described in T . canis- as well as in T . cati-infected mouse brains [28] , indicating potential behavioural alterations caused by either Toxocara species . Based on the affinity to the CNS as well as observed symptoms , T . canis is considered the causative agent of most human cases of NT . Nevertheless , T . cati has previously been associated with human cases of NT and the infection risk should not be underestimated [12 , 29] as environmental contamination with T . cati eggs is considered to be high due to the uncontrolled defecation behaviour of cats [30 , 31] . Due to the high infection risk in combination with the impact of Toxocara-infection on neurological and behavioural processes , it was aimed to assess the impact on neurobehaviour as well as the progression of neurological symptoms during the course of NT directly comparing T . canis- and T . cati-infected mice as models for human NT . Obtained data will aid in further characterization of possible influences of infection on the paratenic host and give a closer insight into the pathogenesis of NT as data about neurological disorders and cognitive deficits due to toxocarosis are still rare [32] .
Animal experiments were performed in accordance with the German Animal Welfare act in addition to national and international guidelines for animal welfare . Experiments were permitted by the ethics commission of the Institutional Animal Care and Use Committee ( IACUC ) of the German Lower Saxony State Office for Consumer Protection and Food Safety ( Niedersächsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit ) under reference numbers 33 . 9-42502-05-01A038 and 33 . 12-42502-04-15/1869 . T . canis and T . cati eggs were obtained from faeces of experimentally infected dogs and cats , respectively , kept at the Institute for Parasitology , University of Veterinary Medicine Hannover , for continuous maintenance of Toxocara spp . strains ( reference number 33 . 9-42502-05-01A038 ) . Eggs were purified by a combined sedimentation/flotation technique and allowed to embryonate for 4–5 weeks at 25°C with subsequent storage in tap water at 4°C until use . C57BL/6JRccHsd ( Harlan Laboratories , Horst , Netherlands ) female mice were obtained at approximately 4 weeks of age . Mice were allowed 5 days of acclimatization followed by 8 days of training for behavioural assessments prior infection ( see sections below ) . At 6 weeks of age , a total of 16 mice were orally infected with 2000 embryonated T . canis as well as 7 mice with 2000 T . cati eggs in a total volume of 0 . 5 ml tap water . The control group ( n = 7 ) received tap water only . Maintenance of mice included a 12/12 hours dark/light cycle , standard rodent diet ad libitum as well as daily assessment of physical condition . Behavioural assessments were conducted weekly starting one day prior infection ( -1 dpi ) until 97 days post infection ( dpi ) for T . canis-infected mice or until day 118 pi for T . cati-infected and control mice ( reference number 33 . 12-42502-04-15/1869 ) . Differences in duration of trials are based on delayed clinical symptoms in T . cati-infected mice as well as on severe progression and resulting ethical concerns in T . canis-infected mice [24] . Brain sections of all infection groups were stained exemplary with haematoxylin and eosin following termination of respective trials to demonstrate brain infection and resulting structural damage in Toxocara-infected mice . Mice were tested weekly regarding their physiological status as well as neurobehavioural alterations based on a modified Irwin Screen protocol [33] . Acclimatization prior examination was allowed for 30 min in a designated testing room separate from the maintenance room . Assessment was conducted in two phases: The first phase included a total of 17 general observations regarding appearance , health and motoric function . Briefly , body weight of mice was recorded as well as their responsiveness to being lifted up by the tail . Additionally , tail suspension and arousal upon transferring mice from the scale to the observation cage was evaluated . Within the cage , body position , tail elevation , respiratory rate , skin colour , coat appearance , eyes ( exophthalmos , ptosis as well as lacrimation ) and salivation were recorded as parameters for physiological status . As indicators for neurological and motoric dysfunctions , pelvic elevation , leg position and tumbling motion of mice were recorded in addition to startle response in combination with the ability to hear by evaluating responsiveness/excitability of mice to a sudden sound stimulus . The second test phase was conducted after the assessment of general activity ( see section below ) and included handling of mice which comprised balance in a moving cage , inquisitiveness upon a presented object , escaping upon gentle touch , vibrissae reflex , exploration and placing behaviour , forelimb placing reflex , “vertical screen test” ( ability to hold on to an upside down screen ) , righting reflex and general handling behaviour . Body temperature of mice was measured at the end of assessments . Regular behaviour was scored with “2” , deviating behaviour was scored higher or lower according to a defined scoring system . Deviation of the group mean from the standard score was calculated as percentage deviation from the standard score , if more than 25% of mice showed the respective alteration . Detailed description of the individual testing parameters as well as the corresponding scoring system is provided in S1 Appendix . The experimental cage was cleaned with 0 . 1% acetic acid following each individual mouse assessment . General activity was assessed by observing activity of mice in the experimental cage for three minutes , recording the activity and position every 10 sec . Position was determined by subdividing the cage into 12 squares along the wall ( 7 . 7 cm x 7 . 8 cm ) as well as the centre ( 7 . 7 cm x 23 . 4 cm ) . Activity was evaluated by determining the speed of movement ( inability to move , slow , normal and rapid ) as well as the following parameters: walking , grooming , sitting and sitting in a corner , sniffing , rearing , rearing against the wall as well as urinating/defecation . Besides normal activities , stereotypic behaviour like circling or head flicking as well as seizures were noted . Detailed information about assessed activities is listed in S2 Appendix . Functional deficits after Toxocara-infection were evaluated based on the approach described by Bouet et al . [34] . An approximately 0 . 2 x 0 . 2 cm piece of tape was applied to each forepaw of mice with subsequent immediate release in the experimental cage . Time was recorded upon first contact for left and right paws separately as well as upon removal of the tape for each paw . If 150 sec were exceeded without removal , mice were classified unable to sense or remove the piece of tape and recorded with 150 sec for subsequent analyses . The experimental cage was cleaned with 0 . 1% acetic acid following each individual mouse assessment . To assess the effect of Toxocara-infection on memory function as well as potential differences between T . canis- and T . cati-induced NT , mice were conditioned in a classic maze to find a food reward starting 8 days prior infection ( initial training ) . In addition to the weekly experimental evaluation , mice were allowed to find the food reward once a week ( continuous training ) . Mice were deprived of food for 2 hours before the maze test , which was recorded by camera and subsequently analysed in terms of duration until discovery of food reward and entries into dead-end arms . Changes of direction were also considered as dead-end entries . Mice were allowed three trials during experimental procedure with individual mean times used for final analyses . Mice were removed from the maze when 360 sec without completing the task were exceeded which was used in analyses as the maximum duration . The maze measured 36 cm x 28 cm with a height of 15 cm and included a series of vertical walls without ceiling . Vertical walls were removable and the floor of the maze was covered with plastic coated covering paper , which was disposed and replaced after each individual run . Maze structure is provided in S1 Fig . Obtained data was tested for statistical differences using an unpaired t-test with Welch's correction comparing the respective infection group to the control group each experimental day . If datasets did not pass normality tests , Mann-Whitney U-test was applied . Statistical analyses were conducted with GraphPad PrismTM software ( version 6 . 03 ) . The level of significance was set at α = 5% .
Severe progression of T . canis-infection was observed resulting in a total of 3 mice being euthanized prior termination of experiments due to poor general health . Additionally , two mice were euthanized within the T . cati-infection group before completing the assessments during the course of infection . Therefore , experiments were conducted with the following numbers of mice: control mice: n = 7 at day -1 to 118 pi; T . canis-infected mice: n = 16 at days -1 to 55 pi , n = 15 at day 62 pi , n = 14 at days 69 to 90 pi and n = 13 at day 97 pi; T . cati-infected mice: n = 7 mice at day -1 pi , n = 6 at days 6 to 111 pi and n = 5 at day 118 pi . Histological stains revealed cross sections of Toxocara larvae as well as vacuolization , meningitis , perivascular cuffs and haemosiderophages in T . canis- as well as in T . cati-infected mouse brains . Additionally , gitter cells were detected as a result of T . canis-induced NT . Exemplary sections are provided in Fig 1 . Regarding the body weight , T . canis-infected mice showed significantly higher body weights on day -1 and continuously lower weights than control mice starting day 27 pi . T . cati-infected mice also showed a significantly higher body weight day -1 and 13 pi , however , continuously lower body weights were not observed until day 62 pi . General progression of mean body weights of mice during the course of infection is provided in S2 Fig . Body temperature of uninfected control mice was consistent throughout the experiment in contrast to T . canis-infected mice , which showed significantly lower body temperature starting day 69 pi . T . cati-infected mice partially showed significantly higher body temperatures during the course of infection , except for significantly lower body temperatures than control mice on day 118 pi . Determined body temperatures are mostly considered within physiological range with Toxocara-infected mice being at the upper or lower limit . Mean body temperatures of all experimental groups are provided in S2 Fig . Overall , differences in general health as well as behaviour compared to uninfected controls ( Fig 2 ) were observed for T . canis- and T . cati-infected mice . Both infection groups showed an increased respiratory rate [parameter ( p ) 6] as well as bend body position ( p 4 ) with higher intensity in T . cati-infected mice during the course of infection . In T . canis-infected mice , neurological and motoric dysfunctions such as tumbling motions ( Fig 3; p 17 ) and ataxia ( p 15 , 16 ) in addition to disturbed balance ( p 18 ) and impaired righting reflexes ( p 26 ) were observed . Day 27 pi , T . canis-infected mice started to show the inability to hold on to an inverted screen during the vertical screen test ( p 25 ) and presented as less inquisitive ( p 19 ) and explorative ( p 22 ) during the chronic phase of infection ( starting day 48 pi ) . Less inquisitive ( p 19 ) and explorative ( p 22 ) behaviour was also observed in T . cati-infected mice ( Fig 4 ) ; however , the most striking feature were reduced reactions during fear- and flight-related assessments instead of neurological alterations . Particularly , assessments of excitability ( p 1 , 13 ) as well as escape parameters ( p 1 , 20 ) were affected . T . cati-infected mice additionally presented with ruffled as well as partially dirty coats ( p 8 ) starting day 62 pi . Even though all three experimental groups showed habituation to being handled ( p 27 ) during the course of infection as well as a reduced flight reaction when being touched with an object ( p 20 ) , these alterations of behaviour were observed with higher intensity and more frequently in Toxocara-infected mice . A detailed schematic overview of altered health as well as behavioural parameters is presented in Figs 2–4 . Both infection groups showed differences in activity patterns during the course of infection compared to the uninfected control . T . canis-infected mice revealed most severe differences to control mice mainly in walking activity , which was significantly more frequently observed starting day 34 pi . During the acute phase of infection from days 6–20 pi , T . canis-infected mice were also sitting more frequently than control mice . Rearing on cage walls was significantly reduced during almost the entire course of infection . In contrast , grooming was significantly more frequently observed in T . cati-infected mice compared to control mice starting day 27 pi , whereas T . canis-infected mice showed increased grooming behaviour solely day 27 pi . As similarly described for T . canis-infected mice , T . cati-infected mice were more frequently walking than uninfected controls starting day 34 pi until day 104 pi . Even though more frequent sitting was only observed day 20 pi , T . cati-infected mice were moving significantly slower than uninfected controls starting day 20 pi . In contrast , T . canis-infected mice were not significantly slower than uninfected controls until day 62 . Rearing as well as sniffing was irregularly significantly reduced in both infection groups over the course of infection . Unusual behaviour such as spatial disorientation and stereotypical circling was exclusively observed in T . canis-infected mice , starting day 27 pi and 34 pi , respectively . Urination/defecation as well as corner sitting was rare ( <5% ) and no preference for any position within the cage was detected in all experimental groups . An overview of most commonly conducted activities is provided in Fig 5 . Toxocara-infected mice showed altered sensory as well as motor function over the course of experimental infection with T . canis-infected mice showing more impaired reactions than T . cati-infected mice . Initial significantly increased time-to-contact of infected mice may be attributed to unsatisfactory adaptation to respective handling prior infection . During progression of infection , T . canis-infected mice showed significantly increased values days 41 and 48 pi as well as the remaining five examination days 69–97 pi . Similarly , T . cati-infected mice showed significantly increased time-to-contact compared to control mice as of day 76–118 pi with exception of day 111 pi . Consistent significantly longer times for removing the piece of tape from their paw were observed days 41–97 pi for T . canis-infected mice . Contrary , T . cati-infected mice only showed significantly increased time-to-remove days 83 and 90 pi . Detailed information about time-to-contact and time-to-remove are provided in Fig 6 . T . canis-infected mice required significantly more time to find the food reward than control mice starting day 27 pi . The mean duration upon finding the food reward increased over time , which was also observed regarding the number of entries into dead-end arms , starting to be significant as of day 20 pi . T . cati-infected mice required significantly more time than control mice 4 weeks later than T . canis-infected mice ( starting day 55 pi ) , but did not show continuous entries into dead-end arms . Significantly more dead-end arms than control mice were solely entered on day 62 pi as well as days 83–97 pi . Detailed progression of impaired memory functions is provided in Fig 7 .
Toxocara-induced NT may result in a variety of neurological symptoms such as behavioural alterations as well as neurological disturbances in paratenic hosts , including humans . To date , most studies are focusing on the influence of T . canis as the presumably more frequent causative agent of human toxocarosis . However , diagnosis remains challenging and the potential infection risk with T . cati larvae may be underestimated as it is generally considered high due to uncontrolled defecation habits of cats in public places like playground sandpits [29–31] . Nevertheless , studies regarding influence of T . cati larvae on behaviour and neurological functions as well as the direct comparison of T . canis- and T . cati-induced NT in terms of behavioural alterations and influence on memory function are rare . The presented study aids in filling the gap to provide a comprehensive picture of T . canis- and T . cati-induced NT and its progression in the mouse as model for human NT . Both infection groups showed differences in general health and neurobehavioural parameters compared to uninfected mice , demonstrating the need to consider T . cati as a causative agent of human NT . Nevertheless , more severe pathogenesis in T . canis- than in T . cati-infected mice was observed . Alterations were mostly detected during the chronic phase of infection with effects of T . cati-infection appearing delayed compared to T . canis-infection . These phenotypic results considerably complement data regarding more severe progression in T . canis-infection , with concurrent delayed onset in T . cati-infection in terms of clinical symptoms as well as transcriptional and histopathological alterations in the brain [22 , 24 , 35 , 36] . In contrast to the presented study , previously conducted behavioural studies showed no differences in general health of T . canis-infected mice compared to uninfected controls , but behavioural differences were observed [37] , possibly indicating adaptive manipulation . Differences to presented data may depend on different mouse strains as well as infection doses as these factors influence progression of infection [21 , 35 , 38 , 39] . These factors also demonstrate the importance of directly comparing effects of T . canis- and T . cati-induced NT under identical conditions to thouroughly characterize effects of the parasites on the host . In T . canis-infected mice , most parameters associated with neurological dysfunction were deviant from behaviour of uninfected controls . Tumbling motions were well observed starting day 20 pi indicating an early onset of neurological involvement due to early migration of larvae to the brain . Less severe neurological dysfunctions in T . cati-infected mice may be attributed to generally lower larval numbers in the brain , even though early migration of T . cati larvae to the brain has also been observed [21 , 22 , 26] . Circling as stereotypical behaviour was solely detected in T . canis-infected mice supporting the observation of severe neurological involvement . Even though mentioned neurological dysfunctions result in a selective advantage regarding uptake by the definitive host , behavioural alterations are most likely attributed to side-effects of pathology—rather than adaptive manipulation—and may be coincidentally beneficial for the parasite [38 , 40] . This is also supported by increased excitability levels ( indicated by response to a sudden sound stimulus ) in T . canis-infected mice until day 41 pi , which is not in accordance with adaptive manipulation as this behaviour would prevent exposure of infected mice in the environment . By contrast , T . cati-infection mostly provoked reduced excitability as well as flight reaction when being picked up by the tail resulting in increased probability of ingestion by predators within the environment . In addition to a milder course of infection , neurological disturbances like tumbling motions were observed only sporadically leading to the assumption of a better adaptation of T . cati larvae to the paratenic host . Therefore , contrary to T . canis-induced NT , T . cati-induced behavioural alterations are likely to be attributed to adaptive manipulation to facilitate transmission of larvae . Reasons for this clearly demonstrated deviant influence of the two Toxocara species remain unclear . It may be speculated that lower larval numbers in addition to less severe pathology in T . cati-infected mice provokes the parasite to induce alternative strategies to facilitate transmission of larvae from the paratenic to the definitive host . Reduced defence during physical handling was most intensely observed in T . cati- and to a lesser extent in T . canis-infected mice . In T . cati-infected mice , this observation might be attributed to generally reduced flight behaviour; whereas reduced defence in T . canis-infected mice is most likely a result of described pathology . Reduced defence in combination with reduced inquisitiveness and exploration in both Toxocara-infected groups are in accordance with results of activity assessments in terms of reduced rearing and increased walking . Tendency of incuriosity about their environment as well as high activity levels ( walking ) in familiar or novel areas results in higher vulnerability to predators [41 , 42]; however , it remains unclear if this behaviour is attributed to pathology or adaptive manipulation . Even though suggested host manipulation is not beneficial for Toxocara spp . within the human host , pathology accompanied by behavioural as well as neurological alterations such as confusion , ataxia or dementia severely influences quality of life of affected patients [32] . Wether human Toxocara-infections result in personality changes as well as delayed reaction times , increasing the risk of e . g . accidents as described for human Toxoplasma-infection [43 , 44] , has not been investigated yet . Interestingly , the extensive grooming behaviour in T . cati-infected mice during activity assays was observed prior to ruffled , partially dirty , appearance during the chronic phase of infection ( day 27 pi vs . day 62 pi ) . Ruffled appearance may be connected to effects of larval migration on general health as T . cati-infected mice also showed more severe deviations in body position compared to T . canis-infected mice , possibly indicating persistence of T . cati-larvae and pathological involvement throughout the skeletal muscle [22 , 35] . This persistence could also affect muscle function , nevertheless , T . cati-infected mice only showed sporadic inability to hold on to the vertical screen , whereas T . canis-infected mice showed marked inabilities during the course of infection , possibly attributed to reduced muscle strength [45 , 46] as well as poor general health . Furthermore , ruffled appearance of T . cati-infected mice and especially enhanced grooming may result from disturbances in hormone and neurotransmitter levels as these are inducing factors for altered self-grooming behaviour in rodents [47–49] . Self-grooming has therefore been discussed as an indicator for repetitive behaviour and consequently as potential model for human psychiatric disorders [48] . Disturbed levels of neurotransmitters e . g . depressed levels of dopamine are implicated in behavioural alterations in a variety of psychiatric diseases like schizophrenia and depression [50 , 51] . A potential correlation between Toxocara-seropositivity and neuropsychiatric disorders like schizophrenia , seizures and cognitive deficits in human patients has been discussed [32] . Reduced dopamine as well as serotonin and GABA levels in T . canis-infected mouse brains [50] highlight the need for a more detailed characterization of NT to confirm potential correlation . As a variety of neurotransmitters induce a reduction of rearing behaviour and also modulate stress- and anxiety-related behaviour in rodents [48 , 49] , recorded behavioural alterations may also be attributed to respective dysregulations . However , contrary to T . canis-infection , altered neurotransmitter levels during T . cati-induced NT have not been investigated yet . It may be speculated that T . cati-infection induces a different pattern of neurotransmitter dysregulation , which may account for observed marked increased grooming and reduced flight behaviour . Different neurotransmitter dysregulations may therefore be hypothesised as a key factor for differences in behavioural alterations during T . canis- and T . cati-induced NT . Sensorimotor assessment indicated reduced sensation on paws , which was also observed earlier in T . canis- than in T . cati-infected mice with motoric impairment being more extensive in T . canis-infected mice . As some of the core functions of the cerebellum are coordination and balance , structural damage influences those functions . Recorded motor incoordination as well as ataxia may therefore be assigned to the pathological alterations in cerebellar tissue observed in previous studies [21 , 52 , 53] . Less severe structural damage during T . cati- in comparison to T . canis-induced NT [22] may contribute to less severe impairment in T . cati-infected mice . The altered sensorimotor behaviour in both infection groups may particularly be attributed to occurring axonal damage which is associated with neuromotor and sensory dysfunction [22 , 24 , 54] . Motor impairment as well as sensory impairment and dysesthesia in Toxocara-positive humans [12 , 55–57] may therefore also be attributed to axonal damage , particularly requiring further investigations . As mice predominantly sense the strip on their paw even if delayed , reduced removal may be an effect of impaired balance of T . canis-infected mice . In contrast , prolonged removal times of T . cati-infected mice may not result from motoric impairment in terms of balance , but rather the neurological phenomenon “slower motion” already recorded during activity assays . The general slower motion in T . cati-infected mice also contributes most likely to significantly longer duration in the maze before finding the food reward compared to uninfected controls . Nevertheless , maze results show a significant effect on memory function in both infection groups . To our knowledge , this is the first report on T . cati larvae influencing memory function of the paratenic host . Again , T . canis-infected mice showed an earlier onset of impaired memory function as several attempts to fulfil the task starting day 20 pi were required . T . cati-infection had a less pronounced effect on memory and condition may have been transient as infected mice were not consistently requiring more dead-end entries than uninfected controls during the chronic NT phase as was the case with T . canis-infected mice . Transient behavioural alterations have been described in Toxocara- as well as in Toxoplasma-infected rodents , nevertheless , strains and infection doses were not comparable to the present study , which is an influential factor in disease progression [21 , 35 , 46 , 58] . Less pronounced alterations in T . cati-infected mice may be on account of fewer larvae in the brain , as Toxocara larval or Toxoplasma cysts numbers correlate with the degree of behavioural alteration [22 , 26 , 59] . Memory impairment in rodents may strongly influence survival factors like recalling resources and use of visual cues . As in the presented study , similar experiments provided evidence of memory impairment as T . canis-infected mice required significantly longer times than uninfected controls to locate vital resources like water in a given area after a preceded deprivation period [26] . Cerebellar lesions and damage in Toxocara-infected mice may contribute to impairments as parts of the cerebral cortex involved in cognitive regulation are connected to the cerebellum . Therefore , cerebellar integrity is required for cognitive functions [54] and may be influenced by the infection . Suspected neurodegeneration in Toxocara-infected mice [28] may particularly influence observed memory dysfunction as neurodegenerative cerebellar diseases in humans have been shown to influence explicit memory without affecting implicit memory [24 , 32 , 52] . Therefore , results obtained in mice may be comparable to observed memory impairment of Toxocara-positive human patients [7 , 10 , 11 , 13] as explicit memory , an intentional process to recall facts and experiences , is required to recall the location of the food reward . It may therefore be speculated that neurodegeneration constitutes an important part in human NT . Additionally , particular brain regions may be affected by larval persistence and adverse immune reactions . Damage of e . g . the hippocampus , which is implicated in development and maintenance of spatial learning as well as reduced exploratory behaviour [54] , may result in recorded alterations . However , a detailed characterization of brain areas involved during NT is not yet available . Such investigations are desirable to better comprehend behavioural or neurological alterations in the infected host . Correlation of learning impairment and Toxocara-seropositivity in children additionally highlights the need for further detailed characterization of NT [15 , 17] . The current results clearly show that T . cati-induced NT also influences behavioural parameters , even though to a lesser extent than T . canis-induced NT . It still may be assumed that neurological disturbances such as ataxia and balance incoordination in humans may mostly be attributed to the—extrapolated from the mouse model—likely more severe T . canis-infection . However , based on obtained results , memory impairment as well as cognitive dysfunction frequently observed in NT patients [7 , 10 , 11] may be induced by either Toxocara species . T . cati should therefore strongly be considered as a causative agent of human NT as well . Overall , the conducted study evidently demonstrates an earlier onset of severe neurological and behavioural alterations in T . canis-infected mice with more severe progression of disease than in T . cati-infected mice . Observed differences in behavioural and neurological alterations provide valuable hints towards varying key regulatory mechanisms during T . canis- and T . cati-induced NT , contributing to a broader understanding of NT , which will aid in determining targets for prevention as well as novel diagnostic and therapeutic approaches . | The worldwide occurring zoonotic roundworms Toxocara canis and T . cati may cause the so-called neurotoxocarosis ( NT ) in humans , possibly resulting in a variety of behavioural alterations . Comparable alterations to those in humans have been described in T . canis-infected mice , however , reports on T . cati-induced NT are rare . Therefore , the main causative agent of human NT remains unknown . For a better understanding and more detailed characterization of T . canis- and T . cati-induced NT , infected and uninfected mice were exposed to several examinations regarding behavioural alterations and cognitive dysfunctions . Even though both roundworm species share many characteristics , pathogenicity as well as larval migration in the animal model has been described as quite different . The current study additionally highlights differences between behavioural alterations in infected mice . T . canis-infection mostly provoked neurological alterations , whereas T . cati-infection resulted in reduced excitability and flight-related parameters . Exploratory behaviour was reduced in both infection groups in addition to a negative effect of infection on memory function . The presented work provides a valuable basis for further studies and highlights the need for further investigations concerning consequences and differences between T . canis- and T . cati-induced NT . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"cognitive",
"neurology",
"medicine",
"and",
"health",
"sciences",
"respiratory",
"infections",
"vertebrates",
"mice",
"neuroscience",
"learning",
"and",
"memory",
"animals",
"mammals",
"pulmonology",
"developmental",
"biology",
"cognitive",
"neuroscience",
"animal",
"behavior",
"cognition",
"memory",
"zoology",
"behavior",
"life",
"cycles",
"cognitive",
"impairment",
"rodents",
"neurology",
"biology",
"and",
"life",
"sciences",
"cognitive",
"science",
"amniotes",
"larvae",
"organisms"
]
| 2017 | Abnormal neurobehaviour and impaired memory function as a consequence of Toxocara canis- as well as Toxocara cati-induced neurotoxocarosis |
Chronic intestinal parasite infection is a major global health problem , but mechanisms that promote chronicity are poorly understood . Here we describe a novel cellular and molecular pathway involved in the development of chronic intestinal parasite infection . We show that , early during development of chronic infection with the murine intestinal parasite Trichuris muris , TGFβ signalling in CD4+ T-cells is induced and that antibody-mediated inhibition of TGFβ function results in protection from infection . Mechanistically , we find that enhanced TGFβ signalling in CD4+ T-cells during infection involves expression of the TGFβ-activating integrin αvβ8 by dendritic cells ( DCs ) , which we have previously shown is highly expressed by a subset of DCs in the intestine . Importantly , mice lacking integrin αvβ8 on DCs were completely resistant to chronic infection with T . muris , indicating an important functional role for integrin αvβ8-mediated TGFβ activation in promoting chronic infection . Protection from infection was dependent on CD4+ T-cells , but appeared independent of Foxp3+ Tregs . Instead , mice lacking integrin αvβ8 expression on DCs displayed an early increase in production of the protective type 2 cytokine IL-13 by CD4+ T-cells , and inhibition of this increase by crossing mice to IL-4 knockout mice restored parasite infection . Our results therefore provide novel insights into how type 2 immunity is controlled in the intestine , and may help contribute to development of new therapies aimed at promoting expulsion of gut helminths .
Gastrointestinal parasitic helminth infections are extremely prevalent , affecting nearly one quarter of the world population . Development of chronic infection , defined as the presence of adult worms in the host , results in severe morbidity and health problems and has been heavily linked with promotion of poverty in affected regions [1] . Current treatments involve the use of anti-helminthic drugs to kill the parasite , but this does not prevent rapid re-infection with worms and encounters problems with drug resistance . As infections with these intestinal parasites are usually chronic , it is likely that helminths are able to influence the immune system to prevent their expulsion . Therefore , understanding the cellular and molecular pathways that regulate the immune response during helminth infection will be crucial in identifying novel therapeutic targets for these poorly managed infections . A key cytokine that plays a multi-functional role in controlling immune responses is transforming growth factor beta ( TGFβ ) [2] . TGFβ can affect many different cell types , with data highlighting a crucial role for TGFβ in regulation of CD4+ T-cells , both dampening and promoting effector responses depending on the context of the immune response [3] , [4] . Importantly , although many cells can produce TGFβ , it is always made as an inactive complex that must be activated to produce biological function [5] . Thus , activation of TGFβ is a key regulatory step in controlling the function of TGFβ in the immune system . Given its importance in regulating diverse T-cell responses , it is not surprising that TGFβ plays a crucial role in the maintenance of immune homeostasis and prevention of autoimmunity . Thus , mice lacking TGFβ receptors in T-cells develop multi-organ inflammatory disease [6] , [7] and lack of TGFβ production by T-cells results in autoimmunity and colitis [8] . Interestingly , recent data has implicated TGFβ-like molecules produced by helminths in regulating immune responses during parasite infection [9] . However , the function of TGFβ during helminth infection and how it is regulated to control immune responses to intestinal parasites is poorly understood . Here we show that mice infected with the intestinal parasite Trichuris muris , a homologue of the human pathogen Trichuris trichuria [10] , display enhanced TGFβ signalling in CD4+ T-cells early during infection and that antibody-mediated blockade of TGFβ significantly reduces worm burden during the development of a chronic infection . . We find that integrin αvβ8 expressed by dendritic cells ( DCs ) , which we have previously shown to be a key pathway in activating TGFβ during intestinal homeostasis [11] , [12] , is required for early induction of TGFβ signalling in CD4+ T-cells during development of chronic helminth infection . Importantly , mice lacking integrin αvβ8 expression on DCs are completely protected from chronic infection , with this protection resulting from a specific early upregulation of a Th2-type immune response . Our results therefore provide novel insights into regulatory mechanisms of importance during chronic gastrointestinal parasite infection , and may help contribute to the development of new therapies aimed at promoting expulsion of helminth infection .
Development of a chronic parasite infection is believed to result from an inappropriate suppression of host immunity , although the exact molecular mechanisms governing these pathways remain unclear . Given the fundamental importance of CD4+ T-cells in regulating parasite infection and the key role for TGFβ in regulating many aspects of T-cell biology , we analysed TGFβ signalling in T-cells during development of a chronic infection with the helminth Trichuris muris . In C57BL/6 mice receiving 30 T . muris eggs , a dose shown previously to induce a chronic infection [13] , we observed a specific increase in phosphorylation of Smad 2/3 ( pSmad2/3 ) in mLN CD4+ T-cells , which is the initial signalling event triggered by engagement of TGFβ with its receptor [14] . This increase in TGFβ signalling was observed as early as day 3 post-infection , and was still evident at day 7 post-infection ( Figure 1A and B ) , before returning to levels seen in uninfected mice by day 14 post-infection ( Figure 1B ) . Similar early increases in CD4+ T-cell pSmad2/3 were also observed in cells taken from the lamina propria of the parasite's niche , the caecum and proximal colon ( Figure S1 in Text S1 ) . These data indicate that TGFβ signalling in CD4+ T-cells is an early hallmark of chronic T . muris infection . To directly examine the functional importance of TGFβ in the development of a chronic T . muris infection , we injected C57BL/6 mice with a TGFβ function-blocking antibody before and during infection . Interestingly , mice receiving TGFβ function-blocking antibody were significantly protected from worm infection ( Figure 1C ) . Thus , our data indicate that , during development of chronic infection , TGFβ plays an important role in promoting infection by the intestinal parasite T . muris . We next sought to determine the mechanisms responsible for enhanced TGFβ signalling and function during T . muris infection . One potential explanation for enhanced TGFβ signalling observed in CD4+ T-cells is enhanced activation of host latent TGFβ during infection . We have recently identified integrin αvβ8 , expressed by DCs , as a key activator of latent TGFβ in the intestine during immune homeostasis [11] , [12] . Thus , to determine the importance of this pathway in promoting TGFβ signalling in CD4+ T-cells during T . muris infection , we analysed T-cell responses in C57BL/6 control mice and mice lacking integrin αvβ8 on DCs ( Itgb8 ( CD11c-Cre ) mice ) [11] during infection . Interestingly , the increase in TGFβ signalling observed in CD4+ T-cells early during T . muris infection was significantly reduced in Itgb8 ( CD11c-Cre ) mice , with pSmad2/3 levels remaining similar to those observed in uninfected mice during the first week of infection ( Figure 2A and B ) . This integrin αvβ8-dependent induction of Smad2/3 phosphorylation was confirmed by Western blot analysis for pSmad2/3 in purified CD4+ T-cells from infected mice ( Figure 2C ) . In contrast , we did not observe any differences in pSmad2/3 induction in dendritic cells between control and Itgb8 ( CD11c-Cre ) mice ( Figure S2A and B in Text S1 ) , indicating that the integrin αvβ8-mediated TGFβ activation does not trigger autocrine TGFβ signalling in DCs during early infection . To directly test whether DCs produced enhanced levels of active TGFβ via expression of integrin αvβ8 during T . muris infection , we isolated DCs from control and Itgb8 ( CD11c-Cre ) mice and measured their ability to activate TGFβ using an established active TGFβ reporter cell line [15] . Indeed , we observed an enhanced ability of intestinal DC activation to produce active TGFβ early during the development of chronic T . muris infection , which was completely absent in DCs lacking expression of integrin αvβ8 ( Figure 2D ) . Thus , during development of chronic T . muris infection , enhanced TGFβ activation by integrin αvβ8 on DCs is important in triggering TGFβ signalling pathways in CD4+ T-cells . To determine whether TGFβ activation by integrin αvβ8 on DCs was functionally important during development of chronic infection with T . muris , we analysed worm numbers in control and Itgb8 ( CD11c-Cre ) mice infected with a chronic dose of T . muris eggs . Strikingly , Itgb8 ( CD11c-Cre ) mice were completely protected from chronic infection by T . muris at day 35 post-infection , with mice showing protection as early as day 14 post-infection ( Figure 2E ) . Indeed , protection from infection observed in Itgb8 ( CD11c-Cre ) mice was even more pronounced than that observed using antibody-mediated blockade of TGFβ function ( Figure 1C ) . It has been reported that expression of CD11c-Cre may drive recombination in a subset of CD4+ CD11clo activated T-cells [16] , and we have previously reported that integrin αvβ8 is expressed by CD4+ T-cells [11] . Thus , to test whether protection from infection in Itgb8 ( CD11c-Cre ) mice could be due to deletion of the integrin in T-cell subsets , we infected mice lacking integrin αvβ8 on T-cells via expression of CD4-Cre ( Itgb8 ( CD4-Cre ) mice ) [11] . In contrast to Itgb8 ( CD11c-Cre ) mice , Itgb8 ( CD4-Cre ) mice showed no protection from infection with T . muris ( Figure S3A in Text S1 ) and showed an identical parasite-specific IgG2a/IgG1antibody bias which is associated with development of a chronic infection ( Figure S3B in Text S1 ) . Taken together , these data suggest that integrin αvβ8-mediated TGFβ activation by DCs is essential in the promotion of chronic T . muris infection . We next sought to determine the mechanisms responsible for protection from infection in mice lacking the TGFβ-activating integrin αvβ8 on DCs . CD4+ T-cells are key in promoting expulsion of intestinal parasite infection , including T . muris [17] , and TGFβ signalling is triggered in these cells early during infection ( Figure 1A and B ) . However , recent evidence has proposed that novel innate lymphoid cells can play crucial roles in the expulsion of several parasite infections [18] , [19] , [20] , [21] . Thus , to determine the function of a CD4+ T-cell response in the expulsion of T . muris observed in Itgb8 ( CD11c-Cre ) mice , we first bred mice onto a C57BL/6 SCID background lacking all lymphocytes . In the absence of total lymphocytes , protection from infection was completely absent , with Itgb8 ( CD11c-Cre ) SCID−/− mice showing similar susceptibility to infection as control mice ( Figure 3A ) . To specifically test the role of CD4+ T-cells in protection from infection observed in Itgb8 ( CD11c-Cre ) mice , we depleted CD4+ T-cells using an anti-CD4 antibody ( Figure S4A in Text S1 ) . Absence of CD4+ T-cells restored susceptibility to infection in Itgb8 ( CD11c-Cre ) mice ( Figure 3B ) . Taken together , these results indicate that protection from infection in the absence of integrin αvβ8 expression on DCs is not via a direct effect of innate lymphoid cells , but driven by a classical CD4+ T-cell response , although a role for innate cells in initial priming cannot be ruled out . CD4+ Foxp3+ regulatory T-cells ( Tregs ) have been implicated in inhibiting immune responses to helminths [22] including some strains of T . muris [23] . Additionally , we have previously shown that integrin αvβ8-mediated TGFβ activation by specialised intestinal DCs is a crucial mechanism by which Foxp3+ Tregs are induced in the gut [12] , and that Itgb8 ( CD11c-Cre ) mice have reduced Foxp3+ Treg levels in their intestine [11] . Thus , one potential explanation for protection from infection in Itgb8 ( CD11c-Cre ) mice is that there is a reduced Treg response induced during infection in these mice . To address this possibility , we first directly assessed the role of Foxp3+ Tregs during development of chronic T . muris infection by using the DEREG mouse model on a C57BL/6 background , which allows specific ablation of Foxp3+ Tregs by injection of diphtheria toxin [24] . Despite robust depletion of Foxp3+ Tregs ( Figure S4B in Text S1 ) we did not see any enhanced ability of Foxp3+ Treg-depleted mice to expel worms ( Figure 3C ) . In agreement with a lack of role for Foxp3+ Tregs in the development of chronic T . muris infection , we did not see any enhancement of Foxp3+ Treg levels during the course of infection ( Figure 3D ) . Additionally , to directly assess whether reduced Foxp3+ Treg numbers Itgb8 ( CD11c-Cre ) mice was responsible for protection from infection , we rescued Treg numbers by adoptively transferred Foxp3+ Tregs from GFP-Foxp3 mice [25] prior to infection . Despite enhancement of Treg numbers in Itgb8 ( CD11c-Cre ) mice after adoptive transfer of GFP-Foxp3+ Tregs ( Figure S5 in Text S1 ) , Itgb8 ( CD11c-Cre ) mice were still highly protected from development of a chronic infection ( Figure 3E ) . Taken together , these data indicate that the protection from infection observed in Itgb8 ( CD11c-Cre ) mice is driven by CD4+ T-cells , but independently of Foxp3+ Treg cells . During development of a chronic infection with T . muris , mice develop a Th1-type immune response at the expense of a protective Th2-type response [13] . Thus , an alternative explanation for the expulsion of a normally chronic infection of T . muris by Itgb8 ( CD11c-Cre ) mice is that , in the absence of early CD4+ T-cell TGFβ signalling , mice produce a Th2-type response instead of the usual non-protective Th1 response . To test this possibility , we analysed the production of Th1 and Th2 cytokines during infection in control and Itgb8 ( CD11c-Cre ) mice . Strikingly , as early as 3 days post-infection , we observed a significant increase in production of the Th2 cytokine IL-13 , which was still elevated at 7 days post-infection ( Figure 4A ) . In contrast , although there was a slight enhancement of the Th1 cytokine IFNγ 3 days post-infection , this was not significantly different between control and Itgb8 ( CD11c-Cre ) mice ( Figure 4B ) . Control mice developed a marked enhancement in IFNγ production by day 18 post-infection , as expected during development of a chronic infection , and this was not observed in Itgb8 ( CD11c-Cre ) mice ( Figure 4B ) . Neither control nor Itgb8 ( CD11c-Cre ) mice produced any detectable IL-4 at any tested timepoint post-infection , a cytokine previously shown to be involved in protection from T . muris infection ( Figure S6 in Text S1 and data not shown ) . We next investigated the cellular source of the early IL-13 production in Itgb8 ( CD11c-Cre ) mice using flow cytometry . We observed a significant population of IL-13+ CD4+ T-cells within the intestinal lamina propria early during infection in Itgb8 ( CD11c-Cre ) which was not apparent in control mice ( Figure 4C ) . We also observed a slight increase in IFNγ+ lamina propria CD4+ T-cells in Itgb8 ( CD11c-Cre ) mice early post-infection; however , these levels were not significantly different from those seen in control mice ( Figure 4C ) . Interestingly , in mice treated with a TGFβ function-blocking antibody which resulted in protection from infection ( Figure 1C ) , we observed a similar increase in CD4+ T-cell IL-13 production , with no difference in IFNγ production observed ( Figure 4D ) . Furthermore , mice treated with TGFβ blocking antibody developed a skewed parasite-specific IgG1 response during infection , indicative of an enhanced type2 immune response ( Figure 4E ) . Taken together , these data indicate that TGFβ activation by DC-expressed integrin αvβ8 is important in controlling IL-13 production in CD4+ T-cells early during development of chronic infection . To test whether the enhanced production of IL-13 early during infection was responsible for expulsion of a chronic T . muris infective dose , we crossed the Itgb8 ( CD11c-Cre ) mice with C57BL/6 IL-4 knockout mice , which have previously been shown to lack the ability to generate an IL-4/13-mediated Th2 response during T . muris infection [26] . As both control mice and Itgb8 ( CD11c-Cre ) mice did not display production of IL-4 early during T . muris infection ( Figure S6 in Text S1 ) , these mice allowed us to test the role of the enhanced IL-13 response seen early during infection in Itgb8 ( CD11c-Cre ) mice . As expected , Itgb8 ( CD11c-Cre ) ×IL-4−/− mice no longer demonstrated an early IL-13 production in the intestinal CD4+ T-cells ( Figure 5A ) . Strikingly , Itgb8 ( CD11c-Cre ) ×IL-4−/− mice were completely susceptible to infection , with parasite burdens comparable to those seen in control mice ( Figure 5B ) . Taken together , these data indicate that lack of the TGFβ-activating integrin αvβ8 on DCs results in a heightened CD4+ T-cell Th2 immune response during T . muris infection which is responsible for rapid parasite expulsion .
Infection with intestinal helminths can result in either expulsion or development of chronic infection , often depending on the type of CD4+ T-cell response generated . Generally , a chronic infection results when inappropriate Th1 cytokine production occurs , as opposed to an inability of CD4+ T-cells to mount a response . Expulsion of the parasite relies on the production of Th2 cytokines , in particular IL-13 which drives a combination of cytokine-mediated expulsion mechanisms such as increased epithelial cell turnover in the intestine [27] , enhanced mucus production [28] and increased production of RELM-β [29] . Our data now demonstrate an essential role for TGFβ and the TGFβ-activating integrin αvβ8 expressed by DCs in promoting chronic intestinal parasite infection , using T . muris , a mouse model of the prevalent human parasite Trichuris trichuria . We observed that TGFβ signalling in CD4+ T-cells is triggered early during T . muris infection , and antibody-mediated blockade of TGFβ function significantly protects mice from infection . Mechanistically , we find that enhanced TGFβ signalling in T-cells during infection occurs via expression of the TGFβ-activating integrin αvβ8 on DCs and that lack of this integrin on DCs completely protects mice from infection due to an enhanced protective Th2 response . We have therefore identified a novel pathway that regulates Th2 immune responses in the gut that could potentially be targeted to upregulate host protective immune responses during gut parasite infection . Recent data suggest that in certain chronic parasite infections , induction of Foxp3+ Tregs is important in suppression of protective immunity and development of chronic infection [30] , [31] , [32] . Given the fundamental role of TGFβ in induction of Foxp3+ Tregs from naive CD4+ T-cells , and the fact that Itgb8 ( CD11c-Cre ) mice have previously been shown to have impaired induction of intestinal Foxp3+ Tregs [12] , we hypothesised that protection from T . muris infection observed in Itgb8 ( CD11c-Cre ) mice was due to reduced induction of Foxp3+ Tregs . However , when Foxp3+ Tregs were depleted before and during the course of infection no protection from infection was observed . Indeed , in contrast to Itgb8 ( CD11c-Cre ) mice , no enhancement of CD4+ T-cell IL-13 production was observed early during infection in Foxp3+ Treg-depleted mice ( Figure S7 in Text S1 ) . Additionally , in agreement with previous reports [23] we did not see a significant increase in Foxp3+ Tregs during T . muris infection . Instead , TGFβ activation by DC-expressed integrin αvβ8 appears important in suppression of IL-13 production by CD4+ T-cells early during T . muris infection . This is in agreement with previous data from in vitro studies , suggesting that TGFβ can downregulate expression of GATA-3 in T-cells ( a key transcription factor in promoting Th2 cell differentiation ) [33] , [34] . Indeed , recent data suggest that TGFβ-mediated induction of the transcription factor Sox4 is important in preventing GATA-3 transcription to drive Th2 development [35] . Furthermore , we only observed an early increase in CD4+ T-cell specific pSmad2/3 signalling during a chronic Th1-promoting low dose infection and not during an acute Th2 promoting high dose infection in C57BL/6 mice ( Figure S8 in Text S1 ) Thus , our data suggests that activation of TGFβ by integrin αvβ8 early during T . muris infection is important in suppression of protective Th2 cell development , which leads instead to production of an inappropriate Th1 response and development of chronic infection . Although we did not detect any IL-4 production in Itgb8 ( CD11c-Cre ) mice during infection , given that we crossed these mice to IL4 KO mice to eliminate enhanced IL-13 production by T-cells , we cannot rule out a potential role for low level production of IL-4 ( below our limits of detection ) in protection from infection . In addition to effects on T-cells , TGFβ has wide-ranging effects on multiple other immune cell types [36] . Recent reports have highlighted an important role for novel innate lymphoid cells in promoting protective type 2 immunity during certain parasite infections [18] , [19] , [20] , [21] . Hence , it could be postulated that protection from infection seen in the absence of integrin αvβ8 results from an enhanced innate lymphoid cell response . However , protection from chronic T . muris infection observed in Itgb8 ( CD11c-Cre ) mice did not correlate with enhanced type 2 cytokine production from any cell types apart from CD4+ T-cells ( data not shown ) , and protection from infection was completely dependent on CD4+ T-cells . Thus , although innate lymphoid cell depletion would be required to definitively rule out their role in this enhanced Th2 response , it appears unlikely that lack of integrin-mediated TGFβ activation by DCs is promoting expulsion of the parasite via effects on non-CD4+ T-cells . Given the crucial importance of TGFβ in regulating CD4+ T-cell responses , our current model is that TGFβ activated by DCs acts directly in CD4+ T-cells to regulate type 2 responses during T . muris infection . A recent study by Heitmann et al . ( 2012 ) suggests that CD4+ T-cell type 2 responses can be regulated via TGFβ signalling in DCs [36] . Thus , mice expressing a dominant negative construct of the TGFβ receptor II in myeloid cells ( hence are refractory to TGFβ signalling ) display enhanced Th2 responses during infection with the helminth Nippostrongylus brasiliensis [36] . However , we observed no difference in pSmad2/3 induction in DCs from control versus Itgb8 ( CD11c-Cre ) mice early during infection ( Figure S2 in Text S1 ) . Thus , our data indicate that activation of TGFβ by integrin αvβ8 on DCs does not regulate Th2 cells indirectly via autocrine TGFβ signalling during T . muris infection . Velhoden et al 2008 [37] have demonstrated that mice expressing a dominant negative TGFβ receptor specifically on CD4+ T-cells ( CD4-DN-TGFβRII mice , thus T-cells are refractory to TGFβ ) are more susceptible to infection with T . muris using an acute model of infection . This finding initially appears to conflict with our data , as we demonstrate that both antibody-mediated blockade of TGFβ and lack of the TGFβ activating integrin αvβ8 on DCs promotes expulsion of the parasite . However , recent data suggests that CD4-DN-TGFβRII mice display high levels of IFNγ level during intestinal helminth infection [38] , [39] which , given the known role of IFNγ in promoting chronic T . muris infection [13] , may explain the enhanced levels of infection observed in CD4-DN-TGFβRII mice . An important question that remains are which specific subset of intestinal DCs are involved in modulating CD4+ T-cells to suppress Th2 responses ? Although a functionally important gut population of CD11c+ T-cells does exist [16] , which may be targeted in our CD11c-cre knock-out system , mice lacking integrin αvβ8 on T-cells ( Itgb8 ( CD4-Cre ) mice ) were completely susceptible to T . muris infection ( Figure S3 in Text S1 ) . These data indicate that it is indeed an αvβ8-expressing DC population ( or a related CD11c+ mononuclear phagocyte population ) that is important in inhibiting Th2 responses in this infection . An important DC subset likely involved during infection are the migratory CD103+ DC [40] , as we have previously demonstrated that this cell subset expresses high levels of integrin αvβ8 [12] . We have observed some integrin αvβ8 expression on the CD11c+ CD103- DC subset in the colon [12] , which has been suggested to include both DCs and macrophage-like cell populations [41] . However , although some subsets of CD11c+ CD103- intestinal cells have been shown to migrate to lymph nodes to modulate T-cell responses [42] , a large population do not normally migrate . Of note , we did not observe any alteration in the levels of αvβ8 expression on either CD103+ or CD103- LILP subset during the development of a chronic infection ( Figure S9 in Text S1 ) . Therefore , the exact DC population involved in downregulation of Th2 responses via integrin αvβ8 remains to be determined . Nevertheless , this key role for DC-expressed integrin αvβ8 in modulating Th2 responses , in addition to its previous essential roles in the induction of Foxp3+ Tregs [12] and Th17 cells [43] , places DC-expressed integrin αvβ8 as a key orchestrator of CD4+ T-cell immunity . In summary , we have highlighted an important cellular and molecular pathway by which the TGFβ-activating integrin αvβ8 expressed by DCs represses protective Th2 immunity during intestinal parasite infection with T . muris . Thus , given the poor treatments currently available for chronic parasite infection , further work should focus on the potential for targeting integrin αvβ8 therapeutically to enhance protective immunity during Trichuris infection . Additionally , whether the pathway is involved in the development of other chronic infections and Th2-associated disease is the focus of current studies .
C57 BL/6 mice were purchased from Harlan Laboratories . Mice lacking integrin αvβ8 on DCs via expression of a conditional floxed allele of β8 integrin in combination with CD11c-Cre ( Itgb8 ( CD11c-Cre ) mice ) [11] , DEREG mice [24] , GFP-Foxp3 mice [25] and IL-4−/− mice [44] , all on a C57BL/6 background , have been previously described . All mice were maintained in specific pathogen-free conditions at the University of Manchester and used at 6 to 8 weeks of age . All animal experiments were performed under the regulations of the Home Office Scientific Procedures Act ( 1986 ) , specifically under the project licence PPL 40/3633 . The project licence was approved by both the Home Office and the local ethics committee of the University of Manchester . The techniques used for maintenance and infection of T . muris were as previously described [45] Mice were orally infected with 20–30 eggs for a low-dose infection and 150 for an acute infection . Worm burdens were assessed by counting the number of worms present in the caecum as described previously [45] . To block TGFβ , mice were injected i . p with 0 . 5 mg of anti-TGFβ blocking antibody ( clone 1d11 . 16 . 8 ) ( BioXCell , NH , USA ) or control IgG1 every 2 days from 4 days prior to infection . CD4+ cells were depleted via i . p . injection of 2 mg anti-CD4 antibody ( YTS 191 ) 47 every 2 days from 6 days prior to infection . Foxp3+ Tregs were depleted in DEREG mice as described [24] , via i . p . injection of 200 ng diphtheria toxin ( Merck ) every 2 days from 2 days prior to infection . Spleens were removed from Foxp3GFP mice , disaggregated and red blood cells lysed . Cells were blocked with anti-FcγR antibody and labelled with anti-CD4 antibody ( clone GK1 . 5; eBioscience ) before sorting for CD4+ , GFP+ populations using a FACS Aria . Cell purity in all experiments was >99 . 8% . Mice were injected i . p . with 0 . 5×106 cells 3 days prior to infection . mLNs were excised from mice and incubated with shaking for 20 min at 37°C in RPMI with 0 . 08 U/ml liberase blendzyme 3 ( Roche ) or 1 mg/ml collagenase VIII and 50 U/ml DNAseI , respectively . Cells were blocked with anti-FcγR antibody ( clone 24G2; eBioscience ) before enrichment using a CD11c enrichment kit and LS MACS column ( Miltenyi Biotec ) . Enriched DCs were labelled with anti-CD11c antibody ( clone N418; eBioscience ) and sorted using a FACS Aria . In all experiments , subset purity was >95% . DCs were incubated with mink lung epithelial cells transfected with a plasmid containing firefly luciferase cDNA downstream of a TGFβ-sensitive promoter [15] in the presence of 1 µg/ml lipopolysaccharide . Co-cultures were incubated overnight in the presence of 80 µg/ml control mIgG or anti-TGFβ antibody ( clone 1d11 ) and luciferase activity detected via the Luciferase Assay System ( Promega ) according to manufacturer's protocol . TGFβ activity was determined as the difference in luciferase activity between control mIgG-treated samples and samples treated with anti-TGFβ antibody . Mesenteric lymph nodes ( mLNs ) were removed from mice and disaggregated through a 100 µm sieve . Caecum and proximal colon were excised and lamina propria lymphocytes were prepared essentially as described [46] with slight modification in the tissue digestion step ( digestion medium used was RPMI with 10% Foetal calf serum , 0 . 1% w/v collagenase type I and Dispase II ( both Invitrogen ) , and tissue was digested for 30 min at 37°C ) . Cell suspensions were blocked with anti-FcγR antibody ( clone 24G2; eBioscience ) before labelling with antibodies specific for CD4 ( clone GK1 . 5; eBioscience ) , Foxp3 ( clone FJK-16s; eBioscience ) , IL-13 ( clone eBiol13A; eBioscience ) , IFNγ ( clone XMG1 . 2; eBioscience ) or p-Smad 2/3 ( Santa Cruz ) . For pSmad2/3 staining , an Alexa Fluor 594-labelled donkey anti-goat secondary antibody was used ( Invitrogen ) . All samples were analysed on a FACS LSRII . mLN and LILP cells were prepared as described above before incubating with 50 ug/ml of concavelin A or T . muris excretory/secretory ( E/S ) antigen for 48 hours . Cell-free supernatants were analysed for cytokine production via cytometric bead array ( BD ) or paired ELISA antibodies ( anti- IFNγ , clone XMG1 . 2 and biotin anti- IFNγ , clone R4-6A2; anti-IL-13 , clone eBio13A and biotin anti-IL-13 , clone eBio1316H and anti-IL-4 , clone 11B1and biotin anti-IL-4 , clone BVD6-2462 ( eBioscience ) . For intracellular cytokine analysis cells were incubated for 12 hours with 50 ug/ml T . muris E/S antigen followed by PMA/ionomycin stimulation for 4 hour with addition of monensin for the final 3 hours . Cells were then stained with antibodies against IL-4 , IL-13 and IFNγ using the eBioscience Foxp3 permibilization kit according to the manufacturer's instructions . CD4+ T-cells were isolated from mLN via negative selection using a CD4+ T-cell isolation kit ( Miltenyi Biotec ) during the course of a chronic T . muris infection and lysed using 1% Triton-X100 in Tris buffer ( 50 mM Tris-HCl , 150 mM NaCl pH 7 . 4 ) plus 5 mM EDTA , 20 µg/ml leupeptin and aprotinin , 0 . 5 mM AESF and 2 mM NaVO3 . Lysates were analysed by Western blot with antibodies to detect p-Smad 2/3 ( Millipore ) and β-actin ( Sigma Aldrich ) , using the Invitrogen Nupage gel system according to manufacturer's instructions . Total RNA was purified from sorted DC subsets using an RNAeasy minikit according to manufacturer's protocol ( Qiagen ) . RNA was reverse transcribed using Oligo dT primers , and cDNA for specific genes detected using a SYBR green qPCR kit ( Finnzymes ) Gene expression normalised to HPRT expression . HPRT Forward: GCGTCGTGATTAGCGATGATGAAC , HPRT Reverse: GAGCAAGTCTTTCAGTCCTGTCCA , Integrin β8 Forward: GGGTGTGGAAACGTGACAAGCAAT , Integrin β8 Reverse: TCTGTGGTTCTCACACTGGCAACT . Results are expressed as mean ± SEM . Where statistics are quoted , 2 experimental groups were compared using the Student's t-test for non-parametric data . Three or more groups were compared using the Kruskal–Wallis test , with Dunn's multiple comparison post-test . P≤0 . 05 was considered statistically significant . | Infection with intestinal parasitic worms is a major global health problem , with billions of people infected world-wide . Often these worms ( known as helminths ) develop a long-lasting chronic infection , due to failure of the host to mount the correct type of immune response that would normally expel the parasite . However , how the immune system is controlled leading to chronic helminth infection is not well understood . Here we identify a novel pathway of importance in the development of chronic helminth infection . Using a model parasite which infects mice , we find that a protein called transforming growth factor beta ( TGFβ signals to T-cells early during the development of chronic infection and that blocking this signal protects mice from infection . We have also uncovered a key pathway and cell type that controls TGFβ function during development of chronic infection . When a protein called integrin αvβ8 is absent from dendritic cells of the immune system , TGFβ is no longer activated to signal to T-cells and mice are able to mount a protective ( type 2 ) immune response resulting in worm expulsion . Our findings therefore provide new insights into how chronic infections develop and identify potential molecular targets for the prevention of chronic helminth infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2013 | Loss of the TGFβ-Activating Integrin αvβ8 on Dendritic Cells Protects Mice from Chronic Intestinal Parasitic Infection via Control of Type 2 Immunity |
The surface coat of Trypanosoma cruzi is predominantly composed of glycosylphosphatidylinositol-anchored proteins , which have been extensively characterized . However , very little is known about less abundant surface proteins and their role in host-parasite interactions . Here , we described a novel family of T . cruzi surface membrane proteins ( TcSMP ) , which are conserved among different T . cruzi lineages and have orthologs in other Trypanosoma species . TcSMP genes are densely clustered within the genome , suggesting that they could have originated by tandem gene duplication . Several lines of evidence indicate that TcSMP is a membrane-spanning protein located at the cellular surface and is released into the extracellular milieu . TcSMP exhibited the key elements typical of surface proteins ( N-terminal signal peptide or signal anchor ) and a C-terminal hydrophobic sequence predicted to be a trans-membrane domain . Immunofluorescence of live parasites showed that anti-TcSMP antibodies clearly labeled the surface of all T . cruzi developmental forms . TcSMP peptides previously found in a membrane-enriched fraction were identified by proteomic analysis in membrane vesicles as well as in soluble forms in the T . cruzi secretome . TcSMP proteins were also located intracellularly likely associated with membrane-bound structures . We demonstrated that TcSMP proteins were capable of inhibiting metacyclic trypomastigote entry into host cells . TcSMP bound to mammalian cells and triggered Ca2+ signaling and lysosome exocytosis , events that are required for parasitophorous vacuole biogenesis . The effects of TcSMP were of lower magnitude compared to gp82 , the major adhesion protein of metacyclic trypomastigotes , suggesting that TcSMP may play an auxiliary role in host cell invasion . We hypothesized that the productive interaction of T . cruzi with host cells that effectively results in internalization may depend on diverse adhesion molecules . In the metacyclic forms , the signaling induced by TcSMP may be additive to that triggered by the major surface molecule gp82 , further increasing the host cell responses required for infection .
The kinetoplastid protozoan Trypanosoma cruzi is the etiologic agent of Chagas’ disease or American Trypanosomiasis which infects 6–7 million people worldwide , mostly in Latin America [1] . In the last few decades , the disease has spread to non-endemic regions , such as the United States and Europe , posing a new worldwide challenge [2] . T . cruzi is transmitted to humans by hematophagous triatomine vectors , that during blood meals , deposit feces containing the infective parasite forms , which enter the host through a lesion in the skin or mucosal surfaces . Other means of contamination include blood transfusion , congenital transmission and oral infection . In recent years , there have been frequent outbreaks of acute Chagas [3–8] . Host cell invasion is a crucial step for the establishment of T . cruzi infection . The first step of invasion is the adhesion of trypomastigotes to host cells by different surface molecules such as glycoproteins and proteases [9] . Among the most abundant surface molecules are glycoproteins anchored by the glycosylphosphatidylinositol ( GPI ) moiety and the GPI-related complex glycoinositolphospholipids [10 , 11] . Abundantly expressed GPI-anchored surface glycoproteins , encoded by multigene families , include mucins , mucin associated surface proteins ( MASP ) and trans-sialidases ( TS ) [12–14] . These genes comprise approximately 17% of protein-coding genes in the T . cruzi genome and are involved in host-parasite interactions [12 , 14–16] . Genomic comparison of the human parasites T . cruzi , Trypanosoma brucei and Leishmania spp revealed that most genes are common and arranged into syntenic regions [15 , 17 , 18] . However , genes encoding surface proteins are highly divergent , which is compatible with specific adaptations and survival strategies . Among these parasites , most species-specific genes encode surface antigens , such as variable surface glycoproteins ( VSG ) in T . brucei , TS and mucins in T . cruzi and promastigote surface proteases in Leishmania spp [15 , 17] . While the major components of T . cruzi surface coat are well characterized , little is known about less abundant surface proteins and their roles in infection and transmission . Recently , low abundance surface proteins ( SAP , TcTASV and DGF-1 ) have been characterized [19–24] . Screening of T . brucei cDNA libraries allowed the identification of a minor surface component of procyclic forms named Procyclic Surface Specific Antigen-2 ( PSSA-2 ) [25] . PSSA-2 cDNA was isolated and predicted to encode a membrane-spanning protein with a C-terminal domain containing proline-rich tandem repeats . Fragoso et al . [26] demonstrated the importance of this cytoplasmic tail in targeting PSSA-2 to the plasma membrane and suggested that PSSA-2 could function as a sensor to transmit signals from the tsetse fly to the parasite . The PSSA-2 null mutant was fully competent to establishing midgut infections in tsetse , but was defective in colonizing the salivary glands and in producing infectious metacyclic forms [26] . In a previous study , we identified PSSA-2 orthologous proteins in T . cruzi epimastigotes and metacyclic trypomastigotes by tandem proteomic analysis while analyzing membrane enriched fractions [27] . In our study , we focused on characterizing PSSA-2 orthologous genes in T . cruzi , as well as studying their expression and potential functions in host parasite interactions . In the course of this study , we decided to name this new protein family the T . cruzi Surface Membrane Protein ( TcSMP ) .
All experiments involving animals were carried out under Brazilian National Committee on Ethics Research ( CONEP ) ethics guidelines , which are in accordance with international standards ( CIOMS/OMS , 1985 ) . The protocol was approved by the Ethical Committee of the Universidade Federal de São Paulo for animal experimentation ( permit number: CEP 1877/08 ) . Trypanosoma species used in this study were: T . cruzi [clones CL Brener ( CLB ) and Dm28c , and CL strain] , T . cruzi marinkellei and T . brucei rhodesiense . T . cruzi epimastigotes were maintained in axenic cultures at 28°C in liver-infusion tryptose ( LIT ) medium [28] , supplemented with 10% heat-inactivated fetal calf serum ( Vitrocell , Brazil ) . For in vitro metacyclogenesis , epimastigotes of the CL strain were harvested at early stationary phase ( 5–7 days ) and allowed to differentiate in TC 100 medium ( Vitrocell , Brazil ) . Metacyclic trypomastigotes were purified by chromatography on a DEAE-cellulose column ( Sigma-Aldrich , St Louis , MO ) , as previously described [29] . Tissue culture trypomastigote ( TCT ) was obtained by differentiation of intracellular amastigotes in Vero cell monolayers infected with metacyclic trypomastigotes . Extracellular amastigotes were obtained by TCT differentiation in LIT medium as previously described [30] . Epimastigotes from T . cruzi marinkellei were maintained in LIT medium supplemented with 15% FCS at 28°C . Procyclic forms of T . brucei rhodesiense YTAT 1 . 1 were cultured in semi-defined medium ( SDM-79 ) supplemented with 10% heat-inactivated fetal bovine serum ( FBS ) at 27°C . Vero cells and human epithelial HeLa cells ( Instituto Adolfo Lutz , São Paulo , Brazil ) were grown at 37°C in a 5% CO2 humidified atmosphere in DMEM medium ( Sigma-Aldrich , St Louis , MO ) supplemented with 10% fetal bovine serum , 10 μg/mL streptomycin ( Sigma-Aldrich , St Louis , MO ) , 100 U/mL penicillin ( Sigma-Aldrich , St Louis , MO ) and 40 μg/mL gentamicin ( Sigma-Aldrich , St Louis , MO ) . Cell invasion assays were performed by seeding CL strain metacyclic forms onto 24-well plates containing 13-mm diameter round glass coverslips coated with 1 . 5 x 105 HeLa cells . The multiplicity of infection ( MOI ) was 10 . After 1 h of incubation with parasites , the duplicate coverslips were washed in PBS , fixed in Bouin solution , stained with Giemsa , and sequentially dehydrated in acetone , a graded series of acetone:xylol and xylol . The number of intracellular parasites was counted in 250 cells . Binding of the recombinant proteins TcSMP or GST to target cells was determined by ELISA . HeLa cells , grown in 96-well microtiter plates , were fixed with 4% paraformaldehyde in PBS for 30 min , washed and blocked with PBS containing 2 mg/ml BSA ( PBS/BSA ) . Following 1 h incubation with the recombinant protein in PBS/BSA , cells were incubated sequentially with anti-TcSMP or anti-GST antibodies , and peroxidase-conjugated anti-rabbit IgG , all diluted in PBS/BSA . The final reaction was revealed by o-phenylenediamine , and the absorbance was read at 490 nm in ELx800TM absorbance microplate reader ( BioTek , Winooski , VT ) . HeLa cells were grown overnight in DMEM with 10% fetal bovine serum on ibidi multichamber dishes ( Hi-Q4 , ibidi ) and incubated at 37°C for 30 min with Fluo-4 Direct 2x ( Invitrogen ) diluted 1:1 in HBSS solution ( Hank’s Balanced Salt Saline ) . Fluo-4 is a fluorescence indicator that binds intracellular free calcium ions . It presents a non-fluorescent AM grouping ( acetoxymethyl ester ) in its structure that when internalized , is cleaved by intracellular esterases and released to bind to cytoplasmic calcium . The emitted fluorescence intensity ( 500–550 nm ) is related to the concentration of cytoplasmic calcium [31] . Images were captured with a microscope ( Leica , Wetzlar , Germany ) equipped with a HCX PL APO 40X dry objective 0 . 85 numerical aperture . The Fluo-4 probe was excited at lEx = 485/15 nm and light emission was detected at lEm = 535/48 nm . Images were acquired at 2 sec intervals for 8 min ( 16 bit ) . The first two minutes corresponded to the basal fluorescence before application of the stimulus ( recombinant TcSMP or GST ) . Fluorescence intensity was analyzed and normalized with reference to the basal fluorescence using Leica LAS AF software ( Leica , Wetzlar , Germany ) . Genomic DNA was isolated from 1 x 108 epimastigotes of the clone CL Brener by homogenization in TELT buffer ( 50 mM Tris , pH 8 . 0; 62 . 5 mM EDTA , pH 9 . 0; 2 . 5 M LiCl; 4% Triton X-100; 20 μg/mL RNase ) , followed by extraction with phenol/chloroform and chloroform and precipitation with absolute ethanol . The pellet was washed with 70% ethanol , air-dried and suspended in TE ( 1 mM EDTA , pH 8 . 0; 10 mM Tris-HCl , pH 8 . 0 ) . Fifteen micrograms of genomic DNA were digested with 10 U of BamHI , BglII , EcoRI , EcoRV , HaeIII , HindIII , KpnI , PstI , SmaI or XhoI restriction enzyme ( Invitrogen ) , and the fragments were resolved on a 0 . 8% agarose gel , stained with ethidium bromide ( 0 . 5 μg/mL ) and photographed under UV light . Gels were sequentially treated with depurination solution ( 0 . 25 M HCl ) for 45 min , denaturation solution ( 0 . 5 M NaOH; 1 M NaCl ) for 20 min and neutralization solution ( 1 M Tris-base; 0 . 5 M NaCl ) for 20 min and the DNA transferred to nylon membranes in 20x SSC for 2h . Membranes were prehybridized in the hybridization solution ( 50% formamide , 5x SSC , 5x Denhardt’s solution [Invitrogen] , 0 . 1 mg/mL salmon sperm DNA , 0 . 5% SDS , 5 mM EDTA ) for 2 h at 42°C and incubated in the hybridization solution containing 32P -labeled TcSMP derived probe for 16 h at 42°C . Membranes were washed twice in solution containing 2 x SSC , 0 . 1% SDS and 0 . 1% sodium pyrophosphate for 30 min at 42°C followed by two additional 30 min washes in 0 . 1 x SSC , 0 . 1% SDS and 0 . 1% sodium pyrophosphate at 56°C . Finally , membranes were exposed to X-ray film in light-tight cassettes at -70°C . The TcSMP probe ( clone 23C ) shares 99% identity with other TcSMP sequences deposited on TriTrypDB . T . cruzi chromosomal DNA was resolved by pulsed-field gel electrophoresis ( PFGE ) in a Gene Navigator System ( Pharmacia , Amersham , GE Healthcare Life Sciences ) using a hexagonal electrode array as previously reported [32] . Plugs of agarose-embedded chromosomal DNA were prepared from epimastigotes of the clone CL Brener as previously described [32] . After electrophoresis , DNA was stained with ethidium bromide ( 0 . 5 μg/mL ) , transferred to nylon filters and hybridized as described above . To clone TcSMP genes from T . cruzi clone CL Brener , four oligonucleotide primers ( P1F , P2F , P3R and P4R ) were designed based on conserved regions of sequences deposited in the TriTrypDB ( S1 Fig ) . PCR amplification was carried out with genomic DNA as follows: initial denaturation ( 5 min at 94°C ) , 40 cycles of denaturation ( 30 s at 94°C ) , annealing ( 1 min at melting temperature ) and elongation ( 30 s at 72°C ) . PCR products were resolved in 0 . 8% agarose gels and purified using the Wizard SV Gel and PCR Clean-up System ( Promega , Madison , WI ) . Purified amplicons were cloned into the pGEM-T Easy vector ( Promega , Madison , WI ) , transformed into Escherichia coli DH5α strain and sequenced by the dye terminator method on an ABI PRISM 3130xl Genetic Analyzer ( Applied Biosystems , Foster City , CA ) . Similarity searches using BLASTN and BLASTP [33] algorithms were carried out to search for TcSMP nucleic acid and protein sequences of T . cruzi in the TriTrypDB genomic resource ( http://www . tritrypdb . org ) and GenBank . Alignment was performed using sequences from clone CL Brener excluding the truncated sequences . Nucleotide and deduced amino acid sequences were aligned using ClustalW [34] and manually adjusted . The search for orthologous genes in other trypanosomatids was performed using BLASTn and BLASTp algorithms . The TcSMP sequences from T . cruzi ( clones CLB , Sylvio X/10 and PCR amplified products from CL Brener ) , T . cruzi marinkellei , T . brucei ( strains Lister 427 and TREU927 ) , T . brucei gambiense , T . congolense and T . vivax were aligned using the MUSCLE algorithm [35] and manually inspected with Seaview ( 4 . 0 version ) [36] . Truncated sequences were excluded from ML analysis . The TcSMP phylogenetic analysis was inferred with maximum likelihood ( NJ ) method , using the LG substitution model with a bootstrap value of 1 , 000 replicates . Gene synteny analysis was carried out between chromosomes containing TcSMP orthologous genes from different trypanosomatids . Regions were aligned in pairs in the BLASTN program and analyzed by the ACT program . Analyses were conducted via TBlastN using the Artemis Comparison Tool ( ACT ) [37] . The prediction of functional domains and posttranslational modifications was performed using bioinformatics tools available at the ExPASy proteomic server ( http://www . expasy . org ) . Signal peptide and signal anchor were determined using SignalP3 . 0 and SignalP4 . 0 . SignalP3 . 0 also reports the probability of signal anchor , which is also named uncleaved signal peptide [38] . Transmembrane helices , O- , N-glycosylation and phosphorylation sites were determined using TMHMM , NetOGlyc , NetNGlyc and NetPhos software , respectively . A 311-bp EcoRI fragment from clone 23B ( see S5 Fig ) was inserted into pGEX-1λT to produce the TcSMP-GST fusion protein . E . coli BL21 bacteria transformed with this construct were grown in LB medium , induced with 1 mM isopropyl β-D-1-thiogalactopyranoside ( IPTG ) for 3 h at 37°C and harvested by centrifugation . Cells were suspended in PBS containing protease inhibitors and sonicated for 10 min on ice . Extracts were centrifuged at 1 , 600 x g for 20 min at 4°C to separate the insoluble proteins ( pellet ) from the soluble proteins ( supernatant ) . The pellet was suspended in Laemmli’s sample buffer and resolved on 10% SDS-PAGE . Gels were stained with ice-cold 250 mM KCl and the bands corresponding to the recombinant protein were excised from the gels . Gels slices were dialyzed against ammonium bicarbonate for 48 h at 4°C under agitation followed by 36 h at 4°C against distilled water . Purification was checked by SDS-PAGE stained with colloidal Coomassie Blue and immunoblotting . Purified protein was quantified with Coomassie Plus ( Pierce , Thermo Fisher Scientific ) in 96-well plates at 620 nm . Anti-TcSMP polyclonal antibodies were obtained by immunization of 6 weeks old BALB/c mice with four doses of purified TcSMP recombinant protein via the intraperitoneal route . Each mouse received a first dose consisting of 50 μg of antigen with aluminum hydroxide as adjuvant , followed by another three doses of 25 μg each , under the same conditions . The doses were given at 15 days intervals and the mice were bled two weeks after the fourth dose . Rabbits were immunized intradermally with 1 mg of TcSMP recombinant protein ( protein concentration 1 mg/mL ) emulsified with Freund’s adjuvant ( Sigma-Aldrich , St Louis , MO ) , followed by another three doses of 1 mg each , under the same conditions . The doses were given at 15 days intervals and the rabbits were bled two weeks after the fourth dose . Sera were separated by centrifugation and kept frozen at -20°C . In order to verify the specificity of the antibodies used in this study , we carried out careful control experiments to ensure that they recognized the correct protein . First of all , the polyclonal antibodies were purified after incubation with the purified recombinant TcSMP in phase with GST immobilized on nitrocellulose membrane followed by elution with glycine . Next , we performed western assays by incubating the immunopurified antibodies against TcSMP-GST and MVK ( mevalonate kinase ) with recombinant proteins ( S2A Fig ) . Anti-TcSMP specifically reacted against TcSMP while anti-MVK recognized only the MVK protein . As an additional control , the coding DNA sequence of the TcSMP ( TcCLB . 510129 . 30 ) gene was synthetized and cloned in phase with GFP into pTREX-GFP vector [39] . Epimastigotes ( CL strain ) were electroporated with 70 μg of the construction TcSMP-GFP by two pulses of 0 . 3 kV/ 500 μF in the BioRad Gene Pulser apparatus . Transfected cells were selected in the presence of 500 μg/mL G418 in LIT medium supplemented with 20% heat-inactivated fetal calf serum ( Vitrocell , Brazil ) . Parasites expressing TcSMP-GFP were incubated with anti-TcSMP antibodies following the same protocol described in the item “immunofluorescence” ( methods section ) . Confocal images obtained from each fluorescence channel were overlapped , and co-localized pixels are shown in panel CP ( S2B Fig ) . The presence of co-localized pixels between fluorescence emission of GFP ( TcSMP-GFP ) and Alexa Fluor 488-labeled anti-mouse immunoglobulins confirmed the antibody specificity for TcSMP protein . Our results with transfected parasites corroborated the immunofluorescence staining pattern obtained with anti-TcSMP antibodies as the recombinant protein TcSMP-GFP displayed the same labeling pattern . T . cruzi developmental forms ( 1 x 107 cells ) and procyclic forms of T . brucei ( 1x107 cells ) were washed in PBS and incubated in 4 x Laemmli sample buffer for 5 min at 100°C . Proteins were separated in 10% SDS-PAGE , transferred to Hybond ECL membranes ( Amersham , GE Healthcare Life Sciences ) and blocked with PBS/7% skim milk overnight at 4°C . Membranes were incubated with polyclonal anti-TcSMP antibody ( 1:200 ) for 3 h , washed with PBS/0 . 005% Tween 20 and incubated for 1 h with a secondary goat anti-mouse antibody ( H+L ) ( 1:10 , 000 ) ( Bio-Rad ) . The immunocomplexes were detected by chemiluminescence using the ECL-Plus Western Blot Detection System ( Amersham , GE Healthcare Life Sciences ) . T . cruzi developmental forms were harvested from culture medium and washed with PBS . Live parasites were incubated in 1% BSA for 10 min on ice followed by 1 h with anti-TcSMP antibodies diluted 1:20 in 1% BSA , washed with PBS and fixed in 4% formaldehyde for 15 min on ice . Thereafter , parasites were incubated with fluorescein-labeled goat anti-mouse IgG ( Sigma-Aldrich , St Louis , MO ) diluted 1:100 in BSA and 1 mM DAPI ( 4' , 6-diamidino-2-phenylindole; Molecular Probes ) . After three washes with PBS , the coverslips were mounted in glycerol buffered with 0 . 1 M Tris pH 8 . 6 containing 0 . 1% p-phenylenediamine . Confocal images were obtained using a Bio-Rad 1024UV system coupled to a Zeiss Axiovert 100 microscope or a Leica TCS SP5 II system . Images were acquired with 100X ( 1 . 4 NA ) oil immersion objectives . In parallel , parasites were first fixed with 4% formaldehyde for 15 min at room temperature and then processed as described above , except by the fact that parasites were incubated by a 1 h with anti-TcSMP followed by 1 h incubation with fluorescein-labeled goat anti-mouse IgG antibody in the presence of 0 . 1% saponin ( 0 . 15% gelatin in PBS containing 0 . 1% sodium azide and 0 . 1% saponin ) . Flow cytometry experiments were performed following the procedures described above . Parasites incubated without antibody or only with the fluorescein-labeled goat anti-mouse IgG antibody were included as controls for auto-fluorescence and background , respectively . The number of fluorescent parasites was estimated with a FACScanto cytometer ( BD Biosciences , Franklin Lakes , NJ ) . In parallel , fixed parasites were placed in 12-well slides for immunofluorescence reaction ( IF reaction ) . The wells were washed with PBS and blocked in PBS/1% BSA solution for 30 min . Parasites were incubated with anti-TcSMP ( 1:20 ) antibodies , washed with PBS and incubated for 1 h with an Alexa Fluor 568 anti-mouse IgG antibody made in goat solution diluted 1:100 in 0 . 1% saponin and 1 mM DAPI ( 4'6 -diamino-2-phenylindole , Molecular Probes ) . Finally , the wells were washed with PBS and the slides were mounted using Glycerol-PPD . The fluorescence was observed by confocal microscopy using the Bio-Rad 1024UV system adapted for a Zeiss Axiovert 100 microscope with Leica TCS SP5 system or II . Images were acquired with a 63X objective with immersion oil .
PSSA-2 orthologs identified in T . cruzi genomes have been annotated as “procyclic form surface glycoprotein” or “procyclic form surface phosphoprotein” in the TriTryp and GenBank databases . In the present work we demonstrate that these proteins are associated with cellular membranes and are expressed at the cell surface of all T . cruzi developmental forms . For this reason , it is more appropriate to refer to these proteins collectively as T . cruzi Surface Membrane Proteins ( TcSMP ) . With the aim of characterizing the TcSMP gene family , we used the BLASTP algorithm to retrieve sequences from the TriTrypDB and GenBank databases by entering the surface glycoprotein ( XP_803885 . 1 ) and procyclic form surface phosphoprotein ( XP_819855 . 1 ) of T . cruzi clone CL Brener ( CLB ) as queries . We identified 19 TcSMP genes in the T . cruzi genome database whose products share approximately 40% identity with the T . brucei PSSA-2 protein , distributed as follows: 9 sequences in CLB , 3 in clone Sylvio X10/4 , 2 in clone Dm28c and 5 in T . cruzi marinkellei ( Table 1 ) . Gene identification in parasite genomes was performed by automatic annotation software that can produce some erroneous predictions . To rule out errors in prediction and annotation , TcSMP sequences were also manually curated . The initial predicted CDSs for the TcCLB . 507711 . 100 and TcCLB . 507711 . 110 genes missed a 44- and a 34-amino acid-extensions at N-terminus , respectively ( S1 Fig ) . In this study we used the correct CDSs , which can be found in S3 Fig . In addition four sequences were clearly misannotated as TS ( TcCLB . 508173 . 120 , TCSYLVIO_001920 , TCDM_01936 , Tc_MARK_706 ) ( Table 1 ) , besides their lack of similarity to the TS gene , they share 70–80% similarity with the TcSMP genes ( Table 2 ) . TcSMP genes were classified into two groups based on the length of their coding sequences ( CDS ) : TcSMP_L comprising a large CDS encoding 45 . 7 to 51 . 3 kDa proteins and TcSMP_S comprising a small CDS encoding 41 . 7 to 43 . 4 kDa proteins ( Table 1 ) . The predicted molecular masses of TcSMP proteins listed in Table 1 were based on the translation of each open reading frame without considering additional processing of the protein , i . e . , removal of the signal peptide of TcSMP_L members after trafficking through the endoplasmic reticulum , as discussed below . Seven TcSMP truncated sequences were found in the T . cruzi genome , one was a pseudogene and the remaining sequences were truncated ( Table 1 ) , as judged by the existence of gaps in the genome assembly . Sequence analysis of TcSMP genes showed a high degree of conservation between CLB sequences , ranging from 82 to 98% identity ( S1 Fig ) . PCR amplification using primers designed based on conserved regions of TcSMP genes allowed the identification of three new members of the TcSMP family in the CLB ( GenBank KJ682657 , KJ682658 and KJ682659 ) , which exhibited 88–91% similarity with previously identified genes but also displayed nucleotide differences , which indicates that they correspond to new copies of TcSMP ( S1 Fig ) . TcSMP genes shared 82–98% nucleotide identity ( S1 Fig ) , whereas amino acid identity varied from 70–90% and 65–93% for TcSMP_L and TcSMP_S , respectively ( Fig 1 ) . Sequence alignment showed that TcSMP_L proteins share an N-terminal extension of 69 amino acids containing seven in frame putative initiator codons . The region immediately after the fourth ATG codon encodes a typical 25-amino acid signal peptide with a putative cleavage site between residues RSA-FF ( Fig 1 ) . Upstream sequences in the vicinity of the third ATG best fit the Kozak consensus sequence ( S4 Fig ) . Among other aspects of the mRNA structure , the context surrounding the AUG codon can modulate the initiation of translation [40] . The Kozak sequence located upstream of the initiation codon is expected to facilitate ribosome binding and thus the beginning of protein synthesis . In view of this , we suggest that the translation of large sequences initiates at the fifth methionine . The presence of several initiation codons in the same reading frame is common among T . cruzi surface protein genes such as GP82 and GP90 [41 , 42] . TcSMP proteins have 2–3 hydrophobic domains at the N- and C-termini ( Fig 1 and S5 Fig ) . Considering that the translation of TcSMP_L starts at the 3rd methionine , the first hydrophobic domain corresponds to a typical amino-terminal signal peptide with basic residues followed by a hydrophobic stretch with a putative cleavage site RSA-FF ( Fig 1 ) . TcSMP_S proteins have two transmembrane domains and the first one is located immediately after the initiator methionine , which was predicted to be a signal anchor [38] . An exception was Tc_MARK_3606 in which the first hydrophobic domain was predicted to be a signal peptide ( Fig 2 ) . A similar pattern is present in PSSA-2 of T . brucei ( S5 Fig ) which is attached to the plasma membrane by a stable transmembrane anchor characteristic of membrane proteins [25] . The C-terminus of TcSMP contains 4 proline residues centered on the YGQ motif that can also contribute to size differences among the TcSMP proteins ( Fig 1 ) . Interestingly , there are 8 tandem proline repeats on the cytoplasmic tail of PSSA-2 that are predicted to form tight helices [25] . TcSMP and PSSA-2 repeats share low degrees of similarity . Systematic BLAST searches revealed the presence of TcSMP_S orthologs in the genomes of Trypanosoma rangeli and mammal-dwelling African trypanosomes . The identified TcSMP orthologs are depicted with selected examples from other species in Fig 2 . Trypanosoma grayi , a trypanosome from crocodilians , has a single transmembrane domain at the C-terminus ( Fig 2 ) . Expression of TcSMP was analyzed using polyclonal antibodies raised against a recombinant protein carrying a small fragment of TcSMP that encodes a predicted transmembrane helix flanked by cytoplasmic and extracellular hydrophilic regions ( see S5 Fig ) . Western blot analysis of whole parasite extracts using mouse anti-TcSMP polyclonal antibodies detected a ~43 kDa protein band expressed in all parasite developmental forms , which is the predicted molecular mass for TcSMP_S ( Fig 3 ) . Assuming that TcSMP_L has a signal peptide that is removed after trafficking through the endoplasmic reticulum , all members of this family can be expected to generate proteins of the same molecular mass as TcSMP_S . An additional ~70 kDa band consistently reacted against anti-TcSMP antibody; this size is close to that of two of these proteins together . Fragoso et al . [26] found some evidence that PSSA-2 forms homodimers or multidimers . As with its ortholog , there are few cysteine residues in the TcSMP protein sequence that could be responsible for the formation of disulfide bridges . Although the sample buffer contains 2-mercaptoethanol or dithiothreitol ( DTT ) , our protein samples were not prepared with antioxidants; hence , intermolecular disulfide bonds could be reconstituted during electrophoresis . Interestingly , when we performed a western blot containing the recombinant TcSMP protein excised from a single 38 kDa band of bacterial extract , two larger bands ( ~80 kDa and ~160 kDa ) reacted against anti-TcSMP antibody after SDS-PAGE ( Fig 3 ) . It is possible that the larger bands are dimer/multidimers and that this could be the stable conformation of native TcSMP . Whilst some evidence supports the hypothesis that this band may be a dimer , further experiments should be carried out to prove this hypothesis . The PSSA-2 coding region size can vary in T . brucei isolates mostly due to insertion/deletion of tyrosine/proline-rich repeats . The predicted size of these sequences in T . brucei_TREU927 and T . brucei gambiense_DAL972 was 47 kDa according to the translation of the open reading frame , while the predicted size for PSSA-2 in T . brucei Lister ( strain 427 ) was 50 kDa , the same size reported by Fragoso et al . , 2009 [26] who used T . brucei AnTat 1 . 1 in western assays . Anti-TcSMP antibodies recognized a 55 kDa protein in T . brucei procyclic forms due to the amino acid-conserved region between TcSMP and PSSA-2 used for immunization procedures . As the T . brucei rhodesiense genome has not yet been sequenced , we cannot predict the exact protein size . In addition , we do not have any information as to whether there are glycosylation sites or whether there were any other post-translational modifications in the processed protein that could justify its larger size compared with PSSA-2 AnTat 1 . 1 . ” Recently , Bayer-Santos et al . [43] performed a proteomic analysis of the T . cruzi secretome in which two populations of extracellular vesicles ( exosomes and plasma membrane-derived vesicles or ectosomes ) and soluble proteins released by epimastigotes and metacyclic trypomastigotes were characterized . Although TcSMP proteins were not reported in this study [43] , we decided to revisit our raw data to search for TcSMP proteins among those identified by only one peptide , which were excluded from the previous work in which only proteins identified with two peptides were considered . Four peptides were identified in samples from epimastigotes and metacyclic trypomastigotes , either in the soluble protein fraction or the small membrane vesicle fraction ( fraction V16 ) ( Fig 1 and Table 2 ) . Two peptides identified in epimastigotes share identity to an ortholog protein from T . vivax ( TvY486_1010870 ) , which may indicate the existence of new TcSMP genes in T . cruzi . These data suggest that TcSMP proteins and their orthologs can be secreted by T . cruzi and T . vivax ( Fig 1 and Table 2 ) . Live and permeabilized parasites were analyzed by indirect immunofluorescence , using anti-TcSMP antibodies ( Fig 4 ) . TcSMP distribution varied from dispersed throughout the cytosol in permeabilized parasites to punctate and concentrated in discrete spots on the cell surface of live parasites . From these results , we conclude that TcSMP proteins are located on the surface as well as intracellularly in T . cruzi developmental forms . Surface proteins follow the parasite’s secretory pathway through the endoplasmic reticulum ( ER ) , where the signal peptide is removed before being addressed to the cell membrane . To confirm that the signal peptide directs TcSMP to the ER , parasites expressing TcSMP-GFP were incubated with anti-BIP , an endoplasmic reticulum marker . Overlapping of confocal images obtained from each fluorescence channel showed co-localized pixels between TcSMP-GFP and anti-BIP ( S2B Fig ) . This result suggests that TcSMP goes to the ER and then it is addressed to the cell surface membrane . The PSSA-2 protein was located on the surface of T . brucei procyclic forms of parasites transfected with the complete ORF protein [26] . We confirmed the presence of TcSMP and PSSA-2 on the cell surface of T . cruzi tissue culture trypomastigotes ( TCT ) and T . brucei procyclic forms by flow cytometry analysis , labeling live and permeabilized parasites with anti-TcSMP antibodies . S6 Fig shows the labeling of live T . cruzi TCT forms and T . brucei procyclic forms with anti-TcSMP . Approximately 90%-92% of live T . cruzi and T . brucei parasites labeled with anti-TcSMP antibodies exhibited greater fluorescence intensity than those incubated with pre-immune serum . Besides indicate that anti-TcSMP antibodies recognized an ortholog in the T . brucei surface membrane , these results confirm that TcSMP is located on the cell surface of T . cruzi tissue culture trypomastigotes ( TCT ) . As the TcSMP protein was detected on the T . cruzi cell surface ( Fig 4 ) , we examined whether it is implicated in parasite-host cell interactions . First , the target cell binding capacity of the recombinant TcSMP protein fused to GST was examined . HeLa cells , immobilized on the bottom of microtiter plates , were incubated with increasing concentrations of recombinant TcSMP or GST , and the bound protein was detected using antibodies directed against TcSMP or GST . Binding of TcSMP to HeLa cells was dose-dependent whereas GST failed to bind ( Fig 5A ) . Next , we determined the ability of the TcSMP protein to trigger host cell Ca2+ signaling , which is required for T . cruzi internalization [44–46] . Recombinant TcSMP or GST , at 40 μg/mL , was added to HeLa cells loaded with the Ca2+ indicator fluo-4 , and Ca2+ signal-inducing activity was monitored by fluorescence microscopy . TcSMP , but not GST , triggered an increase in the fluorescence intensity ( Fig 5B ) . We also evaluated variation of the cytosolic free Ca2+ concentration after challenge with 40 μg/mL TcSMP or GST . The recombinant protein TcSMP induced a Ca2+ signal that , although moderate , was significant when compared with variation in the fluorescence intensity after stimulation with GST ( Fig 5C ) . As the metacyclic stage surface molecule gp82 , which mediates host cell invasion , binds to HeLa cells and induces Ca2+ signal [47] and lysosome scattering to the cell periphery followed by exocytosis [48] , an event that contributes to parasitophorous vacuole biogenesis [49 , 50] , we tested whether the TcSMP protein could induce lysosome mobilization . HeLa cells were incubated for 30 min with 40 μg/mL recombinant TcSMP or GST and then processed for immunofluorescence to visualize lysosomes . Scattering of lysosomes from the perinuclear region to the cell periphery was induced by TcSMP protein but not by GST ( Fig 5D ) . The finding that the TcSMP protein triggered Ca2+ signaling and lysosome scattering suggested that it could be involved in parasite internalization . To determine the involvement of TcSMP in host cell invasion , assays were performed by incubating CL strain metacyclic forms with HeLa cells for 1 h in the absence or presence of the recombinant TcSMP or GST , at 40 μg/mL . TcSMP protein , but not GST , significantly inhibited parasite invasion ( Fig 5E ) . An additional experiment consisted in comparing the effects of TcSMP and gp82 on lysosome scattering and host cell invasion . HeLa cells were incubated for 30 min with TcSMP or the recombinant gp82 , which is also fused to GST , and then processed for immunofluorescence . At 40 μg/mL , TcSMP induced lysosome-scattering comparable to that of gp82 at 20 μg/mL ( Fig 6A ) . The effect of TcSMP at 20 μg/mL was much less pronounced , although it appeared to be higher than that of the GST control ( Fig 6A ) . For cell invasion assays , HeLa cells were incubated for 1 h with CL strain metacyclic forms in the absence or presence of TcSMP or gp82 or in the presence of the two proteins at concentrations of 20 μg/mL and 40 μg/mL . Fig 6B shows that TcSMP significantly inhibited parasite internalization at 40 μg/mL but not at 20 μg/mL , in contrast to gp82 , which exhibited an inhibitory effect at 20 μg/mL . The levels of inhibition by the combination of the two proteins were similar to that of gp82 alone ( Fig 6B ) , indicating that the effects of TcSMP and gp82 are not additive . T . cruzi ( CLB ) genomic sequences were assembled into 41 chromosome-sized scaffolds designated as TcChr1 to TcChr41 ( T . cruzi in silico chromosomes ) [51] . Due to the hybrid nature of clone CL Brener , the chromosome-sized scaffolds were designated S and P to denote the Esmeraldo and non-Esmeraldo haplotypes , respectively [51] . In silico , TcSMP genes were located in chromosomes TcChr37-S , TcChr37-P and TcChr27-P ( Fig 7 ) . The homologous chromosomes Tc-Chr37-S and TcChr37P contain a cluster of 3 and 5 TcSMP genes , respectively , while chromosome TcChr27-P contains only one copy . TcSMP genes in the cluster have the same transcriptional orientation ( Fig 7A ) . TcSMP genes located on chromosome TcChr37-P share ≥92% identity with each other or with those in the homologous TcChr37S . Moreover , these sequences exhibit 82–86% identity with TcSMP ( TcCLB . 508173 . 120 ) located on chromosome TcChr27-P . Due to the in tandem organization of TcSMP genes and high similarity between the sequences , we suggest that TcSMP genes have undergone duplication events followed by mutations that resulted in differences between these genes . Mutation may be one of the mechanisms involved in the generation of sequence diversity and may have contributed to the evolution of these genes in T . cruzi . Gene duplication and in tandem organization may be important strategies to increase protein production in the absence of transcriptional regulation [52] . TcSMP clusters could have evolved by duplication of a chromosomal region of TcChr37 and dispersion to one chromosomal location in TcChr27 . Restriction-mapping data for the genome sequence in which TcSMP loci are located were utilized to select 10 restriction enzymes that cut inside and/or flanking regions of TcSMP genes . Genomic Southern blot hybridization revealed a relatively simple hybridization profile consistent with a gene family with few members ( Fig 7B ) . Most computed DNA fragments predicted from in silico restriction analysis were superimposed on the images of gels obtained by experimental genomic DNA restriction , as indicated by the colored boxes ( Fig 7B ) . In some cases , the hybridization signals could not be assigned to any fragment predicted by the in silico chromosomal digestion analysis . This could be due to the presence of TcSMP genes close to regions of undetermined nucleotides on those chromosomes [51] , called N regions , which preclude accurate determination . The novel copies of TcSMP described in this study ( KJ682657 , KJ682658 and KJ682659 ) that were not assigned to any chromosome-sized scaffold can also be responsible for not assigned hybridization signals . The mapping of TcSMP was carried out by chromoblot hybridization with a specific probe that labeled three chromosomal bands of 3 . 27 , 1 . 02 and 0 . 97 Mb in CLB ( Fig 7C ) . Our previous data assigned the chromosomes TcChr37-S and TcChr37-P to a chromosomal band of 3 . 27 Mb , while the chromosome TcChr27-P was assigned to two chromosomal bands of 1 . 02 and 0 . 97 Mb [53] . Fig 7C shows the hybridization of the TcSMP probe with two bands in the G strain ( 2 . 83 and 2 . 00 Mb ) and in T . cruzi marinkellei ( 2 . 20 and 0 . 98 Mb ) , which could correspond to homologous chromosomes with different sizes . In contrast to T . cruzi , mammal-dwelling African trypanosomes and reptilian trypanosomes appear to have reduced the TcSMP repertoire to a minimum . While T . cruzi CLB harbors 9 TcSMP genes , T . brucei and Trypanosoma grayi genomes are limited to one ortholog . We compared the corresponding syntenic regions around the TcSMP genes in the genomes of T . cruzi CLB , T . brucei and T . grayi , this last was isolated from African crocodiles ( Fig 8 ) . The analysis was performed using a region of approximately 30 kb from chromosome Chr10 of T . brucei , TcChr37-S and TcChr37-P of T . cruzi and contig Tgr_12_V1 of T . grayi . This region exhibits a conserved synteny , the same genes are located in these genomic surroundings in trypanosomes . The conserved genome structure was punctuated by structural divergence , including the insertion/deletion of individual genes and intergenic regions . The bottom of Fig 8 shows the synteny between the chromosome TcChr37P of T . cruzi and the T . grayi contig . While conserved synteny relationships between T . cruzi and T . grayi can be defined , the exact orientation of T . grayi genes in relation to T . cruzi cannot be inferred because the genome of T . grayi has not yet been assembled . The maintenance of this chromosome region during trypanosome evolution suggests that its genomic organization may be functionally important . A phylogenetic tree calculated from the alignment of TcSMP sequences , excluding truncated sequences , is shown in Fig 9 . TcSMP proteins appear to have followed different evolutionary pathways , with two main branches for the clustering of American and African trypanosomes . Within the American trypanosomes , sequences located on the same T . cruzi chromosomes tended to cluster on the same branch: Tc . CLB . 510129 . 30 and TcCLB . 509639 . 10 belonging to TcChr37-P ( highlighted in green ) and TcCLB . 507711 . 100 and TcCLB . 507711 . 110 belonging to TcChr37-S ( highlighted in purple ) , excluding two peptides that did not cluster with the others ( TcCLB . 510129 . 20 belonging to TcChr37-P and TcCLB . 507711 . 90 belonging to TcChr37-S ) . It is worth noting that the TcSMP sequences misannotated as TS in the CLB ( TcCLB . 508173 . 120 ) , clone Sylvio ( TCSYLVIO_001920 ) and T . cruzi marinkellei ( Tc_MARK_706 ) genomes tended to group on the same branch , showing that these sequences may exhibit some differences compared to other TcSMP peptides . T . vivax , T . congolense and T . brucei brucei-T . brucei gambiense and T . brucei proteins are clustered into three separate branches . T . brucei species comprises three morphologically indistinguishable subspecies T . b . gambiense , T . b . rhodesiense and T . b . brucei [54] . Studies based on SSU and 5 . 8S ribosomal sequences had indicated the location of T . vivax branch marginal to the T . brucei , T . b . gambiense and T . congolense branch [55] . T . grayi and T . ralphi were in sister subclades . T . grayi , a crocodilian trypanosome , clustered apart from American and mammal-dwelling African trypanosomes . Interestingly , T . grayi was closer to T . cruzi , T . c . marinkellei and T . rangeli than to African trypanosomes ( T . brucei gambiense , T . congolense and T . vivax ) . The subclade T . grayi comprised trypanosomes from African crocodilids and tsetse flies [56 , 57] , while the subclade T . ralphi comprised trypanosomes from South American alligatorids represented by T . ralphi sp . from crocodilian caimans of Brazilian river basins [58] . ”
We demonstrated that TcSMP is a novel gene family that is conserved among different T . cruzi lineages and with orthologs in other Trypanosoma species , including T . grayi , a trypanosome of reptiles . Genome annotation update is an important step to validate the accuracy and relevancy of genetic information and may provide new information about genomic structure and organization as well as gene function . In the T . cruzi databases ( TcruziDB and GenBank ) TcSMP genes were annotated as surface glycoproteins or procyclic form surface phosphoproteins and a few were annotated as trans-sialidases . An extensive body of literature exists regarding the structure and function of highly expressed T . cruzi surface proteins ( mucin , TS , cruzipain , amastin ) [10 , 12–14 , 59] however , little information is available for the less abundant proteins ( MASP; TcTASV , DGF1 , SAP ) [19–24 , 60–62] . Recently large-scale experimental approaches comprising multistep protein separation strategies and proteomic analysis allowed the identification of a number of low abundance proteins . TcSMP proteins were first identified by proteomic analysis of enriched membrane-enriched fractions of T . cruzi isolated by phase partitioning with Triton X-114 [27] . Several observations led us to conclude that TcSMP is a membrane-spanning protein located at the cellular surface and is also released to the extracellular milieu . First , the deduced amino acid sequence of TcSMP showed the key elements typical of surface proteins in trypanosomes , namely the presence of an N-terminal signal peptide or a signal anchor and a C-terminal hydrophobic sequence predicted to be a TM domain ( Fig 1 and S5 Fig ) . Second , immunofluorescence of live parasites with anti-TcSMP antibodies clearly labeled the surface of all T . cruzi developmental forms ( Fig 4 ) . Co-localized pixels between TcSMP-GFP and anti-BIP in confocal images obtained by immunofluorescence of parasites expressing TcSMP-GFP suggests that TcSMP goes to the ER and then it is addressed to the cell surface membrane ( S2B Fig ) . Moreover , anti-TcSMP antibodies also reacted with PSSA-2 , an ortholog expressed at the cell surface of T . brucei procyclic forms ( S6 Fig ) . Third , TcSMP peptides , previously found in a membrane-enriched fraction , were also identified by proteomic analysis in membrane vesicles as well as in a soluble form in the secretome of epimastigote and metacyclic forms ( Table 2 ) . Taken together , our results are compatible with the assumption that TcSMP are membrane-spanning proteins located at the outer surface and can be released into the extracellular milieu . TcSMP proteins are also located intracellularly in all developmental forms , likely associated with membrane-bound structures ( Fig 4 ) . TcSMP_S and TcSMP_L are α-helical transmembrane proteins that appear to differ from one another by the mechanism with which they are translocated to the plasma membrane ( Fig 1 and S5 Fig ) . TcSMP_S and TcSMP_L have 2–3 TM spanning domains therefore , they could be classified as polytopic proteins inserted into the plasma membrane through two TM domains ( S5 Fig ) . TcSMP_L has three TM domains , the first is a canonical amino-terminal signal peptide that should target the protein to the ER lumen following cleavage by the signal peptidase . TcSMP_S contains two TM domains , the first domain is also found after the initiator methionine and it has been predicted to be a non-cleavable signal anchor . Prediction of membrane protein topology suggested that TcSMP_L are proteins with the N- and C-terminus outside ( e . g . , TcCLB . 510129 . 20 ) or inside ( e . g . , TCD_04964 , Tc_MARK_306 , Tc_MARK_706 ) and TcSMP_S are proteins with N- and C-terminal inside ( S5 Fig ) . Of the seven TcSMP proteins identified in T . cruzi ( clone CLB ) , four have a similar topology to TCCLB510129 . 20 , i . e . , the larger hydrophilic domain , located in the middle of the protein , is predicted to be located intracellularly . The remaining three TcSMP proteins of CLB have the opposite topology , which means that the larger hydrophilic domain is predicted to be located extracellularly . This topology has also been predicted in other T . cruzi isolates ( clones DM28c , Sylvio X/10 ) , T . cruzi marinkellei and T . brucei . Anti-SMP antibodies were generated against 101 amino acid sequences of the N-terminal domain , comprising the second TM domain ( 22 aa ) and a segment of 79 aa that is exposed to the outer surface in TcSMP_S and several TcSMP_L proteins ( TCD_04964 , Tc_MARK_3605 , Tc_MARK_706 ) . The reaction to live parasites by the anti-TcSMP antibodies confirmed that TcSMP is exposed on the outer surface . Anti-TcSMP antibodies also reacted with live T . brucei parasites confirming that PSSA-2 is attached to the outer surface of the plasma membrane by a stable TM anchor [25 , 26] . Although our findings favor the model in which the larger hydrophilic middle domain is predicted to be located extracellularly , we cannot rule out the presence of both TcSMP protein topologies in the parasite . In African trypanosomes and T . rangeli , the orthologs have two TM domains , and the first functions as a signal anchor ( Fig 2 ) . TcSMP orthologs in T . grayi , a reptilian trypanosome , have a single TM domain at the carboxy-terminal domain . TcSMP proteins are expressed in all T . cruzi developmental stages . Translation of TcSMP mRNAs resulted in a single protein with an apparent molecular mass of 40 kDa , indicating that TcSMP proteins have the same molecular mass . The apparent molecular mass of native TcSMP ( 40 kDa ) on SDS-PAGE is relatively close to that predicted after the processing of TcSMP ( 41 . 3 to 43 . 7 kDa ) . If we assume that the TcSMP_L are processed from the signal peptide and the TcSMP_S from the first methionine and signal anchor , they should have fairly similar molecular masses ( 41 . 3 to 43 . 7 kDa ) . The observed discrepancies between the predicted and the observed SDS-PAGE mobility could be explained in part by the cleavage of a short sequence at the amino-terminal region . Another possibility could be the targeting of the nascent polypeptide to the endoplasmic reticulum through the anchor signal where it would be recognized and cleaved by a signal peptide peptidase . The experimental evidence indicates that polytopic membrane proteins can be cleaved internally by a signal peptidase that recognizes non-canonical sequences [63] which could explain the smaller size of the native protein . Finally , it remains unclear whether an individual trypanosome expresses a single TcSMP gene or co-expresses different members of the TcSMP family . Several pieces of evidence have indicated that the TcSMP protein is implicated in host cell invasion . Similar to the surface molecule gp82 , which mediates the internalization of metacyclic trypomastigotes , TcSMP bound to target cells and induced Ca2+ signaling and lysosome mobilization ( Fig 5 ) , events that are required for parasitophorous vacuole biogenesis [48 , 49] . In addition , the recombinant TcSMP protein was capable of inhibiting metacyclic trypomastigote entry into host cells ( Fig 5 ) . The effects of the TcSMP protein on target cell lysosome scattering and parasite invasion were found to be of lower magnitude compared to gp82 , suggesting that TcSMP may play an auxiliary role in parasite invasion , as was previously ascribed to the SAP protein . Like TcSMP , the SAP protein bound to host cells and triggered Ca2+ signaling and lysosome exocytosis , and its recombinant form exhibited the property of inhibiting metacyclic trypomastigote invasion [19 , 24] . We envisage the possibility that the productive interaction of T . cruzi with host cells that effectively results in internalization may depend on diverse adhesion molecules . In the case of metacyclic forms , the signaling induced by TcSMP and SAP may add to that triggered by the major surface molecule gp82 , further increasing the host cell responses required for infection . TcSMP genes are densely clustered within a ∼16-kb region of the genome , with one cluster composed of five genes on chromosome TcChr37-P and another composed of 3 genes on TcChr37-S ( Fig 7 ) . Each of the TcSMP genes in a cluster is in the same transcription orientation . Adjacent TcSMP genes located within the same chromosome share 91–98% identity , and similar values were obtained for TcSMP sequences located on homologous chromosomes TcChr37-P and TcChr37-S . This family could have originated by tandem gene duplication of an ancestral gene and sequence homogenization . The solitary TcSMP copy was found in the chromosome TcChr27-P , and it could have arisen by gene duplication followed by translocation to this chromosome . The degree of sequence similarity and synteny found among TcSMP and its orthologs indicates that the TcSMP and surrounding regions existed before the branching of the evolutionary tree that resulted in different species of the Trypanosoma genus . The maintenance of this chromosome region during trypanosome evolution suggests that its genomic organization may be functionally important . The role of TcSMP in the invasion of mammalian cells by trypomastigotes was demonstrated in several experiments . However , the reason why all developmental forms express TcSMP proteins remains unclear . TcSMP may have another role in non-invasive T . cruzi forms e . g . , to work as a sensor for parasite interaction with the environment of the invertebrate host similarly to the function suggested for its ortholog PSSA-2 in T . brucei [26] . Although we found orthologous TcSMP genes in several other trypanosome species , there is no evidence of expression of this protein in these protozoans . A detailed , functional characterization should therefore be carried out to explain the role of this protein in extracellular and non-invasive forms . TcSMP can be added to the repertoire of proteins expressed at the cell surface and also secreted by T . cruzi . Although it is a minor component of the cell coat when compared to GPI-anchored molecules such as TS , mucins and GIPLs , it has a role in the invasion of mammalian cells by metacyclic trypomastigotes . TcSMP could also transmit signals to host cells . Recently , proteins secreted by T . cruzi have been implicated in arrhythmias in an ex-vivo model [64] . It has been suggested that pro-arrhythmogenic proteins secreted or released by T . cruzi could act as enhancers causing the cardiac conduction system to cross an arrhythmic threshold in chagasic patients . Thus , characterization of a novel gene family encoding proteins that can be found at the cell coat or secreted by T . cruzi can provide new understanding of the interaction of this parasite with its host cells . | Trypanosoma cruzi is the etiologic agent of Chagas’ disease , which infects 6–7 million people worldwide , mostly in Latin America . Currently , there are no vaccines available , and the drugs used for treatment are toxic and are not fully effective . To infect mammalian hosts , T . cruzi relies on the ability to invade host cells , replicate intracellularly and spread the infection in different organs of the mammalian host . Knowledge of the structure and function of T . cruzi surface molecules is fundamental to understanding the mechanisms by which the parasite interacts with its host . T . cruzi infective forms engage a repertoire of surface and secreted molecules , some of which are involved in triggering signaling pathways both in the parasite and the host cell , leading to intracellular Ca2+ mobilization , a process essential for parasite internalization . Here , we described a novel family of T . cruzi surface membrane proteins ( TcSMP ) , including their genomic distribution , expression and cellular localization . We studied the mechanism of action of TcSMP in host-cell invasion and proposed a triggering role for TcSMP in host-cell lysosome exocytosis during metacyclic internalization . TcSMP genes are conserved among different T . cruzi lineages and share orthologs in other Trypanosoma species . These results suggest that the diversification of TcSMP genes in mammalian trypanosomes occurred after continental drift . In T . cruzi this gene family expanded by gene duplication . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| []
| 2015 | Molecular Characterization of a Novel Family of Trypanosoma cruzi Surface Membrane Proteins (TcSMP) Involved in Mammalian Host Cell Invasion |
West Nile virus ( WNV ) has been circulating in California since its first detection in 2003 , causing repeated outbreaks affecting public , wildlife and veterinary health . Epidemics of WNV are difficult to predict due to the multitude of factors influencing transmission dynamics among avian and mosquito hosts . Typically , high levels of WNV amplification are required for outbreaks to occur , and therefore associated viral strains may exhibit enhanced virulence and mortality in competent bird species resulting in increased mosquito infection prevalence . In our previous study , most WNV isolates made from California during 2007–08 showed increased fitness when competed in House Finches ( HOFI , Haemorhous mexicanus ) and Culex tarsalis Coquillett mosquitoes against COAV997-5nt , a genetically marked recombinant virus derived from a 2003 California strain . Herein , we evaluated the competitive fitness of WNV strains isolated during California epidemics in 2004 , 2005 , 2007 , 2011 and 2012 against COAV997-5nt . These outbreak isolates did not produce elevated mortality in HOFIs , but replicated more efficiently than did COAV997-5nt based on quantification of WNV RNA copies in sera , thereby demonstrating increased competitive fitness . Oral co-infections in Cx . tarsalis resulted in similar virus-specific infection and transmission rates , indicating that outbreak isolates did not have a fitness advantage over COAV997-5nt . Collectively , WNV isolates from outbreaks demonstrated relatively greater avian , but not vector , replicative fitness compared to COAV997-5nt , similar to previously characterized non-outbreak isolates of WNV . Our results indicated that ecological rather than viral factors may facilitate WNV amplification to outbreak levels , but monitoring viral phenotypes through competitive fitness studies may provide insight into altered replication and transmission potential among emerging WNV strains .
West Nile virus ( WNV; Flavivirus , Flaviviridae ) naturally cycles among competent avian hosts and Culex mosquitoes that serve as both amplification and bridge vectors [1] . Outbreaks of WNV in North America generally have occurred during the warm summer and fall months , when transmission spills over to include humans and other susceptible mammals [1] . Since its invasion of California in 2003 [2] , WNV has not only adapted to and persisted in diverse habitats [3] , but also has caused several epidemics , impacting public , veterinary and wildlife health [4–6] . Most regions of California now have experienced either year-round or repeated seasonal WNV activity [3] . Epidemics have been associated with intense WNV amplification and historically have followed a three-year pattern of introduction , rapid amplification and then subsidence [7] . Factors linked to WNV outbreaks have included above average temperatures and drought conditions , leading to enhanced virus transmission [8 , 9] . Warming temperatures increase vector blood feeding frequency and shorten the extrinsic incubation period of the virus and in California are linked with reduced rainfall that results in increased domestic irrigation and mosquito production from urban drainage systems . Rapid amplification also may relate to phenotypic differences between enzootic and epidemic isolates of WNV that enhance avian host and vector infection , facilitating transmission[10] . High titer avian viremias , associated with elevated mortality in certain avian hosts , are required to produce efficient mosquito infection [10] and have been a hallmark of WNV epidemics in North America [11] . Avian virulence in American crows has been related to single nucleotide polymorphisms in the NS3 and NS1-2B proteins [12–14] . After the 2004 WNV epidemic in Los Angeles , elevated passerine seroprevalence and corvid de-population were associated with limited WNV transmission during successive years [4] . However , the progressive loss of passerine flock immunity and corvid population recovery led to WNV resurgence in 2008 and again in 2011 [4] . Alternatively , decades of pathogen-host co-circulation may have led to a trade-off between a reduction in avian susceptibility as previously shown for House Sparrows ( HOSPs; Passer domesticus ) and increased WNV virulence leading to stable host competence [15] . Interestingly , the competence of several Culex spp . populations from California for the NY99 strain of WNV did not change significantly during WNV outbreak years compared to non-outbreak years [16] , suggesting that epidemics were not linked to changes in vector competence . Adaptive changes in arboviruses typically appear to be driven by viral diversification within mosquitoes [17] , with subsequent expectoration of unique virus populations [18] followed by purifying selection in vertebrate hosts [19] . Studies with Venezuelan equine encephalitis virus ( VEEV ) , for example , have shown that epizootic strains of VEEV originated from enzootic strains , but possessed less than 2% genetic diversity encoding for increased replicative phenotypes [20] . Enzootic isolates of WNV collected between 2007–08 from different regions of California possessed less than 0 . 2% genetic diversity and demonstrated increased replicative fitness compared to an isolate made during the initial introduction of WNV into California in 2003 [21] . Included in this study was a spring isolate from 2008 preceding the epidemic in Los Angeles that exhibited elevated fitness and virulence similar to the NY99 phenotype [22] . As a follow-up to these findings , five WNV isolates made during epidemics from 2004–2012 were assessed herein to determine if increased replicative fitness may facilitate amplification and thus be predictive of future outbreaks .
Isolates of WNV were made from mosquito pool homogenates collected as part of the California Mosquito-Borne Encephalitis Virus Surveillance Program that had tested positive for WNV RNA by qRT-PCR [23] . One homogenate each was selected from outbreak years 2004 , 2005 , 2007 , 2011 and 2012 from regions of California with high transmission activity ( Table 1 ) and increased incidence of human cases . Supernatant from the selected mosquito pool homogenates was propagated once in African Green Monkey kidney cells ( Vero cells; ATCC no . CCL-81 ) to generate sufficient material for all competition studies . The location , collection date , mosquito species , Ct value of the original mosquito pool homogenate , and infectious titer following Vero cell culture amplification were summarized in Table 1 . The genetically labeled COAV997-5nt virus was utilized as a reference strain in competitions against all the wild type WNV isolates , because its fitness previously has been well characterized in vitro and in vivo [24 , 25] . COAV997-5nt was generated from a clone of the COAV997 isolate collected in July 2003 from Imperial Valley , California , early in the WNV invasion of southeast California . Replicative fitness was assessed through competition in dually infected HOFI and Cx . tarsalis using 1:1 mixtures of equal titers of outbreak isolates and COAV997-5nt . Depredation permits allowed trapping of wild HOFIs at vineyards near Bakersfield , CA under US Federal Fish and Game permit MB-082812-1 and State Fish and Game collecting permit SC-002281 . Culex mosquitoes from which the WNV isolates were made were collected with dry-ice baited traps as part of the California Mosquito-Borne Encephalitis Virus Surveillance Program . Blood collection from chickens was approved under Institutional Animal Care and Use Committee ( IACUC ) protocol 15892 and permitted by the Kern Mosquito and Vector Control District in Bakersfield . Experiments with infectious WNV were performed in an animal biosafety level 3 facility containing aviaries and an insectary approved under USDA permit 47901 and Biological Use Authorization 0872 by the Environmental Health and Safety Institutional Committee of UC Davis . Animal experiments were conducted under protocol number 15895 approved by the UC Davis IACUC and under strict adherence to the American Veterinary Medical Association ( AVMA ) guidelines on the Care and Use of Laboratory Animals . UC Davis is approved for the use of animals in research under the National of Institutes of Health ( NIH ) assurance number A3433 . Infection , sampling and qRT-PCR assay methods essentially followed Worwa et al . [21 , 24 , 25] . Outbreak and reference viruses were competed within dually infected hosts , with wild type and the labelled COAV997-5nt clone detected by concurrent qRT-PCR assays . Briefly , equal concentrations of plaque forming units ( PFU ) of COAV997-5nt ( 8 . 34 log10 PFU per mL stock ) and each outbreak isolate ( stock titers in Table 1 ) were utilized to inoculate HOFIs and Cx . tarsalis . Mixed inocula were analyzed by qRT-PCR and differences in RNA numbers from COAV997-5nt and outbreak isolates expressed as ratios ( outbreak isolate RNA divided by COAV997-5nt RNA ) in Tables 2 and 3 . To account for these relative RNA differences , the COAV997-5nt RNA copy data were normalized by multiplication with the RNA ratio determined for each inoculum . Fitness competitions in mosquitoes were performed using the KNWR laboratory-adapted colony of Cx . tarsalis , established in 2002 from collections made at the Kern National Wildlife Refuge , Kern County , California , which were reared in an insectary at 22°C with a photoperiod of 16:8 ( L:D ) hours . Females were transferred to a BSL-3 insectary and sugar starved for 24 hours prior to exposure to an infectious blood meal containing 7 log10 PFU per mL of each competition group . Blood meals were composed of heparinized WNV seronegative chicken blood spiked with 8 log10 PFU per mL of a 1:1 mix of outbreak isolates and COAV997-5nt , thereby diluting the virus mixture 10-fold in the blood meal . Sugar starved females were allowed to feed for 1 hour in the dark on pre-warmed blood meals offered through a Hemotek membrane system ( Discovery Workshops , Accrington , Lancashire , UK ) . Five fully engorged females were frozen at -80°C immediately following blood meal exposure and were used for infectious blood meal analysis . The remaining blood-fed females were maintained at 26°C , 12:12 ( L:D ) and 60% humidity and daily offered a 10% sugar solution on cotton pads . After 14 days , all surviving females were anesthetized using triethylamine ( Fisher Scientific , USA ) to collect expectorants and then bodies as described previously [21 , 25] . Expectorants were collected from 32 females per group using the capillary tube method . The remaining females were immediately frozen at -80°C and used to enhance sample sizes to estimate viral infection and numbers of RNA copies of each virus . Wild , hatch-year HOFI were captured during the summer of 2012 using grain-baited traps at vineyards near Bakersfield , CA , transported to the Arbovirus Field Station at Bakersfield , treated with 0 . 2 mg per mL of chlortetracycline ( Fort Dodge , Overland Park , KS ) for two weeks , and tested for the presence of antibodies against WNV , St . Louis encephalitis ( SLEV ) and western equine encephalitis virus ( WEEV ) [26] . Seronegative birds were transported to UC Davis . After a 2-week acclimation period in the BSL-3 aviary , groups of six HOFIs were inoculated subcutaneously with 0 . 05 mL containing approximately 1 , 000 PFU of an equal mixture of COAV997-5nt and one of the outbreak isolates . Following challenge , HOFIs were evaluated for the presence of clinical signs until 14 days post infection ( dpi ) , with 0 . 1 mL of blood collected daily by jugular venipuncture between 1–7 dpi and upon termination of the study at 14 dpi . HOFIs were euthanized by placement in a CO2 chamber when they were found moribund or at termination of the study on 14 dpi . Mosquito bodies were homogenized twice for two minutes at 24 Hz using a Tissue Lyser ( Qiagen , USA ) and the suspension clarified by centrifugation at 5 , 000 × g and 4°C for 10 minutes . Culex tarsalis expectorants were amplified by one passage in Vero cell culture to enhance detection of low amounts of viral RNA . Nucleic acids were extracted from HOFI sera , mosquito homogenates and amplified expectorants using a MagMAX-96 Viral RNA isolation Kit ( Applied Biosystems , USA ) . A previously described qRT-PCR assay was utilized for specific detection and quantification of viral RNA in mixed competition samples based on five nucleotide polymorphisms in the envelope gene region of COAV997-5nt [24] . Outbreak isolates were detected using primers previously described [27] . Infectious titers in mosquito blood meals and HOFI inocula were determined using plaque assay titration on Vero cells [12] . GraphPad Prism version 7 ( La Jolla , CA ) was used for statistical analysis and data plotting .
Concentrations of outbreak isolates and COAV997-5nt were measured in homogenates of five individual fully engorged females collected immediately following blood meal exposure of each group of Cx . tarsalis ( Table 2 ) . The number of WNV RNA copies present in each blood meal was expressed as the ratio of the outbreak isolate to COAV997-5nt ( Table 2 ) , and this was utilized for normalizing competition data [24] . Bodies and expectorants from surviving Cx . tarsalis were collected on 14 dpi and infection rates ( Table 2 ) determined based on the percentage of bodies containing WNV RNA ( outbreak isolate and/or COAV997-5nt ) . For each competition group containing one outbreak isolate and COAV997-5nt , we determined the percentage of Cx . tarsalis females with dually and singly infected bodies ( Fig 1A ) and cell culture amplified expectorants ( Fig 1B ) . Using a binominal proportions test , we compared expected equal proportions of singly/dually infected Cx . tarsalis and their expectorates to observed proportions specific for either the outbreak isolate or COAV997-5nt . Although dually infected bodies were present in all competition groups , GRLA-04-1624 , KERN-07-291 and SAYO-05-912 groups contained greater numbers of singly infected Cx . tarsalis ( P = 0 . 05 ) ; all expectorants following competition with isolates GRLA-04-1624 , KERN-07-291 and SAYO-12-772 contained a single virus strain ( P<0 . 001 ) . The proportion of females infected with the outbreak isolates were not significantly different ( P>0 . 05 ) compared to COAV997-5nt ( Fig 1A ) in all competition groups combined . Similarly , there was no difference ( P>0 . 05 ) in the percentage of outbreak isolates and COAV997-5nt detected in the cell-culture amplified expectorants ( Fig 1B ) , and equal numbers of singly infected expectorants contained either KERN-07-291 or COAV997-5nt in that group . There was no significant difference in the amount of RNA ( transformed by y = log10 ) detected in Cx . tarsalis infected bodies following competition , except for isolate KERN-07-291 which replicated less efficiently ( P = 0 . 006 ) in co-infected bodies compared to COAV997-5nt ( Fig 2C ) . However , despite replicating to higher RNA copies in Cx . tarsalis bodies , COAV997-5nt was not transmitted at a significantly higher rate ( P>0 . 05 ) compared to isolate KERN-07-291 ( Fig 1B ) . Each of the five outbreak isolates was competed against COAV997-5nt by dual inoculation of six seronegative HOFIs per competition group . The infectious titer and RNA concentration of the challenge inocula were determined by plaque assay titration and qRT-PCR , respectively , and RNA ratios determined for each group were used for normalization of mixed competition samples ( Table 3 ) . Following serial blood collection between 1 and 7 dpi , the RNA of the outbreak strains and COAV997-5nt were quantified in HOFI sera using concurrent qRT-PCR ( Fig 3 ) . Based on RNA loads , all challenged HOFIs developed dual viremias indicating replication of both outbreak isolate and COAV997-5nt , albeit to varying levels . To determine which virus exhibited higher replicative fitness on each dpi , the mean RNA copy numbers from outbreak and COAV997-5nt viruses from HOFIs in each group were analyzed by a Wilcoxon matched-pairs signed rank test ( Fig 3 ) . With the exception of SAYO-05-912 , outbreak isolates showed higher levels of replication on 2 and 3 dpi compared to COAV997-5nt as indicated by greater mean number of RNA copies ( Fig 3A and 3C–3E ) . In addition , the duration of viral RNA shedding was longer for GRLA-04-1624 , KERN-07-291 , GRLA-11-6246 and SAYO-12-772 when compared to COAV997-5nt ( Fig 3A and 3C–3E ) in concurrently-inoculated HOFIs . In contrast , the SAYO-05-912 outbreak isolate exhibited neutral fitness compared to COAV997-5nt between 2–6 dpi , as indicated by similar RNA copy numbers ( P ≥ 0 . 05 ) and the persistence of detectable RNA from both virus strains in sera through 6 dpi ( Fig 3B ) . Mortality following acute WNV infection was low and without significant differences among the virus competition groups ( Fig 3F ) . There were no deaths recorded among birds in the GRLA-04-1624 and SAYO-12-772 competition groups; one HOFI each died in groups SAYO-05-912 ( 7 dpi ) , GRLA-11-6246 ( 6 dpi ) and KERN-07-291 ( 7 dpi ) following competition against COAV997-5nt ( Fig 3F ) . Overt neurological signs were not observed for any of the challenged HOFIs . Across all competition groups , replication of the outbreak strains and COAV997-5nt peaked on 3 dpi ( Fig 3 ) . When comparing the means of peak replication of all outbreak isolates to COAV997-5nt , a significantly ( P = 0 . 008 ) greater RNA load of 4 . 1±0 . 5 log10 RNA copies was observed for outbreak strains compared to 2 . 6±0 . 4 log10 RNA copies for COAV997-5nt ( Fig 3 ) . Interestingly , the level of COAV997-5nt replication was lower in mixed infections during the current study ( Fig 4; HOFIs 2012 ) when compared to similar infections in previous competitions ( Fig 4; HOFIs 2010 and 2011 ) [21 , 25] despite using the same challenge dose and COAV997-5nt stock . Because HOFIs for the three sets of competitions were trapped near Bakersfield within the same vineyards and were all hatch-year birds , the mean RNA loads from COAV997-5nt were compared with regard to year of capture ( Fig 4 ) . Levels of COAV997-5nt replication gradually declined in hatch-year HOFIs captured between 2010–12 , particularly when comparing peak RNA loads on 3 dpi by Mann-Whitney test . HOFIs from 2012 developed peak COAV997-5nt RNA loads of 2 . 6±0 . 4 log10 , whereas HOFIs from 2011 showed on average significantly higher RNA loads of 3 . 8±0 . 6 ( P = 0 . 004 ) . In contrast , HOFIs from 2010 were more competent developing significantly higher peak COAV997-5nt RNA loads of 4 . 1±1 . 5 compared to 2011 ( P = 0 . 01 ) and 2012 in particular ( P = 0 . 009 ) ( Fig 4 ) .
The goal of the current study was to evaluate the replicative fitness of WNV isolated during five California outbreaks from 2004 to 2012 by competition against COAV997-5nt in Cx . tarsalis and HOFIs . Our findings demonstrated that isolates from four out of five outbreaks replicated more efficiently in HOFIs compared to COAV997-5nt ( Fig 3 ) , but mortality and viral RNA loads were significantly lower compared to previous WNV isolates [21] , presumably due to decreased competence of HOFIs trapped during 2012 ( Fig 4 ) . Similar to our previous study [23] , competition in Cx . tarsalis resulted in varying numbers of singly and dually infected mosquito bodies , but infection and expectoration rates were not dominated by one virus strain over the other ( Figs 1 and 2 ) . Similar to our previous findings [21] , the amount of viral RNA found in Cx . tarsalis bodies was not a reliable determinant of viral fitness , because RNA copy differences between outbreak and COAV997-5nt viruses were either not significant or not proportionate to the virus that infected greater numbers of bodies and expectorants ( Figs 1 and 2 ) . Salivary glands are the final site of WNV replication within Cx . tarsalis and secretion of infectious virus in the expectorant determines potential WNV transmission during subsequent blood feeding on susceptible hosts [28 , 29] . Although significantly greater COAV997-5nt than KERN-07-291 RNA copy numbers were found in mosquito bodies ( Fig 2C ) , equal numbers of females expectorated either KERN-07-291 or COAV997-5nt viruses ( Fig 1B ) , indicating equal transmission , despite greater COAV997-5nt replication . It is possible that the higher infection rate and subsequently larger sample size may have resulted in this difference compared to other groups . Overall , competitions in Cx . tarsalis did not result in the complete displacement of the outcompeted virus as both strains were retained and transmitted within all mosquito groups ( Fig 1 ) . Mortality in HOFIs during the current five outbreak competitions was not different from mortality rates observed with COAV997 and COAV997-5nt infections alone and combined [25] , and was decreased compared to several enzootic California isolates evaluated from 2007–08 [21] and to the WNV NY99 strain in HOFIs collected during 2003 [30] . Although virulence and mortality rates were not elevated in HOFIs in the current competitions , WNV outbreak isolates ( except for SAYO-05-912 ) replicated to higher RNA copy numbers compared to COAV997-5nt . The magnitude and duration of RNA present in sera potentially indicated a commensurate viremia which would be important for oral mosquito infection and subsequent transmission [10] . Interestingly , peak RNA loads were significantly lower for both COAV997-5nt and outbreak isolates in HOFIs that were trapped in 2012 compared to RNA loads in HOFIs collected from the same sites in 2010–11 when co-infected with the same stock of COAV997-5nt , wild type WNV of similar fitness , and using the same challenge protocol [21] . This may be due to decreased competence of the local HOFI population as a consequence of rising natural resistance against WNV and/or a result of host-pathogen co-evolution previously observed in HOSPs [15] . Following invasion of the NY99 genotype into North America in 1999 , high avian virulence and mortality were hallmarks of WNV infection , with HOFIs developing viremias up to 8 log10 PFU per mL sera [22 , 31] . With the emergence of the WN02 genotype; however , the host competence indices of HOSPs for NY99 significantly decreased over time , with lower viremia and mortality rates compared to infection with more recent SW03 and WN02 genotype isolates [15] . Between 2003 and 2012 , there was a 22% WNV infection prevalence in dead California HOFIs [11] and a decline in population abundance since the arrival of WNV in the state [31] , possibly increasing selection pressure on these birds for resistance . This could be one reason for observing reduced HOFI competence for WNV in the current study . Our competition model has the advantage of revealing fitness differences among WNV strains through intra-host competition , therefore minimizing the confounding effects of differences in host immune response . This also presented a limitation of our study , because we used a single moderately competent avian host and vector species in our in vivo competition model without inclusion of low- or high-competence hosts and vectors . Collectively , the competitive fitness of outbreak WNV isolates associated with five California epidemics between 2004 and 2012 was not markedly different compared to enzootic wild type isolates [21] indicating that additional ecological and environmental factors were necessary to trigger WNV amplification to epidemic levels . Similar to the 2012 WNV epidemic in the Dallas-Fort Worth area which was not tied to increased WNV virulence[32] , above-average temperatures and reduced rainfall may have enhanced transmission by mosquitoes and warrants further study . | West Nile virus ( WNV ) is the cause of the largest mosquito-borne epidemic in the United States in recent history , with over 46 , 000 reported human cases to date , including >2000 deaths . Since the detection of WNV in California in 2003 , repeated outbreaks have been associated with explosive WNV amplification in local mosquito and bird populations . Along with climatic and ecological factors , vector-host-pathogen interactions are potential drivers of WNV epidemics and warrant study . Here , we evaluated the competitive fitness of five WNV isolates made during Californian outbreaks from 2004–2012 by co-infection of Culex tarsalis mosquitoes and House Finches against a 2003 founding virus strain . Results indicated that competitive fitness in HOFIs was greater for four of the outbreak isolates , but coincidentally elevated mortality was not observed and viral RNA loads were lower compared to previously studied enzootic strains of WNV . Mosquito competitions resulted in levels of infection and transmission that did not differ among strains . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"united",
"states",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"california",
"geographical",
"locations",
"microbiology",
"vertebrates",
"animals",
"viruses",
"north",
"america",
"rna",
"viruses",
"rna",
"isolation",
"molecular",
"biology",
"techniques",
"insect",
"vectors",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"birds",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"viral",
"replication",
"molecular",
"biology",
"disease",
"vectors",
"insects",
"arthropoda",
"people",
"and",
"places",
"biomolecular",
"isolation",
"mosquitoes",
"eukaryota",
"west",
"nile",
"virus",
"anatomy",
"flaviviruses",
"blood",
"virology",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"amniotes",
"organisms"
]
| 2019 | Comparative fitness of West Nile virus isolated during California epidemics |
With every breath , the dynamically changing mechanical pressures must work in unison with the cells and soft tissue structures of the lung to permit air to efficiently traverse the airway tree and undergo gas exchange in the alveoli . The influence of mechanics on cell and tissue function is becoming apparent , raising the question: how does the airway tree co-exist within its mechanical environment to maintain normal cell function throughout its branching structure of diminishing dimensions ? We introduce a new mechanical design principle for the conducting airway tree in which mechanotransduction at the level of cells is driven to orchestrate airway wall structural changes that can best maintain a preferred mechanical microenvironment . To support this principle , we report in vitro radius-transmural pressure relations for a range of airway radii obtained from healthy bovine lungs and model the data using a strain energy function together with a thick-walled cylinder description . From this framework , we estimate circumferential stresses and incremental Young's moduli throughout the airway tree . Our results indicate that the conducting airways consistently operate within a preferred mechanical homeostatic state , termed mechanical homeostasis , that is characterized by a narrow range of circumferential stresses and Young's moduli . This mechanical homeostatic state is maintained for all airways throughout the tree via airway wall dimensional and mechanical relationships . As a consequence , cells within the airway walls throughout the airway tree experience similar oscillatory strains during breathing that are much smaller than previously thought . Finally , we discuss the potential implications of how the maintenance of mechanical homeostasis , while facilitating healthy tissue-level alterations necessary for maturation , may lead to airway wall structural changes capable of chronic asthma .
The act of breathing creates a mechanical environment that pervades all structures in the lungs down to the molecular level [1] . As a pressure difference across the lungs draws air through the airway tree for gas exchange , all airways dilate and transmit mechanical stresses and strains to their cellular constituents – including smooth muscle , epithelial , and fibroblast cells . All of these airway cell types reside within a complex bifurcating airway tree , and through mechanotransduction , they actively sense and respond to their mechanical environment [1]–[3] . Indeed , growth and remodeling of the extracellular matrix is consistently observed in response to chronically altered mechanical stress and occurs through a concerted response of multiple airway cell types [4]–[6] . This study asks the following question: What are the fundamental principles guiding the distribution of airway wall properties – in particular , airway wall thicknesses and airway wall material properties - throughout the conducting airway tree residing in a dynamic mechanical environment ? Previous studies have examined airway tree design principles but have ignored both the dynamic mechanical forces to which the airway tree is perpetually subjected and the tissue-level biomechanical properties of the airway wall . Instead , based on the concept of a self-similar rigid tree [7] with optimal space-filling properties [8] , it was presumed that the fractal design of airway luminal radii through the airway tree is optimal to transport fresh air to the periphery of the lung for efficient gas exchange . A tree design based purely on optimizing gas transport through rigid pipes ignores the fact that breathing dynamics can produce mechanically-driven alterations in the cells and tissue of the airway walls . Moreover , it is conceivable that these alterations would eventually modify the material properties and hence the caliber of the pipes themselves perhaps in a fashion destroying the underlying physical optimization related to gas transport [9] . Here , we introduce a new mechanically-based design principle for the airway walls of the conducting airways , in which mechanotransduction at the level of cells occurs in response to an altered mechanical microenvironment and is driven to orchestrate tissue-level structural changes of the airway wall to restore and maintain a preferred mechanical microenvironment . This theory is termed mechanical homeostasis and provides a plausible physiological role for mechanotransduction [10] in mechanically-driven tissue and organ systems . The concept has emerged as a prevalent theory in the vascular system [11] . However , mechanical homeostasis has not yet been conceptualized nor tested in the airway system . Using experimental and modeling approaches , we show that the distribution of tissue-level biomechanical properties of the airway walls within the normal conducting airway tree is consistent with the existence of mechanical homeostasis . We also provide the likely desired homeostatic conditions for a healthy airway tree undergoing tidal breathing and occasional deep inspirations . Lastly , we address the implications of a mechanically-driven design principle with regard to airway disease . We conjecture that changes in the mechanical environment alone would facilitate healthy tissue-level alterations necessary for maturation on the one hand , but could also lead to airway wall structural changes capable of chronic asthma in a “misguided” attempt to sustain such mechanical homeostasis .
We first examine if and how the principles of mechanical homeostasis occur in structurally intact bovine airways in vitro . We measure the quasi-static relationships between airway luminal radius ( Rin ) and transmural pressure ( PTM ) ( Fig . 1a , red circles ) . By adopting a computational model of vascular mechanics using a strain-energy formulation for thick-walled cylindrical tubes [12] , [13] , we estimate the three-dimensional stresses and strains within intact airway wall tissue ( Methods ) . We implement this analysis to determine the optimal model fits to our data for positive PTM ( Fig . 1a , solid lines ) , and calculate the relationships between the circumferential stress at the inner wall ( σθ ) and PTM ( Fig . 1b , solid lines ) and the corresponding incremental circumferential elastic modulus ( Yinc ) and PTM ( Fig . 1c , solid lines ) . This analysis identifies two salient mechanical features present within our airways . First , within the typical operating PTM in vivo ( 0 . 5 to 1 kPa ) , every airway experiences a relatively narrow fixed range of σθ between about 8 and 15 kPa , which is less than 6% of the maximum stress . We approximate this as a fixed circumferential stress of 12 kPa at the mean operating PTM of 0 . 75 kPa in Fig . 1b . Second , for all 11 airways measured , there is a nearly identical relationship between Yinc and PTM with values between 180 and 310 kPa at PTM = 0 . 75 kPa , as defined by the dashed curves in Fig . 1c . Importantly , we find no correlations between Yinc at 0 . 75 kPa PTM and absolute airway size ( Fig . 1c inset ) suggesting that the airway wall stiffness is approximately the same for many generations of the airway tree when evaluated around the physiological operating PTM of 0 . 75 kPa . We next utilize a computational modeling approach to examine how the uniform σθ and Yinc would be maintained throughout an airway tree structure exposed to the transmural pressures of breathing . The substantial decreases in luminal radius from the trachea ( generation 0 ) to the periphery ( generation 26 ) presumably allow for optimal gas transport via fractal branching but would also result in vastly different circumferential stresses for the resident cells within the airways , as evident in simplistic terms by the geometric relationship of LaPlace's law ( σθ = PTM * Rin/H where H = wall thickness ) . To maintain a constant σθ and Yinc at the operating PTM throughout the airway tree , the airway wall areas and material properties would need to play a compensatory role within the thick-walled cylindrical airway . We re-analyze high resolution computed tomography ( HRCT ) data measured in humans by several investigators [14] , [15] ( Methods ) to discover a strong linear relation between airway wall area and airway luminal radius throughout the airway tree ( Fig . 2a ) . Remarkably , this relationship is maintained throughout all stages of lung growth from birth to adulthood . That is , an airway at the periphery of an adult lung , which has the same luminal radius as a central airway in a child , also has the same wall area . Taken together with our results of radius-independent micromechanical environment in Fig . 1 , these findings are consistent with the existence of mechanical homeostasis within the airway tree . Using the geometric relationship in Fig . 2a , we next predict the airway wall material parameters required to maintain the data-derived conditions of mechanical homeostasis ( from Fig . 1 ) throughout the airway tree . By necessity , the model-determined material parameters describing the nonlinear elasticity of the airway wall ( Eq . 1 ) increase from the periphery to the trachea ( Fig . 2b ) . From the generation-dependent geometric ( Fig . 2a ) and material ( Fig . 2b ) relationships , we then compute σθ and the circumferential stretch ratio ( λθ ) ( Fig . 2c , solid lines ) for the entire airway tree and consequently , we obtain Yinc ( Fig . 2c , slopes of solid lines ) . The peripheral airways have smaller Yinc than the central airways at any given λθ , which is consistent with observed decreases in collagen and cartilage content down the airway tree [16] . As a direct consequence , the progressively increasing stretch ratios along the nonlinear σθ- λθ curves allow the peripheral airways to experience the same σθ and Yinc as the central airways when exposed to typical operating PTM ( Fig . 2c , cyan circles ) . To validate our model prediction , we utilize the predicted σθ- λθ curves to calculate the relationships between λθ and PTM for the airway tree ( Fig . 3a ) . These relationships are directly measurable in human lungs in situ for large airways [17] ( greater than 2 mm luminal radius ) and in vitro for peripheral human airways [18] , [19] . From the trachea to the periphery , our analysis predicts that the airways are progressively more compliant . The specific airway compliance ( Δλθ/ΔPTM ) from 0 to 3 kPa PTM modestly increases from the trachea ( 4 . 4 Pa−1 ) to the periphery ( 5 . 1 Pa−1 ) ( Fig . 3b , solid black line ) . Our predictions of specific airway compliance are consistent with the human data in the literature [17]–[19] ( Fig . 3b , black circles ) , our bovine data ( Fig . 3b , blue crosses ) , and data in dogs [20] and rabbits [21] ( Fig . 3b , green squares and cyan stars ) . Importantly , our predictions are also consistently below the estimate of parenchymal hole expansion , which is generally accepted as the limit of airway expansion ( Fig . 3b , dashed line ) . We also performed a sensitivity analysis that shows that these predictions remain essentially the same when the calculations are repeated for an optimal PTM of 0 . 5 kPa ( solid magenta ) instead of 0 . 75 kPa ( solid black ) . Additionally , the root mean square error ( RMSE ) between model and data are similar for both PTM of 0 . 5 kPa and 0 . 75 kPa independent of airway radius ( Fig . 3c ) . Thus , the agreement between our model predictions and the data in Fig . 3b is consistent with the notion that the known generation-dependence in specific airway compliance exists as a means to maintain a constant intrinsic mechanical microenvironment in response to PTM throughout the airway tree . With only modest increases in specific airway compliance down the tree , our results also suggest that cells in the airway wall are well-equipped to respond uniformly to strain-dependent phenomena , such as cellular fluidization from deep inspirations [3] . Since it has not been possible to directly measure small airway mechanical properties in vivo , computational models of airway wall mechanics are vital to estimate the mechanical strains of breathing that are used in mechanobiological cell culture and tissue strip experiments . One prevailing model developed by Lambert et al . [22] was based on the extrapolation of limited empirical fits of radius-PTM data [17] . This model has been used in isolated ASM strips to apply in vivo like loads that would mimic a small airway's structure ( 1 . 1 mm luminal radius ) . In this experimental model , deep inspirations to TLC result in large sustained reductions in ASM constriction [23] . These results are in stark contrast to PTM oscillations applied directly to constricted bovine [24] and human [25] airways , which have little to no impact on airway caliber . Interestingly , we find that while the Lambert model has similar predictions to our model for large airways , the Lambert model also predicts airways having unrealistically large compliance as radius decreases ( Fig . 3b , solid gray line ) , which eventually become much larger than the available data and the parenchymal hole expansion limit . In contrast , our model suggests that the mechanical strains experienced by ASM cells during breathing are much smaller than previously thought as our simulations demonstrate ( Fig . 4 ) . Furthermore , these strains depend only mildly on airway size increasing from 3 . 3% and 6 . 5% at the trachea to 4 . 9% and 9% in small airways during tidal breathing and deep inspiration , respectively . As a consequence , their sustained functional impact on airway responsiveness would be greatly attenuated when tested under physiologically appropriate mechanical conditions [24] , [25] .
Our data and modeling analyses suggest that the organ-level structure of the airway system plays a crucial role in sustaining not just organ level function , namely gas transport , but also the function of its individual components down to the cellular level in every airway wall throughout the airway tree during breathing . Specifically , the airway tree's structural design maintains circumferential stresses and circumferential elastic moduli in a relatively narrow range . The circumferential stress and the incremental modulus determine the effective stiffness of the extracellular matrix which is known to significantly influence cellular behavior [2] . Thus , the regulation of the mechanical microenvironment throughout the airway tree ( Fig . 1 ) , during growth ( Fig . 2a ) , and in response to tidal breathing ( Fig . 4 ) to maintain a homeostatic state ensures normal cellular and tissue-level function within a dynamic PTM environment in a way that maintains compatibility with the physical structure of the airway tree required for optimal gas transport . It is interesting to note that while homeostasis is formulated in terms of stress and modulus , it also results in a nearly constant circumferential strain independent of airway radius ( Fig . 4 ) implying that similar cells in the walls but with different sensitivity to strain did not have to evolve for efficient airway function . Thus , the consistency of tissue-level structure with a mechanical design principle further implies an active and essential role of airway cells as controllers of mechanical homeostasis for the airway wall system . While airway luminal radius and length have evolved to efficiently deliver gas through a tree structure [8] , the airway wall tissue structure has evolved to provide the proper mechanical milieu for its constituent cells . Before examining the implications of a mechanical design principle , we first discuss the limitations of our intact airway experiments and computational modeling approaches . We used bovine intact airways in vitro to obtain pressure-radius relationships . Compared to tissue strips and cell culture , this preparation maintains the 3-dimensional architecture of the airway and applies a physiological transmural pressure as a means of stretching the cells and the extracellular structures in the airway wall . However , neural airway tone that is known to constantly modulate airway diameter in vivo [26] is removed in this preparation . When using these data in our computational approach , we neglected the pressure-diameter relation for negative PTM , which may be important during flow limitation; we found that to describe the negative PTM data , we would need a different form of the strain energy function . While our analysis of the data advances airway wall modeling by invoking a thick-walled cylinder approach , the model uses several simplifying assumptions . We assume that the airway walls are elastic , which is a limitation of the strain energy density formalism itself . We also assume that the airway walls are homogeneous and maintain a cylindrical shape at all PTM . During breathing , the lung is exposed to cyclic stretch and the irregular nature of the stretch pattern has important consequences on cellular function [27] . If cellular growth or remodeling of the airway are of interest , these irregular stretch patterns may influence these processes . The thick-walled hollow cylinder is assumed to be homogeneous and isotropic , which are the usual assumptions in airway wall modeling . At the length scale of single cells , however , this is certainly not true . We also assume that the typical operating pressure is 0 . 75 kPa for all conducting airways and for all stages of life ( birth to adulthood ) . It is probable that the operating pressure may change throughout the course of development , and it may also differ slightly throughout the airway tree due to gravitational differences in pleural pressure , resistive losses in airway pressure down the airway tree , surface tension , and local differences in parenchymal tethering forces . A sensitivity analysis , however , revealed that almost identical results are obtained for operating pressures of 0 . 75 kPa and 0 . 5 kPa . Another limiting assumption in both our data analysis and computational approach is plane strain , in which the properties and deformation of the airway wall are assumed to not change in the axial direction . In our in-vitro experiments , the two ends of the airways are fixed at a pre-stretch ratio that is consistent with in-vivo lengthening that occurs during tidal breathing [28] . In this setup , the plane strain condition is indeed valid for a section of the airway in the middle away from the boundaries which we verified using the ultrasound; data only from this region are used in the analysis . However , in the in-vitro experiments and in our computational approach , airways cannot lengthen dynamically with radial expansion , as presumably occur in-vivo . Relaxing the plane strain assumption to allow for airway lengthening would require complex finite element modeling and knowing the in vivo boundary conditions for the airway which are beyond the scope of this study . We next discuss the implications of mechanical homeostasis on the behavior of airways over various time scales . On short time scales , ASM cells can actively control local stresses in the wall by contraction and relaxation . This has indeed been observed in dogs where airway lumen varied substantially from day-to-day over a period of a year [29] . On longer time scales , many other cell types participate in controlling stresses via remodeling of the airway wall . In fact , the consistency of geometric dimensional relationships from birth to maturity in the airway tree ( see Fig . 2a ) suggests that proper tissue growth is not a pure biochemical process but it also requires significant mechanically-driven feedback . As growth factors cause an airway's luminal radius to increase in size [30] , the circumferential stress and elastic modulus would also initially increase at a given operating PTM . However , cellular compensatory mechanisms would detect this deviation from mechanical homeostasis and trigger the airway cells to build more wall tissue to restore the homeostatic state , characterized by an optimal circumferential stress and incremental Young's modulus . Therefore , a generation 0 airway at birth would grow to a much larger luminal radius in maturity while maintaining a similar preferred mechanical environment . As a consequence of this airway wall growth process , our analysis predicts a generation-dependent decrease in airway wall compliance as the airways mature ( Fig . 5a , black circles ) , which is in agreement with data in the literature [21] ( Fig . 5a , squares and triangles ) . This proposed process is consistent with an emerging hypothesis that the unique geometric and material properties of mature organs derive from mechanical stimuli and feedback throughout development [31] and may add another mechanical stimulus for growth in the respiratory system [32] . In an analogous fashion , our results also have implications for airway diseases . While the airway wall structure is designed to maintain normal cell function under physiological operating conditions throughout life , it remains ill-prepared for persistent ASM activation from environmental sources . ASM contraction immediately changes airway wall circumferential stresses by introducing an active mechanical stress that reduces airway caliber and increases wall thickness . Thus , airway cells would be immediately driven from their physiologic mechanical homeostatic state . Over time , repetitive ASM activation would stimulate airway wall remodeling processes consistent with growth in an attempt to restore mechanical homeostasis . Chronic constriction would result in an airway wall that is thicker and stiffer ( Fig . 5b ) , as is consistent with data on asthmatic airways in the literature [33] . However , as ASM activation is sporadic in nature and occurs on a much faster timescale than structural remodeling , the presence of airway remodeling indicates a tissue-level system struggling to restore a steady-state mechanical homeostasis that may never be fully realized . Applied throughout the airway tree , this deviation from mechanical homeostasis would have disastrous impacts to function at the level of the cell , tissue , organ , and organism [6] , [9] and would not be readily reversible with bronchodilators [33] . Mechanical homeostasis may thus emerge as a governing principle that divides health and disease within the respiratory system , and may also unify respiratory diseases with a host of others whose progression is intimately coupled to mechanical signaling [34] .
Experiments were carried out using bovine lungs obtained from a local slaughterhouse immediately after death ( Research 87 , Bolyston , MA ) . Protocol approval was not required . Our system for intact airway experiments has been described previously [35] . Briefly , bovine lungs were obtained from a local slaughterhouse . A bronchus of the right lung ( generations 10–17 ) was dissected and the side branches were ligated . The airway was cannulated at each end and mounted horizontally in a tissue bath containing gassed ( 95% O2-5% CO2 ) and heated ( 37°C ) Krebs solution . The airway was stretched longitudinally ( 110–120% of its resting length ) and held fixed at its extended length for the entire experiment . Tissue viability was then confirmed with both electric field stimulation and acetylcholine ( ACh; 10 5 M ) . A computer-driven pressure-controlled syringe pump delivered PTM changes to the intact airway . The proximal cannula was mounted in series to a hydrostatic pressure column filled with Krebs solution , which also filled the airway lumen . The difference in fluid height between the horizontally mounted airway and the pressure column determined the intraluminal pressure ( thus PTM ) experienced by the airway . A portable ultrasound system ( Terason 2000 ) , consisting of a high-frequency linear array transducer ( 10L5 ) and an external beamformer module , was used to visualize the intact airway . The hardware was connected directly to a personal computer running software that both controlled the imaging settings ( focal depth , focal length , and gain ) and acquired and stored the images in real time . The ultrasound transducer was mounted above the intact airway and partially submerged in the tissue bath . Using a three-directional micromanipulator , the transducer was positioned over the airway's longitudinal axis at its diameter . The airway was imaged at 30 fps with fixed ultrasound imaging settings ( focal depth: 30 mm , focal length: 13 mm , gain: 0 . 2 ) . We controlled Pin delivered to the structurally intact bovine airways of fixed length while Rin and Rout were directly measured with ultrasound imaging . Quasi-static , passive R-PTM curves were measured via slow ramps in Pin ( −1 . 5 to 3 kPa , 0 . 1 kPa/second ) with Pout set to 0 kPa . The positive expiratory limb ( 3 to 0 kPa ) was used for our analysis to probe the full physiological range of breathing ( 0 . 5 kPa PTM at functional residual capacity to 3 kPa at total lung capacity ) . Eleven bovine airways were obtained from the right lower lobes of eight different animals . | With every breath , mechanical pressures change in the lung and permit air to efficiently traverse the airway tree and undergo gas exchange . These pressure variations also influence cell and tissue function , raising the question: how does the airway tree co-exist within its mechanical environment to maintain normal cell function throughout its branching structure of diminishing dimensions ? We introduce a new mechanical design principle for the conducting airway tree in which mechanotransduction , the process that converts mechanical forces on cells to biochemical signals , is driven to orchestrate tissue-level structural changes that can best restore a preferred mechanical microenvironment; a concept termed mechanical homeostasis . We report in vitro mechanical properties for a range of airway sizes and present a mathematical model that describes the data . Our results indicate that airways indeed consistently operate within a preferred mechanical homeostatic state . We further describe how this mechanical homeostasis while facilitating healthy tissue-level alterations necessary for maturation can inadvertently lead to airway wall structural changes capable of chronic asthma . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"bioengineering",
"tissue",
"mechanics",
"biomedical",
"engineering",
"biology",
"biophysics",
"biomechanics",
"engineering"
]
| 2013 | A Mechanical Design Principle for Tissue Structure and Function in the Airway Tree |
Dengue has emerged as one of the most important infectious diseases in the last five decades . Evidence indicates the expansion of dengue virus endemic areas and consequently the exponential increase of dengue virus infections across the subtropics . The clinical manifestations of dengue virus infection include sudden fever , rash , headache , myalgia and in more serious cases , spontaneous bleeding . These manifestations occur in children as well as in adults . Defining the epidemiology of dengue in a given area is critical to understanding the disease and devising effective public health strategies . Here , we report the results from a prospective cohort study of 4380 adults in West Java , Indonesia , from 2000–2004 and 2006–2009 . A total of 2167 febrile episodes were documented and dengue virus infections were confirmed by RT-PCR or serology in 268 cases ( 12 . 4% ) . The proportion ranged from 7 . 6 to 41 . 8% each year . The overall incidence rate of symptomatic dengue virus infections was 17 . 3 cases/1 , 000 person years and between September 2006 and April 2008 asymptomatic infections were 2 . 6 times more frequent than symptomatic infections . According to the 1997 WHO classification guidelines , there were 210 dengue fever cases , 53 dengue hemorrhagic fever cases ( including one dengue shock syndrome case ) and five unclassified cases . Evidence for sequential dengue virus infections was seen in six subjects . All four dengue virus serotypes circulated most years . Inapparent dengue virus infections were predominantly associated with DENV-4 infections . Dengue virus was responsible for a significant percentage of febrile illnesses in an adult population in West Java , Indonesia , and this percentage varied from year to year . The observed incidence rate during the study period was 43 times higher than the reported national or provincial rates during the same time period . A wide range of clinical severity was observed with most infections resulting in asymptomatic disease . The circulation of all four serotypes of dengue virus was observed in most years of the study .
Dengue is caused by infection with one of the four dengue viruses: dengue virus 1 ( DENV-1 ) , DENV-2 , DENV-3 and DENV-4 [1] . Infection with any of these viruses may result in asymptomatic infection , dengue fever ( DF ) , or the more severe forms , dengue hemorrhagic fever ( DHF ) and dengue shock syndrome ( DSS ) . DHF and DSS were recognized in Southeast Asia soon after multiple serotypes began to circulate in the 1950s [2 , 3] . Since then the burden of dengue has increased rapidly with the number of annual cases worldwide rising from 908 in the 1950s to 925 , 896 in the 2000s[4] . The number of dengue-endemic countries has also expanded from nine to over 110[4 , 5] . Cases of DHF and DSS have also been increasingly recognized in other regions including South Asia , Latin America and the Pacific [6–9] , with pediatric cases being more common . In recent years , DF and DHF/DSS have become more common in adults [10–12] . Because of the increased geographical circulation of the virus and the impact of the infection , dengue virus is widely recognized as the most important arboviral infection worldwide . Many clinical and epidemiological studies on dengue have relied on outbreak investigations and hospital based studies [13–23] . These studies provide a wealth of data regarding clinical manifestations , laboratory parameters , pathology , and management of the disease . However , they also have some limitations . Hospital studies , for instance , mostly represent severe cases and do not cover the wide clinical spectrum of dengue infections in adults . Furthermore , hospital studies lack pre-illness and early illness sera that can be used to characterize an individual’s pre-infection dengue virus immune status or to measure laboratory predictors of disease severity . Therefore , there is a need for prospective population-based studies to complement the hospital-based investigations [24] . This form of study contributes to an overall picture of the spectrum of clinical disease in a given geographical area . To study the epidemiology of dengue virus in Bandung , West Java , Indonesia , we conducted a prospective study in a cohort of adults from August 2000 to June 2004 and from September 2006 to April 2009 . The aims of the study were to determine the incidence of symptomatic and asymptomatic infections; determine the temporal distribution of dengue virus serotypes; characterize the clinical manifestations of dengue in adults and determine whether there is a correlation between severity of disease , pre-illness immune status and sequence of infections . Preliminary results of the first two years were published previously [25] . Here , we report a comprehensive seven year account of the epidemiology , virology , immunology and clinical presentation of dengue virus infections within West Java cohorts .
The study protocol was reviewed and approved by the Institutional Review Boards at the U . S . Naval Medical Research Unit No . 2 and the National Institute of Health Research and Development , Ministry of Health , Indonesia ( DoD 30855 , KS . 02 . 01 . 2 . 1 . 2181 and N2 . 2006 . 0001 , KS . 02 . 01 . 2 . 1 . 2776 ) in compliance with all U . S . Federal Regulations governing the protection of human subjects . Each volunteer provided informed written consent upon enrollment . The study was conducted in two phases: from August 2000 to June 2004 and from September 2006 to April 2009 . The first phase was carried out in factories A and B and the second phase in Factories A and C . Approximately 70% of the total factory workers participated in the study . A cohort of 2978 adult volunteers was prospectively followed during the first phase and 2726 during the second phase . Among these volunteers , 1324 participated in both phases . Details of the study design and procedures are illustrated in Fig 1 and are also described by Porter et al [25] . Briefly , blood was collected during enrollment and every three to four months thereafter . Factory employees who were ill were required to be seen by a factory clinician in order to be officially excused from work and still receive wages for the day . Employees who were absent , but did not report to the factory clinic were visited by a study team nurse . Volunteers who experienced fever were evaluated at the factory clinics for clinical assessment and blood was collected when indicated by study clinicians or nurses . A complete blood count ( CBC ) and dengue virus infection diagnostic tests as described below , were performed . Patients were advised to be hospitalized if their platelet count was less than 150 , 000/mm3 or at the discretion of the clinic attending physicians . In order to account for any illnesses that may have been missed , at each serosurvey volunteers were asked about any history of fever or any other illness since their last serosurvey ( S1 Form ) . For volunteers meeting hospitalization criteria , clinical data was collected every day . For volunteers with confirmed dengue virus infection who did not meet hospitalization criteria , daily observation was conducted by study nurses either through home visits or phone calls . To diagnose dengue virus infection , virus isolation [25] and RT-PCR [26] were performed on blood specimens collected during the acute phase of illness . Dengue virus IgM , IgG antibody ELISA ( FocusTechnology ) , and hemagglutination inhibition ( HI ) assays [27] were performed on acute and convalescent specimens . A plaque reduction neutralization test ( PRNT ) using BHK-21 cells was performed on pre-illness , acute and convalescent specimens from confirmed dengue virus infection patients . PRNT was also used on paired serosurvey specimens to confirm potential asymptomatic dengue virus infections identified using a screening strategy described below . The dilution that produced a 50% reduction in plaque count compared to a negative control sample was determined by probit analysis using SPSS . To estimate the incidence of asymptomatic dengue virus infections , 25% of the total volunteer population from September 2006 to February 2008 ( 675 volunteers ) was randomly selected using SPSS . Basic demographics ( gender , age , factory of employment ) between the randomly selected subset and the entire cohort during that time period were not statistically different . During this period , serum samples were collected every three to four months from each volunteer up to a total of six serum samples . In lieu of testing thousands of samples for which the HI test is too cumbersome , we developed a new method to identify asymptomatic dengue virus infections utilizing dengue virus IgG ELISA ( Focus Technology ) assays for screening followed by PRNT for confirmation . First , we established an IgG index ratio ( IR ) that could be used to identify potential asymptomatic infection cases from consecutive serosurvey samples . In order to do this , we tested 43 paired serosurvey sera , collected before and after confirmed symptomatic dengue virus infection episodes , along with 38 paired sera from confirmed non-dengue virus febrile episodes . A post-/pre-illness IgG IR was calculated in dengue virus and non-dengue virus febrile episodes and receiver operating characteristic ( ROC ) analysis was used to determine the IgG cut-off ratio to identify dengue virus infections . As samples identified through this screening process were to be further tested by PRNT , a conservative cut-off ratio was chosen in order to ensure that no cases were missed . We chose the lower IR between the lowest IR in the dengue virus infection group and the highest IR in non-dengue virus infection group as the cut-off value for screening . This resulted in an IR of 1 . 2 . Of note , this strategy did not work if non-consecutive ( more than 4 months between samples ) serosurvey samples were used . In order to validate this approach , we ran IgG ELISAs on the 13 sample sets from the first two years of the study that were screened for asymptomatic infection using HI and confirmed by PRNT[25] and a random sampling of sample sets that were not identified as asymptomatic dengue virus infections based on HI screening ( n = 47 ) . For all confirmed asymptomatic dengue virus infections ( ≥ 4-fold increase in HI titer confirmed by PRNT ) , the IgG ratio between consecutive serosurvey samples was ≥1 . 2 ( range 1 . 2 to 29 . 1 ) and in all cases tested that were not identified as asymptomatic dengue virus infections based on HI screening ( <4-fold increase in HI titer ) , this ratio was less than 1 . 2 ( range 0 . 9 to 1 . 1 ) ( S1 Fig ) . Upon validating the cut-off IgG IR , six serial serosurvey specimens from 675 volunteers were tested . Specimens with IgG IRs ≥ 1 . 2 between two consecutive serosurvey samples were further tested by PRNT . The serosurvey samples from each volunteer were tested simultaneously according to the manufacturer’s instructions , using the same lot of the dengue virus IgG ELISA kit . The envelope genes from eight DENV-1 , one DENV-2 , three DENV-3 and six DENV-4 isolates were sequenced ( Genbank accession numbers KR604819-35 ) as previously described [28–30] . In brief , viral RNA was extracted from virus isolates using the Qiamp Viral RNA mini kit ( Qiagen , Germany ) . Three overlapping fragments covering the Envelope-NS1 genes ( approximately 2700 bases ) were amplified by RT-PCR using serotype specific primer sets . Amplicons were purified and the BigDye cycle sequencing kit ( Applies Biosystems , USA ) was used for sequencing reactions . Sequencing reactions were run on a 3130 XL Genetic Analyzer ( Applied Biosystems ) and sequence outputs were assembled using Sequencher software ( Genecodes , USA ) . Phylogenetic trees were generated using the Neighbor Joining method with bootstrapping in MEGA 4 [31] . The following definitions were used in this study: Dengue virus infection: a recent dengue virus infection was confirmed when DEN virus was isolated , or the RNA was detected in an acute sample , and/or IgM seroconversion , and/or a four-fold or greater increase in HI antibody titers between acute and convalescent specimens was observed . Primary dengue virus infection: a confirmed dengue virus infection in which dengue virus IgG antibodies were not detected in the acute sample and an HI titer of ≤1:80 in the convalescent specimen was observed . When indeterminate by IgG and HI , cases were also classified as primary infections when PRNT50 seroconversion to at least one serotype was detected between acute PRNT-negative and convalescent specimens . Secondary dengue virus infection: a confirmed dengue virus infection in which dengue virus IgG antibodies or HI antibodies were detected in acute specimens or increased to ≥ 1:1280 in convalescent specimens . In instances indeterminate by IgG and HI , cases were classified as secondary infections when the presence of neutralization antibodies to any serotype in the acute specimens was detected . Tertiary dengue virus infection: the second confirmed dengue virus infection during a volunteers participation in the study when the individual had clear evidence ( by PRNT ) of prior dengue virus infection upon enrollment in the study . Clinical category: clinical data were analyzed using WHO 1997 criteria . Cases with evidence of plasma leakage ( hematocrit increase ≥20% , or pleural effusion/acites by ultrasonogram ) , but no thrombocytopenia <100 , 000/mm3 were categorized as unclassified . Asymptomatic dengue virus infection: An asymptomatic dengue virus infection was confirmed when there was no reported fever day or other illness and a four-fold or greater increase of PRNT50 titer in any serotype between two serosurvey specimens whose IgG IR ≥ 1 . 2 was observed . For primary asymptomatic cases , the infecting serotype was determined by the serotype with the highest PRNT50 titer in the second specimen . Naïve population: a subset of study participants whose serosurvey specimens did not show antibodies to dengue as verified by IgG index <1 . Pre-illness neutralizing antibody: pre-illness neutralizing antibodies , as measured by PRNT , were present when the titer >1:10 and considered protective when the titer was >1:100 [32] . Incidence of symptomatic and asymptomatic DEN virus infection was expressed as the number of infections occurring among the cohort per 1 , 000 person years of follow-up . Volunteers that dropped out from the study were accounted for in the denominator ( total person-years ) by including only the length of time they were available for follow-up . For comparison between two proportions , the chi-square test was used using STATA 9 software ( Texas ) .
The study was conducted in two phases: from August 2000 to June 2004 and from September 2006 to April 2009 . During each phase of the study the goal was to maintain a cohort of approximately 3 , 000 volunteers . Over the course of both phases , a total of 4 , 380 volunteers from three factories were enrolled in the study . Factory A was included in both phases of the study and 1324 volunteers from Factory A participated in both phases of the study . Twenty percent of the volunteers participated in the study for more than six years , 19% participated for 3–4 years , 26 . 4% for 2–3 years and 16 . 9% for 1–2 years . The mean ( SD ) age of volunteers at enrollment was 35 . 6 ( 7 . 7 ) with a range between 18 and 66 years . A higher proportion of the study population was male ( ratio 1 . 87: 1 ) . Demographics of the study population aggregated by factory are shown in Table 1 . The presence of asymptomatic infections was determined in 675 subjects randomly chosen from the September 2006 to February 2008 time period who did not experience an illness in between serosurveys . No evidence of a previous dengue virus infection was found in 15 ( 2 . 2% ) of these subjects . Serological evidence of a previous dengue virus infection was found in 87 . 53% of subjects aged 18–27 years , in 96 . 6% aged 28–37 years , in 99 . 1% aged 38–47 years , and in all subjects above ≥48 years old . This trend was found to be statistically significant with prevalence higher in older age categories ( Chi-squared test for trends , p<0 . 001 ) . Consecutive serosurvey samples were screened for asymptomatic infections using an IgG index ratio of ≥ 1 . 2 as a cut off . An IgG index ratio ( IR ) ≥ 1 . 2 between two consecutive serosurvey samples was found in 35 of 675 volunteers . After confirmatory PRNT testing , seven were excluded as no four-fold increase in any serotype was observed ( of note: all of the excluded cases had low IgG IRs ) . Three of the 28 asymptomatic cases were primary infections , two were due to DENV-4 and one was due to DENV-1 . We evaluated “pre-illness” PRNT titers for the 25 secondary infection asymptomatic cases . In four cases , the highest PRNT antibody titers to any serotype were below the level of the suggested protective neutralizing titer ( ≥1:100 ) [32] . In nine cases , protective neutralizing titers to one serotype were detected ( five to DENV-3 , two to DENV-1 and one each to DENV-2 and DENV-4 ) . In eight cases , protective neutralizing titers were detected for two serotypes ( six for DENV-1 and 3 , one for DENV-1 and 2 , and one for DENV-2 and 3 ) and protective neutralizing titers to three serotypes ( DENV-1 , 2 and 3 ) were identified in four cases . During the same time period , 43 symptomatic dengue virus infections occurred in the cohort resulting in an asymptomatic to symptomatic dengue virus infection ratio of 2 . 6:1 ( 95% CI:1 . 6–3 . 12 ) . The average age of subjects with asymptomatic dengue virus infection was significantly higher than subjects with symptomatic dengue ( 40 . 14 , 95% CI: 37 . 34–42 . 94 vs . 36 . 0 , 95% CI: 35 . 06–36 . 94 ) . The proportion of women with asymptomatic dengue virus infections was higher than the proportion with symptomatic dengue virus infections , but not significant ( 39 . 3% vs . 29 . 5% , p = 0 . 28 ) . A total of 2 , 167 febrile episodes occurred during the course of the study , which encompassed a total of 15 , 454 . 5 person months of observation . DENV infections were confirmed in 268 episodes , giving an overall proportion of dengue virus infection among fever patients of 12 . 4% . This proportion was less than 10% in 2002 , 2004 , 2006 to 2007 , while the highest proportions were observed in 2000 ( 41 . 2% ) and 2009 ( 26 . 6% ) . The overall incidence rate in the cohort was 1730/100 , 000 person years . The annual incidence rate was the lowest in 2006 with 630/100 , 000 person years and the highest in 2009 with 3780/100 , 000 person years ( Fig 2A ) . The incidence rate was highest in the youngest ( 18–27 years old ) age group ( 3984/100 , 000 ) , followed by the 28–37 years old age group ( 2449/100 , 000 ) , the >47 years old age group ( 1661/100 , 000 ) and the 38–47 years old age group ( 1384/100 , 000 ) . In general , cases started to increase during the rainy season in January , peaked in the first half of the year and then slowly decreased in the second half of the year ( Fig 2B ) . Of 268 symptomatic dengue cases , 92 were confirmed by virus isolation , RT-PCR and serology , 104 by RT-PCR and serology , and 72 only by serological assays . According to HI , IgG ELISA and PRNT antibodies , infections were classified as primary in 21 ( 7 . 8% ) cases and secondary in 247 ( 92 . 5% ) cases . In Bandung , all serotypes circulated most years of the study . DENV-2 was absent for 18 months from December 2001 until June 2003 . DENV-1 and DENV-2 were not detected for nine months from September 2006 to June 2007 and DENV-2 was absent from August 2007 to November 2008 . DENV-3 and DENV-4 , conversely , were more evenly distributed throughout the year . From a total of 196 cases where the serotype was identified , DENV-4 was the most frequent ( 28 . 6% ) , followed by DENV-3 ( 26 . 5% ) , DENV-2 ( 22 . 4% ) and DENV-1 ( 22 . 4% ) . The only month that all serotypes were detected among the cohort simultaneously was in March 2009 , at a time during the highest incidence of cases ( Fig 2B ) . In confirmed cases , IgM antibodies were positive in 7 . 9% ( 7/89 ) of subjects who came to the clinics on day two , 20 . 2% ( 18/89 ) on day three , 36 . 7% ( 18/49 ) on day four and 48 . 8% ( 20/41 ) on day five or more . In 15 . 74% ( 42/268 ) of the cases , IgM antibodies were never detected , not even in the convalescent specimens . All of the cases in which IgM antibodies were never detected were confirmed by HI and in 19 cases were also confirmed by RT-PCR . All of the cases in which IgM antibodies were never detected were secondary infections . Envelope gene sequences from isolates identified in this study and sequences available in GenBank were used to generate phylogenetic trees for DENV-1 , DENV-2 , DENV-3 , and DENV-4 ( Fig 3 ) . The similarity of envelope gene DENV-1 sequences within isolates from this study and other Indonesian isolates were between 98–99% and 94–98% , respectively . The similarity of amino acid sequence within this study and other Indonesian isolates was 98–99% and grouped to genotype IV . The similarity with an Indonesian isolate from 2007 ( gb/EU448401 ) was only 97% , resulting in different genotypes . The similarity of envelope gene DENV-2 sequences with Indonesian isolates from 1976 to 2010 was between 97–98% and the similarity of amino acid sequences was 99% . Genotype analysis grouped this isolate into the Cosmopolitan genotype . Three DENV-3 sequences in the study have similarity of around 96 . 8–98% with most Indonesian isolates . The similarity in amino acid sequence was 99% and grouped to genotype I . However , the sequence similarity compared to two Indonesian isolates from 1998 ( AY912454 , AY912455 ) was only 94 . 5% and the similarity in amino acid sequences was 98% , resulting in a different genotype . Genotype analysis of five DENV-4 isolates placed them in genotype II , together with other Indonesia isolates from 1973 to 2010 . The majority of cases were DF ( 78 . 4% ) , followed by DHF grade I ( 11 . 9% ) , DHF grade II ( 7 . 5% ) , unclassified ( 1 . 9% ) and DSS ( 0 . 4% ) . Since patients were advised to come early when they experienced fever , the mean days from fever onset was 3 . 2 ( ±1 . 2 ) days , ranging from day two to eight days . Symptoms and signs that were frequently reported included myalgia ( 91 . 3% ) , headache ( 90 . 9% ) , arthralgia ( 63 . 8% ) , nausea ( 59 . 6% ) and a positive tourniquet test ( 30 . 9% ) . Leukopenia ( <4000/mm3 ) was detected in 29% of patients and thrombocytopenia ( <150 , 000/mm3 ) was detected in 34% of patients . Leukopenia and thrombocytopenia by fever day is presented in Table 2 . The majority of DF cases presented with an undifferentiated fever; 38 . 6% ( 81/210 ) also presented with hemorrhagic signs or mild thrombocytopenia . There were no cases with complications such as organ impairment . Six patients experienced multiple confirmed dengue virus infections resulting in febrile illnesses during their participation in this study . Four of them have been reported elsewhere [33] . For the remaining two , one volunteer had evidence of three sequential dengue infections and one had evidence of two dengue infections ( Table 3 ) . The first case was a 33 year old male ( ID 50877 ) with evidence of a previous DENV-4 infection , followed by DENV-1 infection and a DENV-3 infection five and half years later . The clinical diagnosis was DF . The second case was a 28 year old male ( ID 50743 ) with a DENV-2 infection followed by a DENV-3 infection two years later . Clinically , both episodes manifested as DF with hemorrhagic manifestations that required hospitalization . Details for all six sequential dengue virus infections are presented in Table 3 . The distribution of clinical severity according to type of infection and infecting serotype is listed in Table 4 . Overall , 53 ( 19 . 8% ) cases were classified as a severe form of illness , DHF I , II and DSS . Ninety-six percent of individuals that experienced a severe form of dengue ( DHF and DSS ) had a secondary infection . The proportion of DHF and DSS among primary cases tended to be lower than secondary cases ( 9 . 5% vs 20 . 6% , p = 0 . 21 ) . The subjects that experienced a severe illness from a primary infection did not have any co-morbidities . There were 27 cases in which the serotype of the second infecting virus was determined by RT-PCR and the serotype of the previous dengue virus infection could be determined through pre-illness neutralizing antibody titers . Previous infections with DENV-1 were identified in five cases , DENV-2 in 19 cases , DENV-3 in one case and DENV-4 in two cases . Following a DENV-2 infection , infections with DENV-3 or DENV-4 resulted in more severe illness than DENV-1 ( 40% , 28 . 6% , and 14 . 3% , respectively ) . Two DHF cases occurred when DENV-1 was the first infecting virus ( 2/5 ) , each followed by DENV-2 or DENV-3 , and four cases when DENV-2 was the first infecting virus ( 4/19 ) , two followed by DENV-3 , one DENV-4 and one DENV-1 .
This prospective cohort study in West Java provides several important findings on the epidemiology of dengue virus infections in adults living in an endemic area . First , dengue virus is a major etiology of febrile illness ( 12 . 4% ) in adults in Bandung , West Java , Indonesia . Second , the average incidence rate of symptomatic laboratory confirmed dengue virus infection from 2000–2004 and 2006–2009 was 1734 cases/100 , 000 person year , or 43 times higher than the district rate ( 40/100 , 000 person year ) [34] . Lastly , between September 2006 and April 2008 asymptomatic infections were 2 . 6 times more frequent than symptomatic infections . The proportion of dengue virus infections among acute febrile patients in the outpatient setting has rarely been reported in Indonesia . A previous study from 1976 , which was a bacteriological and serological survey among febrile patients admitted to hospitals in Jakarta , revealed a similar proportion of dengue virus infections [35] . Studies from Malaysia , Myanmar and Thailand report proportions ranging between 5 . 7 to7% , suggesting dengue is more prevalent in Indonesia [36–38] . The incidence rate in our prospective cohort was calculated from the number of dengue cases that were identified in outpatient clinics , representing both mild and severe cases of dengue , whereas previously reported provincial and national incidence rates from Indonesia were based on the hospitalized cases , representing more severe cases . This could explain why we found an incidence rate 43 times higher than the reported provincial and national figures . This prospective cohort study was conducted in textile factories located in West Java , Indonesia . A limitation of this study is that there is a lack of information specifically about the adult population in Bandung . Thus , while we aimed to represent a working adult population , it is difficult to determine how accurately we did so . While it was not possible to determine where the participants were infected with dengue virus , it seems more likely that the participants were infected outside of the factory as every time a dengue virus infection was confirmed in a cohort member , fogging was immediately conducted at the factory . Our study revealed that dengue virus infections in adults were mostly uncomplicated DF ( 78 . 4% ) . This finding is different than previous reports that demonstrate severe cases are more predominant [16 , 17] . One of the reasons for this difference could be that our febrile patients came mostly from outpatients clinics whereas other studies enrolled patients who had indications for hospitalization ( usually with platelet count<100 , 000/mm3 ) . Also , we may have under-estimated the number of DHF cases because we relied heavily on serial hematocrit results which may be influenced by intravenous fluid therapy and we used a strict 20% hematocrit increase to confirm hemoconcentration . Although we also performed ultrasonography , at a maximum of once a day , it is probably not sufficient as plasma leakage is transient . It was not possible to classify five cases as the clinical manifestations did not fit with any of the categories . All showed plasma leakage but no thrombocytopenia was noted . For the clinical categories , we also analyzed using the 2009 WHO Criteria that was introduced to answer the difficulties in the use of the 1997 criteria . Our findings revealed that 52 of the 53 DHF cases and the 5 cases that were not possible to be classified would not be considered as severe dengue because the evidence of fluid accumulation was not accompanied by any required respiratory distress [39] . Based on the 2009 criteria , the clinical categories were 148 ( 55 . 2% ) dengue without warning signs ( WWS ) , 119 ( 44 . 4% ) dengue with warning signs ( WS ) and 1 ( 0 . 4% ) severe dengue . While the 1997 WHO criteria is criticized due to the difficulties in applying it in a clinical setting and the increasing clinically severe dengue cases that did not fulfill the strict criteria of DHF [39] , we found that some warning sign criteria in the 2009 WHO criteria are based on clinical judgment and therefore are non-standard ( i . e . “severe bleeding as evaluated by clinician” , “abdominal pain or tenderness” , “persistent vomiting” , and “lethargy” ) . As dengue with warning signs is indicated for hospitalization , researchers have raised concerns about the increasing hospitalization rates [40] . Besides the classical signs and symptoms such as fever , headache and myalgia , we found that leukopenia and thrombocytopenia only supported the dengue diagnosis if tested on day five or more of illness . Furthermore , the development of leukopenia and thrombocytopenia concurred with a sensitivity of DENV IgM antibodies above 50% . We observed an absence of detectable IgM antibodies in dengue cases in some of our participants ( 15 . 7% ) . This phenomenon has been previously reported with the percentages varying from 5 . 4% [41] , 22 . 1% [42] to 27 . 6% of convalescent specimens collected on day 7–14 of fever [43] . In our study population , DHF cases were predominantly secondary infections ( 96 . 2% ) . While the proportion of DHF cases due to a secondary infection was higher than in primary infections , the difference was not significant . DHF cases in adults have also been reported in Thailand , Martinique and Pakistan [44–46] In addition , we also identified tertiary infections in six patients , all presenting with DF . This finding , similar to a previous report from Thailand [47] , indicates that a previous infection with two dengue serotypes does not necessarily protect an individual against future clinical dengue infections . On the contrary , some level of protection has been demonstrated in several primates studies [48–51] , a hospital based study in Bangkok [52] , and a prospective cohort in Iquitos , Peru [53] . Our study demonstrated that no serotype was significantly more predominant and most serotypes circulated every year in Bandung . The fact that certain serotypes were not detected for some time could be due to the limited size of the study population . Only during the most intensive transmission period ( March 2009 ) were all serotypes identified in the cohort at the same time . Similar findings have been reported during other outbreaks in Indonesia [16 , 17] . The sequence of infecting serotypes has been associated with disease severity [54 , 55] . We found that most cases with well-characterized sequence serotypes had DENV-2 as the previous infecting serotype . Furthermore DHF more frequently occurred when DENV-3 ( 40% ) was the infecting serotype for the second infection compared to DENV-4 or DENV-1 , although this difference was not significant . Our data do not support previous findings from Thailand and Cuba that a DENV-2 infection after DENV-1 is a risk factor to develop severe dengue [54 , 55] . A conclusion cannot be drawn regarding other sequences , as the number of cases was too limited . We suspect that the majority of asymptomatic infections were the result of DENV-4 infections for the following reasons: first , two of the three primary asymptomatic infections were caused by DENV-4; second , all available pre-illness sera from 24 of 25 secondary asymptomatic infections revealed no or very low neutralizing antibodies to DENV-4; third DENV-4 , along with DENV-3 , were the predominant serotypes ( 39 . 4% and 42 . 4% , respectively ) identified in symptomatic cases during the same period in the same cohort . Asymptomatic or mild forms of disease resulting from DENV-4 infections have been reported before from Indonesia [56] and thought to be the reason for the scarcity of DENV-4 hospitalized cases [32] . Finally , our study shows that there are dynamic changes in the various genotypes of the circulating DENV in Indonesia . DENV-1 genotype IV that was detected in our study in 2003 , 2005 and 2008 was first detected in 1968 and since then has been endemic in various areas of Indonesia [57–59] . Regarding other genotypes , DENV-1 genotype I was first reported in Indonesia in 2007 ( gb:EU448401 ) and 2010 in Surabaya [58] and became the predominant genotype in Semarang in 2012 , while DENV-1 genotype II has begun to be detected again after it was last detected in 1964 in Thailand [60] . Monitoring the dynamic changes of DENVs is very important as it may relate to clinical severity and thus may have public health impact . For example , despite the high homology of DENV-1 isolates from our study with the strain from the 1998 Sumatera outbreak isolates , no significant rise in cases was reported during our study . A plausible explanation based on the genotype analysis was that the 1998 outbreak was caused by the introduction of DENV-1 genotype IV , but from a different clade . In comparison to the dynamic changes of DENV-1 genotypes , the circulating DENV-2 isolates in Indonesia from 1976 to 2012 and the isolates from our study were only grouped into the Cosmopolitan genotype [57 , 60] . This Cosmopolitan genotype has previously been associated with severe disease [61] . Similarly , the genotype I of DENV-3 in our study is the common genotype in Indonesia and has remained endemic since 1973 . However , during the1998 outbreak , the circulating DENV-3 was from genotype II . Our study was one of only a few studies in Indonesia that has successfully detected and isolated DENV-4 [60] . All DENV-4 isolates were similar to the Indonesia isolates from 1973–2010 , belonging to genotype II , which were different to the dominant circulating strain in Thailand ( genotype I ) [28] and cause mostly mild illness [62] . A limitation of our study is that the viruses selected for sequencing were from a convenience sample and were not selected to be representative . Thus , it is possible that other genotypes of all four dengue viruses may be circulating in this area . Also , the asymptomatic dengue infection was only based on 1 . 5 years of surveillance . However , this is similar to the ratio from the first two years of the study , using a different approach [25] and another study in Thailand [63] . In conclusion , our study was a population-based cohort study carried out in large factories in West Java , Indonesia with intense monitoring during nearly seven years . We were able to identify dengue virus infections at an early stage and those that presented with minimal symptoms thus providing accurate epidemiological data regarding the spectrum of dengue disease in adults . As dengue is a growing public health threat without effective preventive measures and an unclear pathogenesis of severe illness , further studies examining dengue virus infection in a natural setting need to be conducted . | Dengue is the fastest spreading mosquito borne diseases in the world and is endemic in most tropical and sub-tropical countries with an estimated 96 million infections resulting in clinical disease annually . Population-based longitudinal prospective studies are essential for understanding dengue virus in the natural setting and developing prevention strategies to curtail the impact of disease . We present the results of a longitudinal cohort study of adults in West Java , Indonesia that was conducted between 2000–2004 and 2006–2009 . We found that in adults , dengue virus was a significant cause of febrile illness . The entire spectrum of clinical severity was observed with most dengue virus infections manifesting as asymptomatic . The incidence of symptomatic dengue virus infections observed in this cohort was much higher than reported national or provincial rates . In addition , we observed all four dengue virus serotypes circulating during most years of the study . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"enzyme-linked",
"immunoassays",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"tropical",
"diseases",
"microbiology",
"geographical",
"locations",
"indonesia",
"java",
"viruses",
"age",
"groups",
"adults",
"rna",
"viruses",
"signs",
"and",
"symptoms",
"neglected",
"tropical",
"diseases",
"antibodies",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"proteins",
"medical",
"microbiology",
"dengue",
"fever",
"microbial",
"pathogens",
"immunoassays",
"people",
"and",
"places",
"biochemistry",
"asia",
"flaviviruses",
"fevers",
"viral",
"pathogens",
"physiology",
"oceania",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"viral",
"diseases",
"organisms"
]
| 2016 | The Epidemiology, Virology and Clinical Findings of Dengue Virus Infections in a Cohort of Indonesian Adults in Western Java |
Disruption of T cell memory during severe immune suppression results in reactivation of chronic viral infections , such as Epstein Barr virus ( EBV ) and Cytomegalovirus ( CMV ) . How different subsets of memory T cells contribute to the protective immunity against these viruses remains poorly defined . In this study we examined the compartmentalization of virus-specific , tissue resident memory CD8+ T cells in human lymphoid organs . This revealed two distinct populations of memory CD8+ T cells , that were CD69+CD103+ and CD69+CD103— , and were retained within the spleen and tonsils in the absence of recent T cell stimulation . These two types of memory cells were distinct not only in their phenotype and transcriptional profile , but also in their anatomical localization within tonsils and spleen . The EBV-specific , but not CMV-specific , CD8+ memory T cells preferentially accumulated in the tonsils and acquired a phenotype that ensured their retention at the epithelial sites where EBV replicates . In vitro studies revealed that the cytokine IL-15 can potentiate the retention of circulating effector memory CD8+ T cells by down-regulating the expression of sphingosine-1-phosphate receptor , required for T cell exit from tissues , and its transcriptional activator , Kruppel-like factor 2 ( KLF2 ) . Within the tonsils the expression of IL-15 was detected in regions where CD8+ T cells localized , further supporting a role for this cytokine in T cell retention . Together this study provides evidence for the compartmentalization of distinct types of resident memory T cells that could contribute to the long-term protection against persisting viral infections .
It has recently become evident that protective T cell immunity relies not only on circulating memory T cells but also on non-circulating resident memory populations [1–5] . These resident memory T ( Trm ) cells have been identified in a variety of different non-lymphoid and lymphoid tissues in mice [6–10] . Importantly , when compared to their circulating counterparts Trm cells provide superior protection against reinfection at their site of localisation [6 , 11–16] . The characteristics of these cells in humans however , are poorly understood . A greater understanding of the mechanisms that regulate their development and maintenance is paramount for future vaccine strategies . Residence within tissue environments depends upon the ability of T cells to overcome the egress signals . This is achieved by acquiring expression of receptors that enhance cellular interaction within the tissue and facilitate survival for prolonged periods within that tissue . The exit signal for T cells is largely mediated by the concentration gradient of sphingosine-1-phosphate and expression of its receptor , S1P1 , on T cells [17] . Accordingly , studies in mice have shown that Trm cells completely lack the expression of S1P1 as well as its transcriptional regulator KLF2 [18 , 19] . T cell exit through the efferent lymphatic system is facilitated by the expression of CCR7 . KLF2 is also known to positively regulate the transcription of CCR7 [20] , and therefore the loss of KLF2 may also abrogate the CCR7 mediated T cell exit . Trm cells are further distinguished from their circulating counterparts by constitutively expressing CD69 . This C-type lectin has traditionally been considered as a marker of T cell activation , but its role in promoting tissue residence , through binding to and down-modulating pre-existing S1P1 on the T cell surface , has only recently been recognized [21–23] . A second surface marker associated with tissue residence is CD103 , the alpha chain of the integrin αEβ7 which mediates T cell binding to E-cadherin expressed on epithelial tissues [24] . In the mouse , at least two distinct subsets of Trm cells have been identified based on the presence or absence of CD103 , with the CD103+ subset largely found at barrier surfaces [6 , 13 , 25–27] . To date , studies in man suggest that Trm cells are likely to be present in both lymphoid and non-lymphoid organs [28–32] . A recent study has shown that human skin is populated with at least two distinct memory T cell subsets that are non-circulating resident populations . This was demonstrated in patients who underwent alemtuzumab treatment , which selectively depleted the circulating T cell populations [29] . Most other studies however , have relied solely on the expression of CD69 as a Trm marker and , while large numbers of CD69+CD8+ T cells have been reported in human lymphoid organs [28 , 30] , the significance of such findings is difficult to judge . Firstly , it is not clear whether these CD69+CD8+ T cells possess other aspects of the Trm phenotype , such as loss of S1P1 and KLF2 , or whether they merely express CD69 as a result of recent activation . Furthermore a crucial feature attributed to Trm cells in mouse models , namely the ability of antigen-specific cells to persist at sites of potential antigen encounter [10] , has only been examined in human skin [33] or lungs [34] . Whether antigen-specific Trm cells accumulate in human lymphoid tissues is unknown . In that context , it is relevant to compare the distribution of CD8+ T cells to two common human herpesviruses , Epstein-Barr virus ( EBV ) and cytomegalovirus ( CMV ) in different human tissues . Both elicit numerically strong CD8+ T cell responses , but are harboured at different sites in vivo [35 , 36] . EBV persists as a latent infection of a memory B cell population that preferentially recirculates between the blood and oropharyngeal lymphoid tissues such as the tonsil [37] . Occasional reactivation from the latent reservoir is thought to seed foci of lytic infection in oropharyngeal epithelium , leading to periods of viral shedding into throat washings [36] . By contrast , CMV persists as a latent infection of the myeloid lineage with the capacity to reactivate to lytic infection at various tissue sites [35] . In this study we have focused on two human lymphoid tissues , namely spleen and tonsils , and have used CD69 and CD103 to identify two distinct subsets of memory T cells that are retained within these two organs in the absence of recent T cell activation . These two populations are transcriptionally distinct by S1P1 and KLF2 expression and have different anatomic distributions , with a selective retention of EBV-specific T cells in tonsillar tissues suggesting that they are strategically positioned at sites of possible antigen encounter . Our data also indicated the important roles for two locally-produced cytokines , IL-15 and TGF-β , in determining tissue residence of CD8+ T cells .
We chose to compare circulating T cells with T cell populations isolated from human spleens and tonsils , two lymphoid tissues that are important for infections acquired through blood and the oropharynx respectively . As previously reported [28] , CD69 expression was minimal on circulating T cells; however , was consistently detected on 25–75% of splenic and tonsillar CD8+ T cells ( Fig 1A ) . CD69 expression was largely confined to memory CD8+ T cell populations , with the CCR7—CD45RA—effector memory and CCR7—CD45RA+ TEMRA subsets together accounting for over 75% of CD69+CD8+ T cells in the spleen and over 90% in the tonsils ( Fig 1B ) . Since CD69 is also expressed on recently activated T cells , we examined CD8+ T cell subsets for early ( CD137 , CD25 ) or late ( HLA-DR ) T cell activation markers . When compared to CD69—CD8+ T cells however , there were no differences in the expression levels of these markers on CD69+ CD8+ T cells ( Fig 1C and 1D ) . In addition , recently activated effector T cells can be identified by their high expression of KLRG1 . Both CD69+ and CD69—CD8+ T cells in spleen ( Fig 1C ) and tonsils ( Fig 1D ) however , were largely KLRG1low . This demonstrated that CD69 was expressed on memory CD8+ T cells in the absence of recent T cell activation . Enhanced T cell survival capacity is also crucial for the persistence of memory T cells [38] . In this regard we determined the expression of the pro-survival gene BCL-2 in purified CD69+ and CD69—effector memory CD8+ T cells . Compared to CD69—memory CD8+ T cells , CD69+ memory CD8+ T cells expressed higher levels of BCL-2 ( Fig 1E ) , suggesting that CD69+ memory CD8+ T cells are better equipped for survival than their CD69—counterparts . Together these data clearly demonstrate that CD69 is expressed on memory CD8+ T cells in these lymphoid tissues in the absence of recent T cell activation . Since at least two distinct subsets of Trm cells can be identified in mouse tissues and human skin by differential expression of CD103 [6 , 13 , 26 , 27 , 39] , we also examined CD8+ T cells from human spleens and tonsils for co-expression of CD103 and CD69 . As shown in Fig 2A and 2B , in both tissues CD103 expression was largely restricted to the CD69+ subset; however , while in the spleen CD103+ cells made up only a small fraction of CD69+ population , in the tonsil these cells comprised around 50% of the CD69+ subset . As in the mouse therefore , human lymphoid tissues contain both CD69+CD103—and CD69+CD103+ subsets of CD8+ T cells . To gain further insight into possible differences between these two subsets , we characterized both populations phenotypically using the T cell differentiation markers CD45RA and CCR7 ( the latter known to be involved in T cell exit from peripheral tissues and T cell retention in lymph nodes [22 , 40] ) , and CD11a , the alpha chain of the integrin LFA-1 known to be associated with memory T cell retention in murine tissues [41] . As shown in Fig 2C and 2D , while CD69+CD103—CD8+ T cells were divergent in terms of CD45RA and CCR7 expression , the CD69+CD103+ CD8+ T cell subset was uniformly CD45RA—CCR7—in both the spleen and tonsils . The expression levels of CD11a varied between the organs . In the spleen there were no major differences between the three subsets , however in the tonsils both CD69+ subsets expressed higher levels when compared to the CD69—subset . Studies in mice suggested that Trm cells have high levels of PD-1 and therefore we asked whether any of the subsets in humans expressed PD-1 . We also examined TIM-3 and BTLA , two markers often associated with cell exhaustion . These showed that CD69+CD103+CD8+ T cells had the highest level of PD-1 , followed by CD69+CD103-CD8+ T cells ( Fig 2E ) . Neither of these subsets however , appeared to express both TIM-3 and BTLA ( Fig 2E ) , suggesting that they are unlikely to be exhausted cells . To extend the analysis of phenotypic differences between CD69+CD103+ and CD69+CD103—memory cell subsets , we next assessed expression of S1P1 and KLF2 , both of which are down-regulated in , and are thus a feature of Trm cells [18] . We purified CD103+CD69+ , CD103—CD69+ and CD103—CD69—CCR7—memory CD8+ T cells from spleen and tonsils by sorting and then quantified mRNA levels by Q-PCR . This showed that the CD69+CD103+ subset of memory T cells had significantly down-regulated transcription of both markers , whereas the CD69+CD103—subset had down-regulated S1PR1 substantially but KLF2 only partially ( Fig 2F ) . These data suggested that although both CD69+CD103+ and CD69+CD103—populations are likely to be retained within the tissues , the mechanisms that regulate their retention could be different . To characterize further the differences between CD103—CD69+ and CD103+CD69+ CD8+ T cell populations , we examined the anatomical location of these subsets . Immunofluorescence microscopy showed clear differences in the localization of these two subsets in tonsils and spleen . Fig 3A shows the overview of the stains in tonsils and spleen . Higher magnification of tonsillor sub-epithelial region ( area 1 ) and extra-follicular region ( area 2 ) revealed that CD103+CD69+CD8+ T cells preferentially localized near the epithelial barrier surface while CD103—CD69+CD8+ T cells were largely localized in the extra-follicular regions ( Fig 3B ) . Quantitative analysis of the proportion of CD103+ T cells near the epithelium confirmed that the majority of this subset was indeed localized near the barrier surface ( Fig 3C ) . In the spleen there were only a few CD69+CD103+CD8+ T cells , and these were located within the red pulp ( area 1 ) ( Fig 3D ) . The CD103—CD69+CD8+ T cells however , were localized within the periarteriolar lymphoid sheaths ( PALS ) ( area 2 ) ( Fig 3D ) . Together this reveals that CD103—CD69+ and CD103+CD69+ CD8+ T cells localize to different anatomical locations within human lymphoid tissues . Having identified two distinct subsets of CD8+ T cells that were retained within human lymphoid tissues , we wanted to determine the factors that influenced their retention . Studies in mice have shown that the maturation of Trm T cells is largely determined by cytokine exposure [1 , 19 , 27] . We therefore asked whether cytokines could potentiate the retention of CD8+ T cells in humans by inducing a Trm-phenotype in circulating CD8+ T cells . To this end , CD8+ T cells from peripheral blood were stimulated for 7 days either with candidate cytokines or with T cell activation and expansion beads ( TAE ) as a polyclonal TCR stimulus and the expression of CD69 and CD103 examined by flow cytometry . Stimulation with IL-15 alone consistently up-regulated CD69 on resting human CD8+ T cells ( Fig 4A and 4B ) including a small population of CD103+CD69+ CD8+ T cells ( Fig 4A ) . By contrast , type I interferons IFN-α and IFN-β , which in mice have been shown to induce CD69 expression on T cells , and IL-2 had little effect in these experiments ( Fig 4B ) . Although TGF-β failed to induce CD69 expression by itself , together with IL-15 , TGF-β induced a small proportion of CD103+CD69+ CD8+ T cells ( Fig 4A and 4B ) . Stimulation through the TCR and co-stimulatory receptors using TAE beads resulted in a large proportion of CD69+ and small proportion of CD103+ populations ( Fig 4A ) . Importantly , in contrast to this polyclonal TCR stimulation which up-regulated activation markers such as CD137 , IL-15 induced CD69 expression in the absence of CD137 expression ( Fig 4C ) . In addition , CD8+ T cells isolated from the spleen not only proliferated in response to IL-15 , but also maintained the expression of CD69 ( S1A Fig ) . In order to determine which subsets of circulating CD8+ T cells up-regulated CD69 in response to IL-15 simulation , we purified peripheral blood naïve ( CCR7+CD45RA+ ) , central memory ( TCM , CCR7+CD45RA— ) , effector memory ( TEM , CCR7—CD45RA— ) , and TEMRA ( CCR7—CD45RA+ ) CD8+ cells from blood by cell sorting , labeled them with cell trace violet ( CTV ) to track cell proliferation and stimulated them for 7 days with IL-15 . This revealed that although all 4 subsets responded to IL-15 and underwent robust proliferation , expression of CD69 was mainly induced on TEM or TEMRA CD8+ T cells ( Fig 4D ) . These data suggest that effector memory CD8+ T cells may be more responsive to IL-15-induced CD69 expression , which is consistent with the greatest proportion of CD69+ cells within the TEM subset in human tonsils and spleens ( Fig 1B ) . We next determined whether IL-15 , TGF-β and IL-2 could induce the transcriptional down-regulation of S1PR1 and KLF2 in human CD8+ T cells . We stimulated peripheral blood CD8+ T cells with IL-15 or IL-2 in the presence or absence of TGF-β for 7 days , purified CD69+ and CD69—CD8+ T cells by cell sorting , isolated their RNA and determined expression levels of KLF2 and S1PR1 by Q-PCR . As shown in Fig 4E and 4F , IL-15 markedly down-regulated S1PR1 transcription in the cells that had converted to CD69+ status , whereas its effects on KLF2 expression were partial and amplified considerably when TGF-β was also present . Similarly , IL-2 also down-regulated both KLF2 and S1PR1 , however , the presence of TGF-β did not result in further down-regulation ( S1B Fig ) . In order to determine whether IL-15 induced CD69+CD8+ T cells were unresponsive to S1P gradient we performed trans-well migration assay . This showed that the migration of CD69+CD8+ T cells towards S1P was significantly lower when compared to their CD69—CD8+ T cell counterparts ( Fig 4G ) , despite being able to migrate efficiently to CCL5 ( Fig 4H ) . In view of the above findings , we then examined whether IL-15 expressing cells were present within human lymphoid tissues . Frozen sections of tonsils were stained for IL-15 , CD8+ T cells and B cells . As recently observed in IL-15 reporter mice [42] , IL-15 producing cells were abundant in T cell areas of human tonsils , but largely absent from B cell follicles ( Fig 5 ) . In addition , IL-15 was detected in the squamous epithelial cells lining the tonsils . These data provide evidence for the constitutive expression of IL-15 within human tonsils . In order to understand the significance of CD69+CD103+ and CD69+CD103—CD8+ T cell subsets that persist in human lymphoid tissues , we examined the tissue distribution and phenotype of CD8+ memory T cells specific for two common viral pathogens , the B-cell tropic EBV and largely myeloid cell-tropic CMV . The typical values , including that of some paired blood-tissue samples , of the proportion of EBV and CMV-specific CD8+ T cells in the blood , spleen and tonsils suggested that EBV-specific CD8+ T cells preferentially accumulated in the tonsils , while CMV-specific CD8+ T cells remained in the blood ( Fig 6A ) . This prompted us to examine the phenotype of the virus-specific cells in these tissues to determine the proportion of Trm cells . While both EBV-specific and CMV-specific CD8+ T cells in circulation were CD69—CD103— ( Fig 6B ) , a significant proportion of EBV-specific and a small proportion of CMV-specific CD8+ T cells in the spleen expressed CD69 ( Fig 6B and 6C ) . Strikingly , neither EBV nor CMV-specific CD8+ T cells in the spleen expressed CD103 ( Fig 6B and 6C ) . By contrast , large proportions of EBV-specific CD8+ T cells in the tonsils were CD69+CD103+ and the small fraction of CMV-specific cells found in the tonsils remained CD69—CD103— ( Fig 6C ) . The expression levels of late activation marker , HLA-DR was similar between CD69+ and CD69—EBV-specific cells in the tonsils ( Fig 6D ) , suggesting that the CD69 expression was not due to recent T cell activation . These observations revealed that there was selective accumulation of CD69+CD103+ EBV-specific CD8+ T cells in the tonsils .
Maintenance of memory CD8+ T cells at appropriate anatomical sites appears to be crucial for optimal protection against recurrent virus infections . In this study we demonstrate that two distinct subsets of memory CD8+ T cells , expressing markers of tissue residence but not of recent activation , are retained within human lymphoid tissues such as tonsils and spleen . These two subsets are not only phenotypically distinct , but also anatomically separate within the tissue environment . CD8+ T cell memory to different viruses are differently distributed between the two subsets and show different patterns of tissue retention , with memory to an oropharyngeally-replicating virus acquiring a phenotype that is specifically retained at sites of possible virus reactivation within the tonsil . We also show that IL-15 and TGF-β can potentiate the retention of memory T cells in tissues . Our initial experiments showed that CD69 is expressed on large proportions of CD8+ T cells in human tonsils and spleen in the absence of recent T cell stimulation . Previous studies have also reported similar high proportions of CD69+CD8+ T cells in human lymphoid tissues but left open the question whether CD69 was simply a marker of recent activation rather than tissue residence . The present study , showing that such CD69+ cells lack a range of T cell activation markers , makes it clear that their expression of CD69 reflects tissue retention or tissue-residency . Our data also show that this population is itself heterogeneous , with CD103 as a marker that differentiates at least two distinct subsets . CD103 is an adhesion molecule that has already been associated with Trm cells in the skin , brain , gut and reproductive tract in mice [6 , 7 , 9 , 13 , 19 , 25 , 43] . Although CD103 expression is not a requirement for Trm cell formation [27 , 43] , its presence on the membrane enables T cells to bind to epithelial surfaces where its ligand E-cadherin is expressed [24] . We have identified crucial differences between the CD103+CD69+ and CD103-CD69+ CD8+ T cell populations in human lymphoid tissues . Both subsets exhibited dramatic down-regulation in expression of S1P1 . However , while there was a corresponding strong down-regulation of KLF2 in the CD103+ subset , this was only partial in the CD103- subset . In line with this , the CD103+ subset was uniformly CCR7— , consistent with KLF2 also regulating expression of CCR7 [20] . In addition , the CD69+CD103+ subset preferentially localized to epithelial barrier . Therefore our data from human tissues not only demonstrates the differences between CD69+CD103+ and CD69+CD103—subsets , but also indicates that the CD69+CD103+ subset is more typical of the Trm cell populations described at barrier surfaces in mouse models . Emerging evidence from different mouse models suggest that Trm cell formation is a two-step process [1] . The first step is the infiltration of a memory precursor population from blood into the tissue; a process that may or may not depend on the presence of antigen [7 , 9 , 12 , 13 , 27 , 44] . In the second step , the infiltrated precursor cells mature to become Trm cells through a process that largely depends on cytokines [19 , 27] . During this maturation process the CD8+ T cells not only express CD69 and CD103 , but also substantially down regulate KLF2 and S1P1 [12 , 19] . To this end , IL-15 and TGF-β have been implicated in the development of Trm cells in mouse models , although the mechanism by which IL-15 enables Trm cell formation is unclear [19] . Here we provide evidence for a key role for these two cytokines in humans . Firstly , we demonstrate that IL-15 alone can up-regulate CD69 , probably by down-modulating S1P1 and KLF2 . Our data however , also reveals that down-regulation of KLF2 was only partial in the presence of IL-15 alone and TGF-β was required for the complete down-regulation of KLF2 . IL-15 is a known growth factor for memory T cells and it enables T cells to persist in the absence of continuous antigen stimulation [45 , 46] . Therefore IL-15 is a likely candidate to retain and maintain a pool of memory T cells . Although it has been known that epithelial cells produce TGF-β , here we also demonstrate the presence of IL-15 producing cells at the sites where T cells are retained . We therefore propose that IL-15 and TGF-β could be the key regulators of CD69+CD103+ and CD69+CD103- subsets in human lymphoid tissues . Further insight into the differences between the two subsets came when the specificity of CD8+ T cells in human spleen and tonsils was examined . Although EBV persists within the memory B cell compartment , it largely remains as a true latent infection in these cells , and the evidence for its reactivation outside of oropharynx is limited . For example , the viral load in the spleen of healthy virus carriers was 20-fold lower than that observed in their peripheral blood [47] , suggesting that it is unlikely that regular viral replication takes place at sites such as the spleen . By contrast , the lymphoid tissues associated with the Waldeyer’s ring in the oropharynx are likely sites where virus reactivation takes place , probably initiated when virus-infected B cells infiltrate these tissues and the infection switches from latency into lytic cycle to seed foci of replication in permissive oropharyngeal epithelium [37 , 48 , 49] . This explains the occasional bouts of clinically silent virus shedding detectable in the throat washings of healthy virus carriers [36] . Such shedding is clearly under some form of T cell control , since levels of shedding are significantly raised in T cell-compromised individuals [36 , 50] . In this context the phenotype of EBV-specific CD8+ T cells in the tonsils suggests that more than a third of them were CD69+CD103+ and therefore are likely to be positioned near the epithelial barrier where EBV reactivates . Furthermore , the CD103+CD69+ EBV-specific CD8+ T cells were absent in the spleen and the few CMV-specific CD8+ T cells that were retained in the tonsils and spleen were largely CD103— . Our data reveal that CD103+CD69+ EBV-specific CD8+ T cells are selectively retained at sites of possible antigen encounter , which is consistent with the Trm T cell function implicated in mouse models [10] . Accumulation of virus-specific CD8+ T cells at sites of viral replication has been observed in other infections as well [11 , 34] . Here we show the distinct compartmentalization of different virus-specific resident memory CD8+ T cells that could be a crucial strategy for sustained protective immunity to pathogens . Our study also provides evidence for locally produced cytokines to potentiate the formation and positioning of Trm T cells . Taking together , the evidence we provide here suggest that the development of EBV-specific resident memory CD8+ T cells within the tonsils is likely to be influenced by the virus and the local environment . During active EBV replication , virus-specific effector CD8+ T cells are recruited to the tonsils . These infiltrating EBV-specific CD8+ T cells do not express CD103 [51] . Overtime , under the influence of IL-15 and TGF-β these effector cells are likely to mature to become CD69+CD103+ resident memory T cells , positioned along the epithelial barrier . This is supported by the finding that CD103+ EBV-specific CD8+ T cells within the tonsils only appear during convalescence [51] . The positioning of CD69+CD103+CD8+ T cells is likely to be facilitated by presence of E-cadherin at these sites [52] . More importantly , when compared to their circulating counterparts these CD103+ EBV-specific CD8+ T cells were highly reactive against EBV antigens [52] , suggesting that the primary role of these Trm cells could be to prevent or limit the viral replication at these sites . This strategic positioning of EBV-specific Trm cells is reminiscent of herpes simplex virus ( HSV ) -specific Trm cells in human skin . A recent study has demonstrated that similar persistence of HSV-specific CD8+ Trm cells at sites of viral reactivation is crucial for the containment of the virus [33] . Therefore the fact that EBV reactivation is often asymptomatic or sub-clinical could be largely due to this effective control at the sites where the virus reactivates .
Buffy coats from healthy blood donors along with spleens and paired blood samples from cadaveric organ donors were obtained from the Australian Red Cross Blood Services . Tonsils were obtained from patients undergoing routine tonsillectomy . Spleen and tonsil specimens were disaggregated to single-cell suspensions and the mononuclear cells were isolated using the standard Ficoll separation method . Cells were either used fresh or cryopreserved in liquid nitrogen for later work . All human experiments were approved by jurisdictional ethics committees in Sydney as well as the institutional review boards . Approval for this study was obtained from human ethics committees of the Royal Price Alfred Hospital , St Vincent’s hospital and Sydney South West Area Health Services ( Australia ) . Informed written consent was obtained from next of kin . Mononuclear cells from blood , spleen and tonsils were stained with fluorochrome-conjugated antibodies ( mAbs ) specific for cell surface proteins . The following mAbs were used for identification of CD8+ T cells and the determination of their phenotype; anti-CD3 , anti-CD8 ( Biolegend ) , anti-CCR7 ( R&D Systems ) , anti-CD45RA ( BD Biosciences ) , anti-CD69 , anti-CD25 , anti-CD137 , anti-HLA-DR , anti-CD103 , anti-KLRG-1 and anti-CD11a ( all obtained from Biolegend ) . Fluorochrome-conjugated HLA class I dextramers ( Immudex ) were used to identify virus-specific CD8+ T cells . EBV-specific CD8+ T cells were identified with dextramers specific for the following viral epitopes; GLCTLVAML ( derived from EBV-lytic protein BMLF1 ) , CLGGLLTMV ( derived from EBV-latent protein LMP2 ) , RAKFKQLL ( derived from EBV-lytic protein BZLF1 ) and FLRGRAYGL ( derived from EBV-latent protein EBNA3A ) . CMV-specific CD8+ T cells were identified with dextramers specific for the following viral epitopes; NLVPMVATV , TPRVTGGGAM , RPHERNGFTV ( all derived from pp65 protein ) , VLEETSVML , ELRRKMMYM , ELKRKMMYM ( all derived from IE-1 protein ) and VTEHDTLLY ( derived from pp50 protein ) . Stained cells were analyzed on either FACSCanto II or LSRFortessa flow cytometer ( BD Biosciences ) and the data processed using FlowJo software ( Treestar , Ashland , USA ) . RNA was isolated immediately after ex vivo purification of T cells or from cells after 7 days of culture using RNeasy kit ( Qiagen ) . Total RNA was then reverse transcribed with oligo-dT . For BCL2 , the following Real-time PCR primer set was used; forward , 5’–ttgacagaggatcatgctgtactt– 3’ and reverse , 5’–atctttatttcatgaggcacgtt- 3’ . Q-PCR was performed with assorted commercially available Taqman assays ( Hs00824723_m1 , Hs00984230_m1 , Hs02800695_m1 , Hs00173499_m1 , Hs00360439_g1 ) and Taqman Fast Advanced Mastermix on a StepOnePlus Real-Time PCR cycler ( Life Technologies ) . The threshold cycle of S1PR1 and KLF2 for each cell population was normalized to the arithmetic mean of HPRT , B2M and UBC housekeeping genes ( ΔCt ) . Normalized gene expression of each cell type was compared to the gene expression of a reference population with expression set to 1 according to the 2 ( -ΔΔCT ) method . Frozen sections of the tonsils and spleens were acetone-fixed and stained for different markers with mAbs using standard protocols [25] . The following primary mAbs were used to identify lymphocytes , anti-CD8 ( Abcam ) , anti-CD3 ( Biolegend/ AbD Serotec ) ( to reveal CD8+ T cells ) and anti-IgM ( Life technologies ) ( to reveal B cells ) . The expression of CD69 and CD103 was identified using purified , flurophore-conjugated or biotin-conjugated anti-CD69 mAb ( BioLegend ) and anti-CD103 mAbs ( BD Biosciences ) . The presence of IL-15 producing cells was determined using anti-IL-15 mAb ( Abcam ) . The following secondary antibodies were used to reveal specific staining; affinity purified F ( ab’ ) 2 fragments of AF647 or AF488 or CyTM3 conjugated donkey anti-mouse IgG and CyTM3 conjugated donkey anti-rabbit IgG ( all obtained from Jackson ImmunoResearch ) . Control antibodies or secondary mAbs in the absence of primary mAbs were used to determine background fluorescent levels . Images were acquired using Delta Vision Personal ( Olympus ) or Zeiss LSM700 microscope and analysed Imaris software ( Bitplane ) . CD8+ T cells were isolated from PBMCs using magnetic separation kit ( Dynal ) . Isolated cells or total PBMCs were cultured for 7 days in the presence of cytokines ( 50 ng/ml of IL-15 , IFN-α , IFN-β , TGF-β and 50 U/ml of IL-2 ) or with T cell activation and expansion beads ( TAE; anti-CD3/CD28/CD2 mAb micro beards , Miltenyi Biotech; Polyclonal stimulation ) . Different doses of IL-15 ( 1 , 10 or 50 ng/ml ) were used for stimulation of virus-specific CD8+ T cells . For experiments where T cell proliferation was measured , 1–2 x 106 purified CD8+ T cells were labeled with CellTrace Violet ( CTV; Invitrogen ) prior to cell culture . The proliferative history was determined based on the dilution of CTV of the T cells after stimulation . Sorted CD69+ and CD69—CD8+ T cells were washed in RPMI with 0 . 05% fatty-acid-free BSA ( Sigma ) and tested for transmigration across gelatin coated 5 μm transwell filters ( Corning ) for 4 hours to Shingosine-1-phosphate ( S1P ) ( Sigma ) or CCL5 ( R&D Systems ) . Migrated cell numbers were enumerated by flow cytometry . | Some viruses have the capacity to establish chronic infections in humans . How different T cell populations effectively control these infections has not been clear . Continuous circulation of memory T cells was thought to be crucial for effective immune surveillance against such infections . Recent studies in mice however , have shown that non-circulating tissue resident memory populations can also contribute to protective immunity . In this study we have examined the distribution , localization and characteristics of Epstein-Barr virus and Cytomegalovirus-specific T cells in different human tissues . This showed that virus-specific T cells were differentially distributed in different tissues and there was preferential accumulation of EBV-specific resident memory T cells at sites where EBV reactivates . In vitro studies showed that IL-15 and TGF-β could cooperate to extinguish tissue exit signals in T cells and therefore potentiate their retention within tissues . IL-15 expression was also detected in areas where T cells aggregated within the tissue . Together our study provides insight into how distinct memory T cells are compartmentalized in tissues to maintain long-term protection against persisting viral infections . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"blood",
"cells",
"flow",
"cytometry",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"spleen",
"immunology",
"microbiology",
"throat",
"developmental",
"biology",
"molecular",
"development",
"cytotoxic",
"t",
"cells",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"memory",
"t",
"cells",
"animal",
"cells",
"t",
"cells",
"viral",
"replication",
"spectrophotometry",
"immune",
"system",
"cytophotometry",
"cell",
"biology",
"anatomy",
"tonsils",
"virology",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"neck",
"spectrum",
"analysis",
"techniques"
]
| 2016 | Compartmentalization of Total and Virus-Specific Tissue-Resident Memory CD8+ T Cells in Human Lymphoid Organs |
Reconstruction of the regulatory network is an important step in understanding how organisms control the expression of gene products and therefore phenotypes . Recent studies have pointed out the importance of regulatory network plasticity in bacterial adaptation and evolution . The evolution of such networks within and outside the species boundary is however still obscure . Sinorhizobium meliloti is an ideal species for such study , having three large replicons , many genomes available and a significant knowledge of its transcription factors ( TF ) . Each replicon has a specific functional and evolutionary mark; which might also emerge from the analysis of their regulatory signatures . Here we have studied the plasticity of the regulatory network within and outside the S . meliloti species , looking for the presence of 41 TFs binding motifs in 51 strains and 5 related rhizobial species . We have detected a preference of several TFs for one of the three replicons , and the function of regulated genes was found to be in accordance with the overall replicon functional signature: house-keeping functions for the chromosome , metabolism for the chromid , symbiosis for the megaplasmid . This therefore suggests a replicon-specific wiring of the regulatory network in the S . meliloti species . At the same time a significant part of the predicted regulatory network is shared between the chromosome and the chromid , thus adding an additional layer by which the chromid integrates itself in the core genome . Furthermore , the regulatory network distance was found to be correlated with both promoter regions and accessory genome evolution inside the species , indicating that both pangenome compartments are involved in the regulatory network evolution . We also observed that genes which are not included in the species regulatory network are more likely to belong to the accessory genome , indicating that regulatory interactions should also be considered to predict gene conservation in bacterial pangenomes .
Regulation of gene expression is recognized as a key component in the cellular response to the environment . This is especially true in the microbial world , for two reasons: bacterial cells are often under severe energy constraints , the most important being protein translation [1] and they usually face a vast range of environmental and physiological conditions; being able to efficiently and readily react to ever changing conditions can most certainly give a selective advantage over competitors and give rise to specific regulatory networks . Transcription is mainly regulated by proteins , called transcription factors ( TF ) , which usually contain a protein domain capable of binding to specific DNA sequences , called TF binding sites ( TFBS ) . Depending on the position of the TFBS with respect to the transcriptional start site of the regulated gene , the TF can act either as a transcriptional activator or a repressor , mostly because of its interaction with the RNA polymerase and sigma factors [2 , 3] . The binding of the TF to its cognate TFBS is based on non-covalent interactions whose strength is indicated by the so-called affinity constant . Since TFBS can have variations around a preferred sequence , the affinity of a TF for its TFBSs covers a continuous range of values; however , since the TF binding strength appears to follow a sigmoid behaviour , it is possible to distinguish between ‘weak’ and ‘strong’ TFBSs [4] . As opposed to eukaryotic species , prokaryotic TFBSs are usually distinguishable from the ‘background DNA’ , and they tend to have a simpler structure and a close proximity to the transcription start site [5] . The application of information theory concepts to TFBS identification and analysis , revealed that specificity of the TF for a certain TFBS depends on the length , variability and composition of the TFBS itself with respect to the overall genomic background ( i . e . the sequence composition ) . Intuitively , the minimum information content able to provide specific recognition of the TFBS by the TF mostly depends on the genome size and its composition; increasing the size of the genome clearly increases the number of putatively non-functional TFBSs , and when the TFBS bases composition is close to the background DNA composition it may be impossible to discern a true functional TFBS from the surrounding DNA . Transcription factors recognizing TFBS characterized by low information content usually control the transcription of many genes across the genome; alternative sigma factors usually belong to this class , and their TFBSs also show larger variability between species [5] . Gene targets of these TFs are harder to reliably predict , for the presence of many non-functional sites along the genome . The high gene density of bacterial genomes and its organization in operons results in specific expression or repression of whole functional pathways in response to stimuli . Furthermore , the presence of several TFBSs in the upstream region of a gene can result in a complex transcriptional response that recall the behaviour of logic gates [6] . Prediction of TFBSs in a genome usually relies on the availability of a position specific scoring matrix ( PSSM ) storing the frequency of each nucleotide at each position of a TFBS . PSSM modelling the variability of a TFBS can be built by identifying enriched DNA patterns in promoter regions of genes that are known to be under the control of the TF under analysis , better if guided by other assays , like the binding of the TF to synthetic nucleotides . Several algorithms have been developed to use such PSSM to search for TFBSs in nucleotide sequences , such as the MEME suite [7] , RSAT [8–10] and the Bio . motif package [11] . A recent alternative method relies on the construction of a hidden markov model ( HMM ) from an alignment of nucleotide sequences , which can then be used to scan a query nucleotide sequence [12–14] . Since all these methods and their implementations have different weaknesses , it has been advised to use their combination to run predictions [15] . Regulatory networks evolve rapidly , making the comparisons between distant organisms difficult [16–19] . At broad phylogenetic distances , it has been shown that the conservation of a TF is lower than its targets [16] . Additionally , species with similar lifestyles tend to show conservation of regulatory network motifs , despite significant variability in the gene composition of the network , suggesting an evolutionary pressure towards the emergence of certain regulatory logics [16] . The fluidity of most transcriptional regulatory connections is well known and documented , not only at large phylogenetic distances , but also at the level of intra-species comparisons too [20–23] . Experiments have shown that Bacteria have high tolerance towards changes in the regulatory circuitry , making them potentially able to exploit even radical changes to the regulatory network , without extensive changes in phenotypes [24] . However , this is strongly dependent on which regulatory interaction undergoes changes , since there are also examples where a single change determines an observable difference in phenotype [25 , 26] . Bacteria have therefore a mixture of robust and fragile edges in their regulatory networks and evolution can play with them at different extent to explore: i ) the function of new genes , by integrating them in the old gene regulatory network , and ii ) if genes that are part of the gene regulatory network can be removed without harm to the physiology of the cell . The extent of variability and evolution of the regulatory network inside a species is , however , still poorly understood . The aim of this study is a comparative genomics analysis of regulatory networks , to understand the impact of regulatory network variability on pangenome evolution . We decided to use the Sinorhizobium meliloti species , the nitrogen-fixing symbiont of plants from the genus Medicago . S . meliloti has been deeply investigated as a model for symbiotic interaction and an extensive knowledge on its TFs is present in the literature [27 , 28] . This species presents a marked genomic difference with respect to other well-know bacterial model species , such as Escherichia coli , since S . meliloti genome comprises three replicons of comparable size: a chromosome , a chromid [29] and a megaplasmid , characterized by functionally and evolutionary distinct signatures [30 , 31] . This arrangement raises the question of how TF targets are distributed over the replicons . Recent reports have shown that there are only two genes essential for growth in minimal media and soil encoded in the S . meliloti chromid [32] , even though the chromid harbours many genes shared by all sequenced strains of S . meliloti species . Moreover , S . meliloti has several genomes sequenced to date [23 , 30 , 33–39] and the potential for biotechnological and agricultural applications , which could benefit from this analysis . At the comparative genomics level , different strains show quite a high level of variation . Indeed , the pangenome ( the collection of all genes from different strains [40] ) of this species has an abundant fraction of genes common to all members of the species ( termed core genome , as opposed to the strain-exclusive and/or partially shared fraction , called accessory genome ) of around 5000 gene families; approximately 40% of the genome belongs to the accessory fraction [31 , 35] . A preliminary analysis revealed that some of the TFs of the core genome also control genes of the accessory genome [23] . This allowed to propose that , when comparing the same regulon in different strains , we can define a panregulon , including a set of core ( shared ) target genes and an accessory ( variable ) regulon fraction [23] . It should be noticed that while the core regulon is necessarily formed by genes belonging to the core genome , the opposite can also be true ( i . e . that a gene belonging to the core genome belongs to the accessory regulon ) . However , the dynamics of the panregulon in relation to the evolutionary rules controlling the variability of the accessory regulon fraction are still not understood . We have therefore constructed the regulatory network of the S . meliloti species , using the PSSMs of 41 TFs collected from the literature and public databases . We have applied a combination of TFBS prediction methods , combining their output with information about the core and accessory gene families . We have also predicted the presence of the same TFBSs in five other closely related rhizobial species ( termed ‘outgroups’: Rhizobium leguminosarum bv . viciae , Rhizobium etli , Mesorhizobium loti , Sinorhizobium fredii and Sinorhizobium medicae ) . This regulatory network has been used to highlight the different behaviours that are present within and between species . Our predictions and other comparative genomics observations are publicly available ( https://github . com/combogenomics/rhizoreg/ ) .
Based on COG annotations , all the 51 S . meliloti strains analysed in this study , have been found to encode a similar number of predicted TFs ( an average of 522 ) ; a similar number has been also found in the five outgroups ( an average of 533 ) . This is in accordance with previous reports correlating genome size with the number of TFs [41] . Rhizobia belonging to the Alphaproteobacteria class ( alpha-rhizobia ) , which are known to have larger genomes compared to other bacteria from the same class [42] , have then one of the largest collection of TFs in the known bacterial kingdom . As the accessory genome accounts for about 40% of the proteome size [31 , 35] , it is reasonable to expect that a similar proportion of TFs will belong to the accessory genome . Indeed , about 70% of the TFs encoded in the S . meliloti pangenome belong to the core genome , while the remaining TFs are present in 1–3 genomes only; this orthologous genes distribution is similar to the one observed for the whole pangenome [43] ( S1 Fig ) . However , most of the 41 TFs analyzed in this study were found to belong to the core genome ( 37 ) , with the only notable exception represented by RhrA , the activator of the rhizobactin regulon , which is absent in 35% of the strains under study , confirming previous analysis [23 , 44 , 45] . More interestingly , recent reports have demonstrated how the presence of the rhizobactin operon confers competitive advantage over other S . meliloti strains in iron limited environments [32]; we could therefore speculate that a significant fraction of the S . meliloti strains have a competitive disadvantage in environments with limitation in iron bioavailability . Surprisingly , an ortholog of FixJ ( the component of the global two-component system FixJL , which turns on nitrogen-fixation genes in microaerobiosis during symbiosis ) was not predicted in two S . meliloti strains ( A0643DD and C0438LL ) ; the absence of the gene was further confirmed by PCR . Even though such an important regulator has been found to be absent in these two strains , another gene with similar domains ( orthologous group SinMel7252 , containing gene SMa1686 from the reference strain Rm1021 ) was found to belong to the core genome . SMa1686 was shown to be regulated by RirA [46] , but to the best of our knowledge no indications of its relationships with microaerophilic growth conditions and symbiosis are present . Consequently , we cannot a priori exclude that the regulatory functions of FixJ may be carried on by homologs ( as for instance orthologs of SMa1686 ) in strains A0643DD and C0438LL . Indeed , previous works have indicated that several target genes of FixJ lack a direct symbiotic function , suggesting the presence of functional redundancy in the genome [47] . Sixteen TFs were absent in at least one of the outgroups . Of these , 6 are encoded by pSymA , the symbiotic megaplasmid , including two copies of NodD , FixJ , RctR , SyrM and RhrA ( S1 Fig ) . Such difference between intraspecific and interspecific TF gene content may anticipate a similar difference at the downstream regulatory network , for the absence of cross-regulatory links . To minimize the number of false positives in our predictions , we selected PSSMs with relatively high information content ( over the reference strain minimum information content , see Materials and Methods ) A wide range of information gain for PSSMs was observed; of the starting 83 TFBSs retrieved from literature and databases , 41 have been found to have enough information content to reliably predict their TFBSs ( Fig 1a , S1 Table ) . For FixJ , two separate motifs acting together have been described [48] , one above and one slightly below the threshold: both motifs have been used . We have applied a novel TFBS prediction approach to overcome common problems associated with the prediction algorithms and to maximize accuracy and sensitivity [3] , including operon predictions to recover most of the downstream regulated genes ( see Materials and methods ) . The predictions accuracy was determined with a comparison with the downstream regulons reported in the literature , when available ( Fig 1b and 1c ) ; the average accuracy of the predictions was found to be around 55% , with a tendency to positively correlate with the motif information gain ( S2 Fig ) . This behaviour may be explained by the fact that most regulons have been defined on the basis of gene expression data and therefore contain both direct and indirect targets of the TF; our strategy is then not able to recover the indirect targets which might explain the relatively low accuracy . An example of a known regulatory interaction predicted by our approach is rem ( SMc03046 ) , a putative transcriptional regulator involved in the control of motility in S . meliloti Rm1021 [49] , which was predicted to be under the control of MucR in our analysis ( S1 Material ) . To provide additional validation to our predictions , we used a compendium of S . meliloti gene expression data from the Colombos database [50] ( see Materials and Methods ) . The full compendium contained 424 conditions and was used to calculate average correlation coefficients among the genes of i ) the same predicted regulons , ii ) the regulons reported in the literature and iii ) random groups of genes sampled from the genome ( Fig 1d and S2 Material ) . We have selected the conditions maximising the average correlation for a group of genes using a genetic algorithm ( see Materials and Methods ) . Correlations for our predictions were not significantly different from the experimentally defined regulons; genes belonging to predicted regulons had a slight tendency to be higher than the random regulons , but if this difference was not significant ( p = 0 . 09 ) . We further experimentally confirmed some of the predictions on a subset of predicted promoters of the NodD regulon ( S2 Table ) . Predicted TFBSs in upstream regions against TFBSs predicted in coding regions were considered as signal to noise ratio ( upstream hits on total hits ) to measure the predictions quality ( Fig 1e ) ; for more than 70% of the analysed TF the observed ratio was above 50% , with a very poor correlation with the motif information content . Taken together these results show that our predictions are of fairly good quality . Little variability in the number of genes under the control of each TF was observed among different strains ( Fig 2 and Table 1 ) . Each TF was predicted to control the transcription of 12 genes on average , with RirA showing the largest regulon ( with an average of 71 . 6 genes ) and SyrM the smallest one ( with an average of 1 . 1 genes ) . TFs with lower information content TFBSs showed a tendency to control a larger number of genes ( S2 Fig ) , which confirms the influence of the information content on motif recognition . The predicted regulons were found to have comparable sizes in the outgroups; therefore the regulon is conserved in size between different species; this might be the result of the conservation across the species of the TFBS or of more general energy constraints on transcription/translation . Besides similar regulon sizes , we found that an average 40% of genes belonging to a regulon belong to the accessory genome ( Table 2 ) ; this implies that although variable , each TF recruits a similar number of genes under its control , at least in the species analysed here . Obviously , the variability of the regulons is related with both the variability in upstream regions of core genes and the presence of genes from the accessory genome ( whose presence varies across and between the species ) in the regulons . Predictions for TFs with low information content TFBSs showed a very poor accuracy and precision when compared to experimental data found in the literature; an efficient search strategy for such TFBSs using PSSM has still to be developed . However , from an evolutionary point of view , since those TFs are predicted to bind rather aspecifically to many sites along the genome , this would result in even a larger divergence of regulons between strains , as recently reported in comparison among species [51] . To clarify if the patterns of variability of the regulatory network are related to the phylogenetic distance among strains a comparison between divergence of panregulons and divergence of pangenomes was performed . Following the pangenome analysis , we calculate three sets of distance matrices among the genomes under analysis ( see Materials and Methods ) : the first was obtained from the alignment of core genes ( hereinafter the core distance ) , the second from alignments of the upstream regions of the core genes ( the upstream distance ) , and the third is instead based on the presence/absence profiles of accessory genes ( gene content distance ) . The three distances were then compared with the regulatory network distance of the corresponding strains/species , which was calculated with the same metric defined by Babu and collaborators [16] . Intuitively , the divergence in upstream regions should be paralleled by divergence in the regulatory network , since the former will at some point determine a loss/gain of TFBSs affecting the structure of the regulatory network . Similarly , a larger difference in gene content should also be mirrored by a higher variability in the regulatory network , since new genes may be recruited in the regulatory network and/or TFs may be lost/gained . On the other hand , we don’t expect to observe a strong correlation between core and regulatory network distances; this is also due to the lower divergence at the coding level between strains , implying that regulon diversity inside a species could be driven by gene content variability and upstream sequences variability . These hypotheses on patterns of correlations between pangenome differences and regulatory divergence were confirmed at the species level ( Fig 3a and 3b ) . The comparison between S . meliloti strains showed that the regulatory network distance is correlated with both the upstream distance and with gene content distance . The core distance showed no significant correlation with the regulatory network distance ( Fig 3b ) . When considering the outgroup species , all three distances were found to be similarly correlated with the regulatory network distance ( Fig 3c ) . Since the divergence in coding sequences cannot directly influence transcriptional regulation ( with the exception of non-synonymous mutations in the DNA binding domain of a TF ) , we propose that the most likely explanation of the observed correlations is the overall genome divergence between species , which is ultimately reflected by a higher divergence at the regulatory network level . This is also confirmed by the high correlation coefficients among the three distances . We then concluded that the patterns of regulatory network variation are paralleled , at the species level , by changes in promoter sequences and by the variation in the accessory genome composition , at least in S . meliloti . These two fractions of the pangenome could then be used as bona fide predictors of the extent of rewiring in regulatory networks . However , from these data we cannot confirm a direct causative explanation for the observed regulatory network variation , as this analysis has been focused on the whole pangenome . The striking difference between the slow rate of coding sequence evolution versus the much larger difference in the regulatory networks is however worth noting . Regulatory network evolutionary dynamics showed interesting differences within and between species . Each observed regulatory interaction in the two datasets ( S . meliloti and the outgroups ) and its state across all strains was used to build a hidden markov model to infer the preferred state transitions in our predictions ( see Materials and methods ) , that corresponds to the ways the gene regulatory network can grow and shrink . The possible states of a target gene depend on the presence of the TF , the target gene itself and the upstream TFBS . Therefore , each target gene can be found in one of six different states ( Fig 4a ) . The “plugged” state being the only functional one , which corresponds to a target gene with a TFBS in its promoter region when the TF is present in the genome . The other five are non-functional states but may represent transitory states during the evolution of gene regulatory networks . Each of these states lack: i ) the TFBS ( “unplugged” ) , ii ) the TF ( “ready” ) , iii ) both the TF and the TFBS ( “not ready” ) , iv ) the regulated gene ( “absent” ) or v ) both the TF and the gene itself ( “missing” ) . This HMM can be used to estimate the probability for state transitions , that is the probability of observing a change from one state to another between two strains . This results in a model that is able to provide a general description of the evolution of regulatory networks within and between bacterial species . Since the models is based on observed states in the available strains , we consider it as a “snapshot” of the regulatory network evolution , and not an equilibrium model . According to the model , the most represented state in the S . meliloti regulatory network is the “plugged” one , indicating conservation of regulatory interactions at the species level ( Fig 4b and S3 Table ) . More interestingly , the model predicts that the “unplugged” genes are mostly seen recruited by the regulatory network and that the regulatory link is then maintained with high probability . Very little probability was given to the “plugged” to “missing” and “plugged” to “absent” transitions , indicating that genes belonging to the gene regulatory network are rarely removed from the genome . On the other hand , genes with no TFBS and its cognate TF are more frequently found to undergo loss ( “not ready” to “missing” ) , suggesting that regulatory interactions are important for gene conservation at the species level . When considering a wider phylogenetic level ( the outgroups ) , the broader variability in TF gene targets resulted in the “plugged” and “missing” state as equally probable , indicating that regulons might evolve by adding and removing new elements to a conserved kernel of gene targets ( Fig 4c and S3 Table ) . This is also reflected in a smaller probability that a target gene i ) remains in the “plugged” state when compared to the S . meliloti species level , and ii ) that it acquires a TFBS . On the other hand , the same probability as within the S . meliloti species was observed for the transition “not ready” to “missing” , which seems to confirm the importance of regulatory features in explaining the accessory genome fraction evolution . Consequently , a different evolutionary dynamics of regulatory circuitry changes seems to be present in relation to the taxonomic ranks; at the species level , robust networks are formed and they tend to include new genes from the species pangenome , which then may be conserved . On the contrary , when comparing wider taxonomic ranges , regulatory networks are less conserved and genes are apparently included in each species’ genome directly with their regulatory features ( in a sort of plug-and-play model ) . Transcription factors with replicon preference were found to have functional signatures in accordance with the functions encoded in the three main replicons of S . meliloti . This aspect has been evaluated by mapping each draft genome on the S . meliloti replicons ( see Materials and methods ) and considering the presence of each gene in the replicons for each of the 51 strains analysed here . Using a clustering approach on normalized gene hits on each replicon we have found that 19 TFs preferentially regulate genes belonging to one of the three replicons: five to the chromosome ( NtrR , OxyR , NesR , ChvI and SMc03165 ) , six to the pSymB chromid ( SM-b21706 , SM-b20667 , ChpR , RbtR , SM-b21598 and SM-b21372 ) and eight to the symbiotic megaplasmid pSymA ( SyrM , NodD3 , RhrA , NodD1 , NodD2 , FixJ , FixK1 and NifA ) ( Fig 5a ) ; these TFs are also encoded by the same replicon . The six TFs encoded by the pSymB chromid ( whose regulon is also preferentially located on pSymB ) appear to mostly regulate the transport and metabolism of various carbon and nitrogen sources , including ribitol ( RbtR ) , tagatose , sorbitol and mannitol ( SM-b21372 ) , ribose ( SM-b21598 ) , lactose ( SM-b21706 ) and tartrate , succinate , butyrate and pyruvate ( SM-b20667 ) . The eight TFs present in the symbiotic megaplasmid pSymA ( with regulons preferentially located on pSymA ) were found to be involved in the regulation of key symbiotic processes , including nitrogenase synthesis and functioning through micro-aerophilia ( FixJ , FixK1 and NifA ) , nod-factors biosynthesis ( SyrM , NodD1 , NodD2 and NodD3 ) , and iron scavenging ( RhrA ) . A functional enrichment analysis using COG annotations ( S3 Fig ) on genes belonging to the regulons of the replicon-biased TFs confirmed this general observation: no functional category was enriched in the chromosome . The G category ( carbohydrate metabolism and transport ) was enriched in genes regulated by pSymB encoded TFs , in agreement with the role of chromid pSymB in providing metabolic versatility to S . meliloti . The C ( energy production and conversion ) , U ( intracellular trafficing and secretion ) and T ( Signal Transduction ) categories were enriched in genes under the control of pSymA-harboured TFs , which show some relationship with the establishment on the plant symbiosis . This analysis allowed us to depict a scenario where a significant part of the regulatory network is replicon-specific , with a tendency to maintain the functional signature of the host replicon , thus confirming earlier reports on the evolutionary independence of chromids and megaplasmids in S . meliloti [29 , 31 , 32] . Interestingly , a fraction of TFs have target genes which span over different replicons , and show a preference for cross regulation between the chromosome and the chromid ( Fig 5b ) . The presence of cross-replicon regulons , may indeed allow a stabilization of genomic structure , genetically and metabolically connecting chromosome encoded functions with those present in the other two S . meliloti replicons . In the evolutionary model of the chromid [29 , 31 , 32] , its stabilization within the host genome is related to the acquisition of essential ( core ) genes in a previously introgressed megaplasmid which gained niche-specific genes . Here , we found that for TFs encoded on the chromosome ( as AglR , GlnBK , IolR , BetI , LsrAB , MucR , PckR , RirA , NesR ) a variable number of target genes are present on pSymB ( S1 Material ) . The preference for cross-regulation between the chromosome and the chromid , as opposed to the megaplasmid uncovers an additional mechanism by which a chromid integrates itself in bacterial pangenomes .
Regulatory networks are key components of cell’s response to environmental and physiological changes . In the past years , several works have highlighted a high transcriptomic variability in strains or individuals from the same species [52 , 53] , in addition to genomic variation . Consequently , regulatory network variation might have profound impact on local adaptation and fitness of organisms . Recent studies have confirmed that bacterial regulatory networks are able to tolerate the addition of new genes [24] , which in turn can serve as raw material for selection to operate . Using our original combined search strategy , we indeed found variability in regulon composition within the S . meliloti species , which in fact accounted on average on 40% of the regulon of each strain . On the other hand the regulon size was found to be conserved even outside the species boundary . This could suggest that even though the genes under the control of a TF vary between strains , there is a general constraint on the size of the transcriptional response . Whether this is due to energy constraints or being simply an effect due to the genome base composition is yet to be clarified . We found that the regulatory network distance ( as defined in [16] ) correlates with the upstream distance and also with the gene content distance . This correlations may suggest that regulatory network composition is influenced by both promoter variability and accessory genome variability . Indeed , we may speculate that the sequence divergence in upstream regions can result in the appearance or disappearance of TFBSs , thus changing the regulatory network content . Moreover , gene content dynamics may also have a strong impact on the regulatory network , with the introduction of new gene cassettes containing TFBS recognized by resident TFs . We can consequently hypothesize that the evolution of bacterial regulatory networks , as that of the pangenome , may be influenced by mechanisms of gene acquisitions , such as lateral gene transfer , and it’s not only linked to mutations in upstream regions . The observed changes in the regulatory network also show interesting features with respect to pangenome composition . Indeed , even if a significant difference in the state transitions of regulatory links inside and outside the species boundary has been shown , for genes that lack both a TFBS and their cognate TF , we have observed a similar tendency to disappear from the pangenome . This observation may suggest that the dynamics governing pangenome evolution within a species could depend in part on a ‘gene fitness’ related to being wired into the regulatory network . We can then propose that regulatory networks have an important role in shaping the bacterial gene content and can contribute to gene fitness , which in turn may be linked to environmental adaptation . Moreover , the preference of nineteen TFs for target genes on one of the three replicons of S . meliloti indicates that in multipartite bacterial genomes , similarly to replicon-dependent patterns of evolution in gene and functions content [31] , a replicon-specific transcriptional regulation is to be expected . At the same time , a significant number of cross-links between the chromosome and the chromid suggest for the first time an additional mechanism by which new replicons can be integrated into a bacterial pangenome .
The 51 genomic sequences belonging to Sinorhizobium meliloti and the five genomic sequences from closely related symbiotic species are listed in S4 Table . The orthology relationships inside the 51 S . meliloti strains has been computed using the Blast-BBH algorithm implemented in the DuctApe suite ( version 0 . 13 . 0 ) [54] , using default parameters . The same analysis has been conducted on the five closely related species with the addition of the Rm1021 reference strain , using the BLOSUM62 scoring matrix to account for their greater sequence diversity . The number of regulators present in each genome has been estimated using COG annotations . The similarity of each protein against the COG database has been measured with a rpsblast scan [55] , using an E-value threshold of 1e-10 . Each protein mapped to the COG category K ( Transcription ) has been considered as a putative regulator . To confirm the absence of the fixJ gene in strains A0643DD and C0438LL , PCR primers amplifying a large portion ( from nucleotide position 32 nt to 595 out of 615 nt total ) of the coding sequence of fixJ gene have been designed on the basis based on the ortholog sequence in strain BL225C ( SinmeB_6173 ) with Primer3Plus ( fw: 5′-ACGAAGAGCCGGTCAGGAAGTCGCTGGCATTCATGCTG-3′; rv 5-CGGCGAGAGCCATGCGAACGAGATGGGGGAGGCTC-3 ) [56] . PCR has been performed with the Maxima Hot Start Green Master Mix ( Thermo Fisher ) in 20 microL total volume by using 10 ng of DNA , purified from liquid culture with FAST DNA Kit ( QBiogene ) and 10 pmols of each primer . Cycling conditions were as follows: 5′ 94°C , followed by 30” 94°C , 30” 55°C , 1′ 72°C repeated for 35 cycles . PCR products were resolved after agarose gel electrophoresis ( 1 . 5 w/v ) in TAE buffer with ethidium bromide ( 10 microg/ml ) as staining agent . The 83 regulators whose PSSM has been extracted from the various sources are listed in S1 Table . For those PSSMs retrieved from the literature , we collected the upstream regions of the regulated genes and ( when available ) , the consensus binding sites from bibliographical records; the upstream regions have then been analysed with the meme program [7] ( version 4 . 9 . 0 ) , using the model that retrieved the PSMM with higher similarity to literature . Twenty-two motif files have been generated using the information retrieved from the RhizoRegNet database [27] . Fifteen motif files have been generated using the information retrieved from the RegTransBase database [57] . For the 5 regulators having more than one predicted motif , for instance those having a variable length ( FixJ , RpoD , RpoE2 , RpoH1 and RpoH2 ) , one motif file for each motif length has been generated . All the retrieved PSSMs have been converted to HMM models using the hmmbuild program from the HMMer suite [12–14] ( version 3 . 1b1 ) , using the alignments present in the MEME motif file . It has been previously shown that in bacterial genomes TFBS can be reliably distinguished from background DNA only if their information content is higher than the minimum information content for the target genome , which depends on the genome size and composition [5] ( this simplification of course ignores other factors such accessibility or proximity of the RNA polymerase ) . The information gain of the TFBS with respect to the genome is calculated using the Kullback-Leibler divergence between the corresponding nucleotide frequencies [58] , and it has been shown to correlate with the motif length and base composition of the motif with respect to the surrounding genome sequence . TF motifs with sufficient information content also tend to show less variability in their regulon composition between species [51]; by focusing our analysis on such TFs we ensured a more precise analysis . The information content of each motif has been calculated as suggested by Wunderlich et al [5] , using the Rm1021 reference genome for the calculation of the minimum information content; given the dependence of this variable on genome size and the fact that all the S . meliloti strains have similar genome size , there has been no need to calculate a strain specific threshold . PSSMs whose information content was found to be lower the minimum information content have been discarded with exception of FixJ , which has two distinct PSSM , one of which is above the threshold . In the presence of more than one source for a regulator ( literature , RhizoRegNet or RegTransBase ) , the PSSM having the highest information content has been considered in the final analysis . For each genome , background k-mers frequencies have been calculated using the fasta-get-markov program from the MEME suite ( version 4 . 9 . 0 ) [7] , using 3 as the maximum value for k . Each regulatory motif has been searched inside each genomic sequence using four scanning algorithms . The mast program from the MEME suite ( version 4 . 9 . 0 ) [7] has been used with an E-value threshold of 100 and the use of a genome-specific background file . The matrix-scan program from the RSAT suite [8–10] has been used with a P-value threshold of 0 . 001 , the background file and a pseudocount of 0 . 01 , as suggested by Nishida et al . [59] . The Bio . motifs package from the Biopython library ( version 1 . 62b ) [11] has been used with a false negative rate threshold of 0 . 05 and a pseudocount of 0 . 01 , as suggested by Nishida et al . [59] . The nHMMer program from the HMMer suite ( version 3 . 1b1 ) [12–14] has been used with an E-value threshold of 100 and with all the heuristic filters turned off . Each regulatory motif hit has been parsed , separating the hits being present in the upstream region of a gene from the others . The upstream region has been defined as the intergenic region ( not overlapping any coding sequence ) in front of the first codon with a maximum size of 600 bp . In the case of a palindrome motif , the motif orientation has been ignored . The distributions of the raw scores has been tested using a normality test , as implemented in the SciPy library ( version 0 . 13 . 3 ) [60][61] . The score threshold has been determined through the calculation of the raw scores quartiles ( Q1 and Q3 ) and defining the score threshold ( τS in Eq 1 ) in order to consider only the upper outliers [62] . τ S = Q 3 + ( 1 . 5 ( Q 3 - Q 1 ) ) . ( 1 ) For the Biopython method the bit score has been used , while for the RSAT , HMMer and MEME methods the negative base 10 logarithm of the E-value has been considered . The regulatory motifs predicted by at least three methods have been considered for further analysis . The compendium of gene expression data for S . meliloti str . Rm1021 from the Colombos database [50] was used to calculate correlation coefficients among genes in the regulons reported in the literature , our predictions and random sets of genes . Random regulons were produced by random sampling groups of genes of size 5 , 10 and 15 , for which 500 sets were produced . Correlation was quantified by the squared uncentered correlation coefficient , which was calculated using Matlab , as the square of 1 − cos distance . Values plotted in Fig 1d are averages over the entire set of genes under analysis . We have implemented a strategy allowing to select the conditions maximizing the average squared correlation within a group of genes , since many of the conditions of the compendium are likely not related to our predictions . Selection of the conditions was performed using the genetic algorithm implemented in the GA Matlab function , with default tolerances ( TolCon = 10−6 , TolFun = 10−6 ) . We let the algorithm select the conditions minimizing 1 R 2 where R is the uncentered correlation averaged over all pairwise comparisons made within the group of genes under analysis . Since we noticed that correlations are strongly and inversely correlated with the number N of included conditions , especially when N ≤ 20 , we discarded all cases where the number of conditions was less than 20 ( final N = 950 ) . All conditions containing missing data in at least one of the genes under analysis were discarded before starting the procedure . For some of the known and predicted regulons , correlations were not calculated as the available number of conditions after removing missing data was less than 30 before the optimization . Upstream sequences from selected putative target genes of NodD regulon were analysed ( see S2 Table ) . Sequences ( approximately 400 nt upstream the translation start site of the gene ) were amplified from crude lisates of S . meliloti strains with AccuPrime Pfx DNA Polymerase ( Thermo Fisher ) and cloned into pTO2 vector ( which carries GFPuv as reporter gene [63] ) by using SalI and KnpI restriction sites . Recombinant clones of E . coli S17-1 strain were selected by gentamycin resistance and verified by sequencing of inserted fragments . Positive clones were used for transferring recombinant pOT2 vectors to S . meliloti Rm1021 by bi-parental conjugation by using previously described protocols [64][65] . S . meliloti Rm1021 recombinant strains were then tested for GFP fluorescence after incubation of a 5 ml culture grown at the mid-exponential phase with 1 microM luteolin ( Sigma-Aldrich ) in liquid TY medium at 30°C for 3h . GFP fluorecence was measured on a Infine200 Pro plate reader ( Tecan ) . Measures were taken in triplicate and normalized to cell growth estimates as absorbance to 600nm . The operons belonging to the 56 genomes of this study have been predicted using the Operon Prediction Software ( OFS , version 1 . 2 ) [66] , using a beta threshold of 0 . 7 and a probability threshold of 0 . 5 . The number and length of the predicted operons in each strain are listed in S5 Table . Each contig of the 44 S . meliloti draft genomes has been mapped to the seven complete genomes using CONTIGuator ( version 2 . 7 . 3 ) [67] , using a 15% coverage threshold and considering blast hits over 1000 bp in length . A contig has been considered mapped to a replicon when it has been found mapped to the replicon in at least five complete genomes , or when it has been mapped to the replicon in at least one complete genome and to no replicon in the others . Knowing that very few portions of the S . meliloti genome are shuffled between replicons [31] , we assessed the quality of this mapping procedure by checking whether the S . meliloti orthologs were found to be mapped to more than one replicon; for each orthologous group the genes not mapped to any replicon have been removed , and the relative abundance of the most representative mapped replicon has been computed . A relative abundance of 1 means that the orthologs have all been mapped to the same replicon in all the strains . The vast majority of the orthologous groups was found to map to a single replicon ( S4 Fig ) . The number of average gene hits has been divided for each replicon ( either from a complete genome or a draft genome ) and normalized by the number of genes belonging to each replicon in the Rm1021 reference strain . Regulators with preferential regulatory hits in a specific replicon have been highlighted performing a k-means clustering ( k = 5 , selected using an elbow test [68] ) and plotted using the two principal components of the proportion of hits in each replicon , using the scikits-learn package ( version 0 . 14 . 1 ) [69] . Only the three main replicons ( chromosome , pSymB and pSymA ) have been considered . COG categories enrichments have been tested using a Fisher’s exact test , as implemented in the DendroPy package [70] . Phylogenetic distance inside the S . meliloti pangenome and the pangenome of the five related species has been computed as described in a previous work [31] . The pangenome has been divided in three fractions , allowing the use of three distinct phylogenetic distances . The “core” distance has been calculated through the alignment of all the nucleotide sequences of each core gene , discarding those genes where at least one sequence was 60bp shorter or longer with respect to the other sequences . The “upstream” distance has been calculated through the alignment of the core genes upstream regions , discarding sequences below 5bp in length . The alignments have been calculated using MUSCLE ( version 3 . 8 . 31 ) [71] and the bayesian tree has been inferred using MrBayes ( version 3 . 2 . 0 ) [72] . The distance matrix for both distance categories has been computed from the phylogenetic tree using the textitBio . Phylo package inside the Biopython library ( version 1 . 62b ) [73] . The “accessory” distance has been calculated through the construction of a presence/absence binary matrix for all the accessory genome OGs; the distance between each strain has been then calculated using the Jaccard distance measure , as implemented in the SciPy library ( version 0 . 13 . 3 ) [61] . The distance between each strain inside the S . meliloti and the other five related species regulatory network has been computed using the distance in the presence/absence of regulatory interactions as suggested in the work of Babu and collaborators [16] . The distance between strain A and B is computed using Eq 2 . D A B = 1 - c o r e A B t o t a l A B , ( 2 ) where coreAB and totalAB represent the number of conserved and total regulatory interactions , respectively . Pearson and Spearman correlation coefficients between the pangenome and the regulatory network distance have been calculated using the implementations of the SciPy library ( version 0 . 13 . 3 ) [61] , removing the outliers using a Z-score threshold of 3 . 5 on the mean absolute deviation of the distances . The state transitions of the regulatory network has been inferred by encoding them in a hidden markov model . Each one of the regulatory links observed in at least one strain has been tested for their state in each organism , following the labelling of Fig 4a . Specifically , each regulatory link in the network of each organism could belong to one of the following categories: Plugged: regulator , gene and TFBS present Unplugged: regulator and gene present , TFBS absent Ready: gene and TFBS present , regulator absent Not ready: gene present , regulator and TFBS absent Absent: regulator present , gene and TFBS absent Missing: regulator , gene and TFBS absent The hidden markov model has been constructed using the Baum-Welch algorithm [74] , as implemented in the GHMM python library . For each observed regulatory link in the regulatory network , the observed transition between each permutation of pairs of strains has been used to train the HMM and then compute the states and transitions probabilities . The transition probability has been defined for each state as the probability of observing the transition between two strains . Since each state has different transition probabilities and their sum is one for each state , we do not observe symmetrical probabilities . Regulatory motifs data has been analysed and visualized using the NumPy [75] and matplotlib [76] libraries inside the iPython environment [77] . Regulatory networks have been built using the networkx library [78] and visualized using Gephi [79] . Genomic sequences , regulatory motif files and search and analysis scripts are available as separate git repositories . The rhizoreg repository ( https://github . com/combogenomics/rhizoreg/ ) , contains the input data; the regtools repository ( https://github . com/combogenomics/regtools/ ) contains the main scripts used to conduct the analysis . | The influence of transcriptional regulatory networks on the evolution of bacterial pangenomes has not yet been elucidated , even though the role of transcriptional regulation is widely recognized . Using the model symbiont Sinorhizobium meliloti we have predicted the regulatory targets of 41 transcription factors in 51 strains and 5 other rhizobial species , showing a correlation between regulon diversity and pangenome evolution , through upstream sequence diversity and accessory genome composition . We have also shown that genes not wired to the regulatory network are more likely to belong to the accessory genome , thus suggesting that inclusion in the regulatory circuits may be an indicator of gene conservation . We have also highlighted a series of transcription factors that preferentially regulate genes belonging to one of the three replicons of this species , indicating the presence of replicon-specific regulatory modules , with peculiar functional signatures . At the same time the chromid shares a significant part of the regulatory network with the chromosome , indicating an additional way by which this replicon integrates itself in the pangenome . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Evolution of Intra-specific Regulatory Networks in a Multipartite Bacterial Genome |
Iron is a key pathogenic determinant of many infectious diseases . Hepcidin , the hormone responsible for governing systemic iron homeostasis , is widely hypothesized to represent a key component of nutritional immunity through regulating the accessibility of iron to invading microorganisms during infection . However , the deployment of hepcidin in human bacterial infections remains poorly characterized . Typhoid fever is a globally significant , human-restricted bacterial infection , but understanding of its pathogenesis , especially during the critical early phases , likewise is poorly understood . Here , we investigate alterations in hepcidin and iron/inflammatory indices following experimental human typhoid challenge . Fifty study participants were challenged with Salmonella enterica serovar Typhi and monitored for evidence of typhoid fever . Serum hepcidin , ferritin , serum iron parameters , C-reactive protein ( CRP ) , and plasma IL-6 and TNF-alpha concentrations were measured during the 14 days following challenge . We found that hepcidin concentrations were markedly higher during acute typhoid infection than at baseline . Hepcidin elevations mirrored the kinetics of fever , and were accompanied by profound hypoferremia , increased CRP and ferritin , despite only modest elevations in IL-6 and TNF-alpha in some individuals . During inflammation , the extent of hepcidin upregulation associated with the degree of hypoferremia . We demonstrate that strong hepcidin upregulation and hypoferremia , coincident with fever and systemic inflammation , are hallmarks of the early innate response to acute typhoid infection . We hypothesize that hepcidin-mediated iron redistribution into macrophages may contribute to S . Typhi pathogenesis by increasing iron availability for macrophage-tropic bacteria , and that targeting macrophage iron retention may represent a strategy for limiting infections with macrophage-tropic pathogens such as S . Typhi .
Typhoid fever is a common infection that follows oral ingestion and invasion of the Gram-negative bacterium Salmonella enterica serovar Typhi ( S . Typhi ) . An estimated 26 . 9 million cases occurred globally in 2010 , disproportionately affecting children in resource-limited areas of sub-Saharan Africa and southeastern Asia [1 , 2] . S . Typhi is a human-restricted pathogen . Unlike non-typhoidal Salmonella infection , which is characterized by rapid-onset gastrointestinal inflammation and diarrheal illness in immunocompetent adults , S . Typhi causes a systemic infection . After ingestion , bacteria disseminate through the reticuloendothelial system , where they are thought to incubate for 7–14 days . Clinical illness then develops , characterized by fever and non-specific symptoms including headache , nausea and abdominal pain , and accompanied by bacteremia [3] . However , detailed understanding of typhoid pathogenesis remains limited , in part since convincing small-animal infection models are lacking [4] . An experimental human S . Typhi challenge model was recently reestablished , presenting a unique opportunity to investigate typhoid pathogenesis in a controlled setting in the natural host [5–7] . Conflict exists between hosts and invading pathogens over the control of the critical micronutrient , iron ( reviewed in [8 , 9] ) . To limit free iron availability , mammalian hosts sequester iron using high-affinity iron-binding proteins including transferrin , lactoferrin , haptoglobin , hemopexin and the iron storage protein ferritin . To counteract this , many bacteria express higher affinity siderophores ( e . g . enterobactin ) that appropriate iron from host iron binding proteins; host-produced siderophore-binding proteins such as lipocalin-2 in turn counter these . A further host response to infection involves the rapid induction of hypoferremia , where iron becomes sequestered in reticuloendothelial macrophages and therefore excluded from serum [8 , 10] . This state may be disadvantageous to extracellular pathogens [11] but potentially could be exploited by intracellular , macrophage-tropic bacteria including S . Typhi [12] . Together , the host mechanisms aimed at sequestering iron from invading microorganisms are considered to contribute to innate protection against infection , often termed “nutritional immunity” [8] . In recent years , many genes involved in mammalian iron homeostasis have been discovered [13] , meaning that the molecular basis of iron perturbations during infections can be investigated in a new light . Amongst these , hepcidin stands out as the central regulator of systemic iron balance [14] . Hepcidin dictates dietary iron uptake and recycling of red cell iron by binding and causing degradation of the sole known iron exporter ferroportin , which is expressed on duodenal enterocytes and iron-recycling macrophages [15] . Consequently , high hepcidin levels effect iron exclusion from serum , through blocking dietary iron uptake and preventing macrophage iron release . Hepcidin is induced homeostatically in response to increased plasma and liver iron [16 , 17] , but is also an acute phase protein upregulated by inflammatory cytokines , notably IL-6 [18–20] . Thus , elevated hepcidin concentrations during inflammation and infections contribute to hypoferremia [11 , 21] and , if chronic , to iron-restricted erythropoiesis and anemia [22] . Hepcidin regulation is less well studied in the context of human infection . Analyses to date indicate that hepcidin behavior differs between infections . For example , it is upregulated during uncomplicated malaria [23–26] , and during acute , chronic and advanced HIV-1 infection [27 , 28]; however , it is suppressed during Hepatitis C Virus infection [29] , and in severe malarial anemia , where bone-marrow derived signals indicating erythropoietic iron demand likely dominate , suppressing hepcidin production [26 , 30] . Importantly , hepcidin remains remarkably poorly studied in human bacterial infections . This is despite iron representing a battleground of host-bacterial conflict important enough to have shaped both primate and bacterial genomes alike [31] . Here , we investigate the dynamics of hepcidin in relation to iron and inflammatory indices during acute experimental Salmonella Typhi infection in humans .
Human typhoid challenge was performed with healthy consenting adult volunteers ( 18–60 years ) who had not previously received typhoid vaccination or resided in typhoid-endemic areas for >6 months [7] . Data from two sets of study participants are described: first , from the placebo arm of a vaccine/typhoid challenge study , where participants received an oral placebo vaccine ( sodium bicarbonate solution and excipients ) 28 days before oral challenge with S . Typhi ( n = 30 , Study A , Table 1 ( baseline data from day of typhoid challenge shown ) ) , and secondly , for more detailed longitudinal analysis , from a previously described cohort challenged orally with S . Typhi in a preliminary dose-escalation challenge model ( n = 20 , Study B , Table 1 ) [7] . Full details of the challenge model used in both studies are described in Waddington et al [7] . Briefly , participants ingested a single freshly prepared dose of S . Typhi ( Quailes strain , 104 CFU ) suspended in sodium bicarbonate solution . After challenge , study participants were reviewed daily for 14 days; blood samples were collected on alternate days , at typhoid diagnosis and intervals thereafter . In typhoid-infected and non-infected participants , the mean blood volumes collected during the 28 days following challenge were approximately 920mL and 600mL respectively . “Typhoid diagnosis” was defined a priori by clinical and/or microbiological endpoints: temperature ≥38°C sustained for ≥12 hours and/or blood culture evidence of S . Typhi bacteremia , respectively . Antibiotic treatment ( ciprofloxacin , 500 mg twice daily , 14 days ) was initiated upon attainment of either diagnostic criterion and in all remaining participants at Day 14 . Actual challenge doses were determined , and quantitative blood culture was performed at typhoid diagnosis , as previously described [7] . Serum samples were filtered prior to analyses using Costar Spin-X low protein binding 0 . 22μm cellulose acetate membrane filters ( Corning ) . Spin-filtering had no effect on hepcidin measurement ( n = 4 samples , p = 0 . 36 ( paired t-test ) ) . Hepcidin was quantified by ELISA using the hepcidin-25 EIA kit ( Bachem ) , with the manufacturer’s protocol modified to incorporate a 9-point , 2-fold serial dilution standard curve . Samples were diluted to 10% or 5% prior to analysis . The lower limit of detection ( LLOD ) was 0 . 08 ng/mL , calculated as described previously [27] . Samples returning a reading below LLOD were assigned the value ( LLOD*dilution factor ) /2 . Serum ferritin ( Architect Ferritin Assay ) was quantified using the Abbott Architect 2000R automated analyzer ( Abbott Laboratories ) ; C-reactive protein ( CRP ) ( MULTIGENT CRP Vario Kit , with high sensitivity calibrators ) , serum iron and Unsaturated Iron Binding Capacity ( UIBC , MULTIGENT Iron Kit ) were quantified using the Abbott Architect c16000 automated analyzer ( Abbott Laboratories ) . Transferrin saturation ( Tsat ) was calculated using the formula: Tsat = ( ( Serum Iron ) / ( Serum Iron + Unsaturated Iron Binding Capacity ) ) *100 . CRP concentrations above and below the assay limits of detection ( 160 and 0 . 1 mg/L ) were assigned the values 160 mg/L and 0 . 05 mg/L respectively . Plasma cytokine concentrations were measured in duplicate using a custom TNFα / IL-6 Luminex panel ( MILLIPLEX MAP kit , Millipore ) according to the manufacturer’s instructions . Readings with % Coefficient of Variance >30% were excluded; those falling below the LLOD were allocated the value 1 . 6 pg/mL ( LLOD/2 ) . Hemoglobin , red blood cell counts , and mean corpuscular volume ( MCV ) were quantified by routine hematologic analysis . Statistical analyses were performed using Prism ( version 6 , GraphPad Software Inc . ) , SPSS ( version 16 . 0 , IBM SPSS ) , STATA/SE13 . 1 ( Statacorp ) and R statistical language [32] . All raw data can be found in the file S1 Dataset . For indices that were not normally distributed ( hepcidin , ferritin , and CRP in all cases; additionally serum iron and transferrin saturation when considering data other than baseline data ) , geometric means were compared , or data were log-transformed prior to analysis . Differences between indices pre-challenge and at typhoid diagnosis ( Study A ) were evaluated using paired t-tests . In correlation analysis , Pearson correlation coefficients were computed; in cases where study participants contributed more than a single observation , correlation analyses were adjusted accordingly by using regression with clustered errors ( STATA/SE13 . 1 ) , which adjusts the confidence intervals of the regression coefficients to account for intra-cluster correlation , as is likely when multiple observations from the same individuals are included . Statistical tests returning p<0 . 05 were considered significant . Time-course analyses were performed using the packages fields [33] , nlme [34] and lme4 [35] within R statistical language [32] as follows . ( i ) Normalization of time series: Since the time between typhoid challenge and diagnosis ( TD ) varied between participants , the time variable “day relative to TD” was used , with TD = 0 being day of diagnosis . ( ii ) Smoothing spline regression: Mean analyte measurements across all subjects for each day relative to TD were calculated; samples from day of typhoid challenge ( ‘baseline’ ) and the final visit ( Day 14 post-challenge ) were grouped separately . A smoothing spline regression was applied with smoothness estimated from the data by generalised cross validation ( GCV ) [36] . 95% pointwise prediction intervals and conservative simultaneous Bonferroni bounds were calculated . ( iii ) Assessment of the effect of time relative to TD on analyte concentrations: linear mixed-effects models were fitted using a described model formulation [37] and computational framework [38] . The categorical variable “day relative to TD” was modeled by fixed effects; variability between individuals was captured using random effects . The null hypothesis that there is no significant difference in analyte levels over time after challenge was tested using the Wald test; specific pairwise comparisons between analyte concentrations at baseline ( day of challenge ) and later time-points were tested by t-tests , accounting for subject-specific variability . Tests returning p<0 . 05 were considered significant . The National Research Ethics Service approved both studies ( Oxfordshire Research Ethics Committee A , 10/H0604/53 and 11/SC/0302 ) . They were performed in accordance with the principles of the ICH-Good Clinical Practice guidelines and amendments . All study participants provided written informed consent in accordance with the Declaration of Helsinki on at least one occasion , as previously described [7] .
Characteristics of study participants from two experimental typhoid challenge studies are given in Table 1 . The typhoid attack rates ( percent typhoid-diagnosed participants by Day 14 ) in Study A and B were 67% ( 20/30 ) and 65% ( 13/20 ) , while mean duration between challenge and typhoid diagnosis was 7 . 4 ( 95% Confidence Interval ( CI ) : 6 . 6–8 . 3 days ) and 7 . 7 days ( 6 . 7–8 . 7 days ) , respectively . There were no significant differences in the baseline characteristics of participants recruited to Study A or B , except that the challenge dose and transferrin saturations were marginally higher in Study B . Considering all participants from Studies A and B together , significant associations between log10-hepcidin and log10-ferritin levels ( p<0 . 0001 , r2 = 0 . 642; S1 Fig , panel A ) and between hepcidin and both transferrin saturation ( S1 Fig , panel B ) and hemoglobin ( S1 Fig , panel C ) were found in baseline , pre-challenge samples . Male participants had significantly higher baseline hemoglobin , hepcidin and ferritin levels than females ( S1 Fig , panels D-F ) . These observations are typical of healthy adult populations . In univariate analyses , there were no significant differences in age , sex , weight or challenge dose , or in baseline hematological or iron-related parameters between those subsequently diagnosed or not diagnosed with infection , even when participants from the two studies were pooled together to increase power ( S1 Table ) . Amongst individuals diagnosed with typhoid , we found no association between the time to typhoid diagnosis and baseline iron status as indicated by ferritin ( r2 = 0 . 014 , p = 0 . 505 ) or hepcidin ( r2 = 0 . 015 , p = 0 . 497 ) . Amongst individuals from the two studies who were diagnosed with typhoid , increasing challenge dose was significantly negatively associated with time-to-diagnosis ( r2 = 0 . 174 , p = 0 . 016 ) and positively associated with the number of bacteria quantified at diagnosis ( r2 = 0 . 241 , p = 0 . 007 ) . To investigate the extent to which typhoid infection was associated with changes in hepcidin and other iron indices , we analyzed serum samples collected at baseline and on day of typhoid diagnosis in participants challenged in Study A ( n = 19/20 , 7 females and 12 males ) . Amongst these individuals , the mean time to typhoid diagnosis was 7 . 4 days ( 95% CI: 6 . 6–8 . 3 days ) ; mean oral temperature was significantly higher at typhoid diagnosis than at baseline ( diagnosis: 37 . 6°C [95% CI , 37 . 3–37 . 9°C]; baseline: 36 . 3°C [36 . 1–36 . 5°C]; Fig 1A ) . Hepcidin concentrations at typhoid diagnosis were approximately 10-fold higher than at baseline ( Fig 1B ) . This marked hepcidin response was accompanied by hypoferremia demonstrated by a significant decline in mean serum iron and transferrin saturation ( Fig 1C and 1D ) . In contrast , there was a significant increase in the inflammatory markers , CRP and ferritin at diagnosis compared to baseline , although the relative change in ferritin concentration was less notable than that of hepcidin or CRP ( Fig 1E and 1F ) . There were no significant differences in hemoglobin between measurements at baseline and at diagnosis ( Fig 1G ) . Together , these data demonstrate that significant hepcidin upregulation and concurrent hypoferremia are features of the acute phase response to S . Typhi infection . To assess the kinetics of alterations in hepcidin and other indices following S . Typhi challenge , we analyzed serial samples from Study B , firstly from the 7 participants who did not develop clinical disease following challenge , and secondly from the 13 individuals diagnosed with acute infection . For both groups , up to 7 time points from Day 0 ( baseline , challenge day ) onwards were analyzed ( mean , 6 . 15 time points ) . In those who did not develop typhoid infection , significant reductions in hepcidin , ferritin and hemoglobin concentrations , and in red blood cell counts , were observed during the 14-day study period ( S2 Fig , panels A-D ) . There was also suggestion of decline in serum iron and transferrin saturation ( S2 Fig , panels E/F ) . This likely relates to the repeated phlebotomy required by the study protocol , causing reduction of iron indices including hepcidin . CRP concentrations and oral temperatures remained low/normal throughout confirming the absence of a systemic inflammatory response in challenged but non-infected individuals ( S2 Fig , panels G/H ) . Thus , the following time course data from typhoid-infected individuals must be interpreted in the light of these study protocol effects on hematological parameters . In participants who developed typhoid fever , increases in temperature were measured from 48 hours prior to diagnosis ( Fig 2A ) . A concomitant rise in hepcidin concentration was observed , maximal 2 days after diagnosis; temperature and hepcidin levels normalized towards baseline levels over approximately 4 days following treatment initiation ( Fig 2B ) . Similarly , significant declines in serum iron ( commencing prior to diagnosis and reaching a mean nadir of 4 . 9 μmol/L two days post-diagnosis , down from 14 . 5 μmol/L at baseline , Fig 2C ) and transferrin saturation ( mean 6% at nadir two days post-diagnosis , down from 28% at baseline , Fig 2D ) were observed; these indices , like hemoglobin ( Fig 2E ) , were lower at the final time point ( Day 14 ) than at baseline ( Fig 2C and 2D and 2E ) , likely reflecting the effect of venesection described above ( S2 Fig ) . However , the possibility of a hepcidin-mediated block in iron absorption during infection contributing to this observation should not be excluded . A significant induction of the acute phase protein CRP was also observed , escalating marginally later than the initial perturbations to hepcidin and transferrin saturation , but similarly peaking 2 days after typhoid diagnosis ( Fig 2F ) . The iron storage protein ferritin , also an acute phase protein , was induced later than hepcidin or CRP and took longer to resolve towards baseline levels ( Fig 2G ) . Together , these data indicate that the kinetics of hepcidin perturbations and the associated hypoferremia during acute S . Typhi infection mirror typhoid-associated fever and CRP induction . We next investigated relationships between hepcidin concentration and serum iron status in those exhibiting typhoid-related inflammation and those who were not . In this analysis , we included all data from the study from both diagnosed and non-diagnosed individuals , using regression with clustered errors , thereby accounting for the inclusion of multiple observations derived from the same individuals . When there was evidence of acute inflammation ( defined as CRP >5 mg/L ) , significant negative associations between hepcidin and both serum iron and transferrin saturation were observed ( Fig 3A and 3B ) . In contrast , when acute inflammation was absent ( CRP <5 mg/L ) , significant positive associations between hepcidin and serum iron parameters were found ( Fig 3A and 3B ) . Thus , larger hepcidin responses predicted more profound hypoferremia in the context of inflammation , but the opposite in non-inflamed samples , when they presumably reflected iron status . The latter effect was also noted in baseline challenge day samples ( S1 Fig ) . Unlike hepcidin , ferritin did not correlate with the extent of hypoferremia during inflammation , although it did associate positively with serum iron parameters in non-inflamed samples ( Fig 3C and 3D ) . These data indicate non-equivalence of these two indices of iron status , as noted in previous work [39] , and suggest hepcidin may be more closely linked to hypoferremia in the context of the acute inflammation observed during typhoid infection . Hepcidin upregulation during acute phase responses is typically associated with STAT3 activation following signaling by IL-6 and potentially other cytokines ( e . g . IL-22 ) [18–20] . We therefore assessed IL-6 concentrations in plasma samples from the individuals from Study B who developed typhoid infection . We only observed a weak IL-6 response in a subset of individuals ( Fig 4A , see S3 Fig for individual profiles ) ; in the majority of individuals , IL-6 upregulation was not detected . Modest TNF-alpha responses were more consistent , with the highest levels recorded day 2 post-diagnosis in most individuals ( Fig 4B , see S3 Fig for individual profiles ) . These data suggest the cytokine response during typhoid infection may have been blunted , as previously described [40] , and that determinants other than serum IL-6 may be responsible for the hepcidin upregulation observed in this context .
Salmonella Typhi is a significant human pathogen , leading to a major global burden of disease particularly among children and younger adults in endemic settings [1 , 2] . Evolution from a common Salmonella ancestor is thought to have occurred ~50–100 , 000 years ago [41] . However , the basis for the evolution of its ability to evade host defenses and cause systemic infection remains poorly characterized . Understanding how S . Typhi interacts with the human host environment , including the macrophage niche , is crucial in deciphering its pathogenicity and for devising prevention or eradication strategies . The battle for iron is a key determinant of host-bacterial interactions [8 , 9 , 31] . Here , using an experimental human typhoid challenge model , we track for the first time in an invasive human bacterial infection the behavior of the iron regulatory hormone hepcidin and its relationship to perturbations in iron parameters , inflammatory markers , and fever: significant hepcidin upregulation , accompanied by a profound decline in serum iron was observed in participants diagnosed with typhoid infection . Hepcidin has several characteristics reflecting a likely ancestry in immunity to infection . It is a liver-derived acute-phase peptide induced via the inflammatory JAK/STAT3 signaling pathway [18–20 , 42] . It structurally resembles antimicrobial beta-defensins and has modest antimicrobial activity itself [43 , 44] . Hepcidin’s involvement in human infection pathogenesis has been widely proposed , likely relating more to its ability to rapidly alter systemic partitioning of iron than its direct antimicrobial activity [12] . Despite this , its regulation and influence on the pathogenesis of human bacterial infection remains poorly investigated . In humans , significant hepcidin upregulation has been observed during sepsis [45 , 46] , during tuberculosis ( with and without HIV coinfection ) [47 , 48] , and to a less notable extent in children with concurrent Helicobacter pylori infection and iron deficiency anemia [49] . The longitudinal behavior of hepcidin has been assessed during experimental uncomplicated malaria ( where modest increases in hepcidin and IL-6 , associated with changes in systemic iron parameters , were observed ) [23] and during the acute phases of HIV-1 , Hepatitis B Virus and Hepatitis C Virus infections [27] . However , the nature of longitudinal perturbations in hepcidin during the acute phase of a bacterial infection in humans has never been investigated . In study participants diagnosed with acute typhoid infection , we found a marked upregulation of hepcidin around the time of diagnosis , coincident with appearance of fever . Hepcidin concentrations remained high for at least 48-hours during acute infection irrespective of prompt antibiotic therapy , and resolved to normal levels from 4 days after diagnosis . We predict that hepcidin would remain high for considerably longer if the infection were left untreated . Significant elevations in the acute phase proteins , CRP and ferritin , and striking declines in serum iron and transferrin saturation ( from 28% at baseline to 6% at nadir 2 days post-diagnosis ) were also evident . The data suggested hepcidin activity from 1–2 days prior to typhoid diagnosis , and are consistent with the previous description of hypoferremia in the Maryland typhoid challenge in the 1970s [50] . Furthermore , our data indicated that , when acute inflammation was present , the extent of hepcidin upregulation significantly predicted the degree of hypoferremia; in contrast , in normal non-inflamed conditions , hepcidin positively associated with serum iron parameters . Given these data , hepcidin and the associated hypoferremia should be considered for investigation as potential biomarkers of acute infection . Hepcidin upregulation in the context of inflammation/infection is typically linked to signaling via the IL-6/STAT3 pathway [19 , 42] . However , we only detected a modest elevation in plasma IL-6 around typhoid diagnosis , with several participants maintaining IL-6 levels below detectable levels at each sampling time point . When IL-6 was detected , it was at considerably lower levels than in other conditions where hepcidin is notably induced: IL-6 was typically one or more orders of magnitude higher during uncomplicated malaria [24] , sepsis [45] , or experimental endotoxemia [21] . Similarly , although TNF-alpha was induced , the levels detected were relatively low . Since IL-6 and TNF-alpha data were not available from the day of diagnosis or the following day , we cannot exclude the occurrence of a stronger , transient IL-6 induction during these two days . Similarly , we cannot exclude more local but significant cytokine effects in intestine , portal circulation and liver leading to hepcidin upregulation that may not be detected in systemic circulation . Nevertheless , one established feature of S . Typhi infection is a blunted cytokine response [40] . Several factors , most prominently the Vi-capsular polysaccharide , enable S . Typhi to evade innate immune responses ( for example by enabling evasion of detection by complement [51] ) and to establish systemic infections without clinical sepsis [40 , 52] . It is therefore possible signals besides IL-6 are involved in the significant acute phase response and hepcidin induction during acute typhoid infection . Thus , despite being an immunologically evasive infection , dramatic hepcidin up-regulation and hypoferremia remain features of typhoid in humans . Mechanistic links between hepcidin and hypoferremia should not , however , be concluded from observational data such as these . Nevertheless , based on data from other settings linking hypoferremia with hepcidin upregulation during infection and inflammation [11 , 53] , we hypothesize that hepcidin plays a role in mediating hypoferremia and that the hypoferremia reflects rapid sequestration of iron in macrophages during acute typhoid infection . The interplay between Salmonella enterica infection and iron has been well studied , typically through using in vitro or in vivo S . Typhimurium models . The iron exporter ferroportin is upregulated via NO-mediated Nrf2 activation in ex vivo S . Typhimurium-infected macrophages , reducing macrophage iron availability—a state that restricts bacterial replication [54–56] . Despite this mechanism , hepcidin induction and hypoferremia are still observed during invasive murine S . Typhimurium infection , associating with macrophage iron sequestration via reduced ferroportin activity; interference with hepcidin upregulation in this context , leading to reduced cellular iron levels , is protective for the host [55] . Correspondingly , hepcidin administration to infected ferroportin-expressing cells in vitro enhances bacterial replication [54] . As reflected by their different pathologies , there are key differences between non-typhoidal and typhoidal Salmonella enterica serovars , despite high degrees of sequence homology [57] . These include the expression of virulence determinants ( most notably the Vi-capsular antigen ) and inactivation of over 200 genes in S . Typhi compared with its cousin S . Typhimurium [57] . Interestingly , several of these inactivated genes relate to iron acquisition pathways [58] . There is evidence that S . Typhi relies heavily on the fepBDCG enterobactin ferric iron uptake system [57] , which is upregulated in isolates from typhoid patients [59]; the upregulation of this system likely reflects the difficulty of obtaining iron from a host environment where iron availability is typically scarce . In conclusion , during human S . Typhi infection , where hepcidin is strongly upregulated and a marked hypoferremia is observed , we hypothesize that hepcidin activity and macrophage iron retention are dominant over any macrophage cell-intrinsic protective mechanisms aimed at reducing cellular iron content . Stimulating a strong hepcidin response may represent another bacterial strategy for ensuring iron supply to facilitate replication . Therefore , in typhoid ( and possibly other macrophage-tropic intracellular bacterial infections ) , hepcidin-induced hypoferremia may be actively disadvantageous to the host rather than being a stereotypical protective response to infection [11] . A recent study in humans demonstrated that spiegelmer-based hepcidin neutralization during experimental endotoxemia can prevent induction of hypoferremia [53] . Whether targeted manipulation of hepcidin and host iron distribution offers a potential strategy for treating intracellular infections should be investigated further , particularly in an era of increasing antibiotic resistance . | An adequate supply of iron is essential for both human hosts and their infecting pathogens . Hepcidin is the human hormone that controls the quantity and distribution of iron throughout the body . During infections , hepcidin activity may redistribute iron away from serum and into macrophages , potentially affecting pathogen replication , depending on the niche of the invading microbe . However , the involvement of hepcidin in human bacterial infections remains poorly investigated . Similarly , the pathogenesis of typhoid fever , caused by infection with Salmonella Typhi is also poorly understood . We therefore investigated the behaviour of hepcidin and other iron/inflammation-related parameters during the course of typhoid fever in human volunteers challenged experimentally with Salmonella Typhi . Hepcidin concentrations rose rapidly during acute typhoid infection , in parallel with fever . Hepcidin induction was accompanied by a rapid decline in serum iron concentrations , likely reflecting iron sequestration in macrophages ( a preferred replication site of Salmonella Typhi ) . The extent of hepcidin upregulation associated with the extent of serum iron starvation . We hypothesize that hepcidin activity during acute typhoid infection in humans may elevate iron levels in the niche used by the pathogen for replication . Targeting macrophage iron retention should be evaluated as a potential strategy for limiting typhoid fever . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| []
| 2015 | Rapidly Escalating Hepcidin and Associated Serum Iron Starvation Are Features of the Acute Response to Typhoid Infection in Humans |
Huntington's disease ( HD ) is an autosomal dominantly inherited disorder caused by the expansion of CAG repeats in the Huntingtin ( HTT ) gene . The abnormally extended polyglutamine in the HTT protein encoded by the CAG repeats has toxic effects . Here , we provide evidence to support that the mutant HTT CAG repeats interfere with cell viability at the RNA level . In human neuronal cells , expanded HTT exon-1 mRNA with CAG repeat lengths above the threshold for complete penetrance ( 40 or greater ) induced cell death and increased levels of small CAG-repeated RNAs ( sCAGs ) , of ≈21 nucleotides in a Dicer-dependent manner . The severity of the toxic effect of HTT mRNA and sCAG generation correlated with CAG expansion length . Small RNAs obtained from cells expressing mutant HTT and from HD human brains significantly decreased neuronal viability , in an Ago2-dependent mechanism . In both cases , the use of anti-miRs specific for sCAGs efficiently blocked the toxic effect , supporting a key role of sCAGs in HTT-mediated toxicity . Luciferase-reporter assays showed that expanded HTT silences the expression of CTG-containing genes that are down-regulated in HD . These results suggest a possible link between HD and sCAG expression with an aberrant activation of the siRNA/miRNA gene silencing machinery , which may trigger a detrimental response . The identification of the specific cellular processes affected by sCAGs may provide insights into the pathogenic mechanisms underlying HD , offering opportunities to develop new therapeutic approaches .
Huntington disease ( HD ) , a dominantly inherited neurodegenerative disorder , is caused by an abnormal CAG expansion within the first exon of the Huntingtin gene ( HTT ) , leading to an expanded polyglutamine ( polyQ ) track in the HTT protein . HTT is ubiquitously expressed in the cytoplasm of most cells in the body , with higher expression levels in brain and testis [1] , [2] . However the disease shows a selective pattern of neurodegeneration , with clear effects in the cerebral cortex , and a more pronounced neuropathology in the striatum [3] , [4] . The number of CAG repeats influences the severity and the age of onset of the disease . Longer expansions associate with a more severe form and an earlier manifestation of the disease [5] . It has been widely reported that the polyQ expansion in the HTT protein leads to protein aggregation and cell toxicity [6] , a mechanism thought to be primarily involved in several neurological disorders caused by CAG repeats [7]–[10] . However , whether the mutant HTT aggregates are pathogenic , incidental or neuroprotective is still controversial . It has been shown that mutant HTT aggregates may function as sinks where essential proteins are sequestered [11] , compromising cell survival [12] . Other studies show that increased levels of diffuse mutant HTT are responsible for neuronal cell death [13] . In agreement with the two possibilities , the activation of autophagy , reduce both soluble mutant protein and aggregate levels , and reduces toxicity [14] , [15] . In addition to the widely described pathogenic role of expanded polyQ tracks , several studies have also shown that different neurodegenerative disorders caused by trinucleotide repeat expansions may involve RNA-mediated mechanisms [16] , [17] . These include the sequestration of RNA-binding proteins by the expanded trinucleotide repeats , and activation of a variety of pathways such as RNA interference ( RNAi ) and protein misfolding pathways . The understanding of how expanded-repeat RNAs confer neurotoxicity is crucial to developing effective treatments . A neurotoxic effect for CAG-expanded transcripts has been recently demonstrated in Drosophila models of Ataxin-3 [18] and Myotonic Dystrophy [19] . In the later , the authors propose a pathogenic role of siRNAs derived from complementary sense and anti-sense expanded ( CUG/CAG ) transcripts . In line with this , double-stranded CAG/CUG repeat RNA produced by bidirectional transcription induces neurodegeneration and movement disorder in Drosophila model [20] . This neurotoxic effect is largely dependent on Dicer activity and linked to the formation of ( CAG ) 7mers . In addition , other studies describe that trinucleotide repeated transcripts form secondary structures [21] that can be cleaved by Dicer in vitro [22] , [23] resulting in the generation of trinucleotide repeated short RNAs . Together , these data suggest that different mechanisms lead to the formation of aberrant small RNAs in trinucleotide expansion diseases . Huntington's disease like 2 ( HDL2 ) is caused by a CTG . CAG expansion in the JPH3 gene , and the neuropathologic outcome and clinical features largely resemble HD . CUG expansions in the JPH3 gene correlates both with the formation of RNA foci and cell toxicity , suggesting RNA mediated toxicity [24] , [25] . RNA pathogenic mechanisms have been little explored in HD . Expanded HTT transcripts are retained in the nucleus of human HD fibroblasts and co-localize with the MBNL1 protein [26] , a splicing factor involved in the pathogenesis of CTG/CAG expanded transcripts [27] . In addition , mutant HTT protein alters microRNA ( miRNA ) biogenesis [28] , and a strong miRNA deregulation is observed in HD brains [29]–[32] , which may contribute to the aberrant gene expression observed in HD . Here we provide evidence for a pathogenic role of the mutant HTT RNA . CAG-expanded HTT RNA can be processed to generate CAG-repeated short RNAs with neurotoxic activity . We show that expanded HTT toxic effect is dependent on RNA-induced silencing complex ( RISC ) and further demonstrate that expanded HTT participates in posttranscriptional gene silencing of genes containing pure and interrupted CTG repeats . This , together with HTT polyglutamine toxicity , may contribute to the neurodegeneration pattern observed in HD .
To evaluate the contribution of CAG-expanded RNA in HD pathogenesis , we generated vectors expressing unexpanded and CAG-expanded forms of exon 1 of human HTT ( HTT-e1 ) . HTT-e1 constructs containing 23 CAG repeats ( 23*CAG ) were used as wild-type ( unexpanded ) model . For the expanded HD mutation , we generated HTT-e1 constructs containing 80 CAG repeats ( 80*CAG ) . Each set of vectors was produced as a form that could be translated into protein , and as a variant lacking the translation initiation codon , that was only expressed as RNA . Due to the reduced size of HTT-e1 , the different variants were cloned into a pIRES-GFP expression vector . This strategy allowed the monitoring of the transfected cells avoiding the generation of a GFP fusion protein that could lead to artefactual localizations ( Figure 1A , 1B and Figure S1 ) . A recent study reveals that RNA transcripts with expanded CAG repeats can be translated in the complete absence of a starting ATG [33] . Thus , we evaluated whether the constructs lacking translation initiation codon expressed polyglutamine , using the anti-glutamine monoclonal antibody 1C2 ( Figure 1B ) . The different HTT-e1 constructs were efficiently expressed , as shown by PCR amplification of HTT-GFP ( Figure 1B left panel ) . However , we only detected a polyglutamine track in the constructs containing the ATG starting codon , suggesting that repeat-associated non-ATG translation ( RAN translation ) is not compatible with the type of vector used to clone the different HTT-e1 forms , at least for polyglutamine production . Since RAN translation can occur in all frames [33] , the possibility that CAG expansion produce homopolymeric polyalanine and polyserine proteins cannot be ruled out . It is worth mentioning that 1C2 antibody does not allow quantitative comparison of the levels of 23*CAG-Prot versus 80*CAG-Prot; thus , the differences in the intensity of the 1C2 detected bands is a consequence of the number of glutamines in each HTT-e1 , expressed vector ( Figure 1B right panel ) . We transiently transfected these four different HTT-e1 expressing vectors in differentiated human neuroblastoma cells ( SH-SY5Y ) as a post-mitotic neuronal cell model . Transfection experiments revealed that CAG expansion in HTT mRNA was sufficient to induce a dramatic cytotoxic response in differentiated SH-SY5Y cells ( Figure 1C ) . Cell toxicity assays demonstrated that both CAG-expanded constructs ( translated and non-translated forms ) drastically affected neuronal cell viability , only differing in the timing of the response , that was earlier for the 80*CAG-RNA construct . However , a expanded polyglutamine expressing vector using CAA instead of CAG repeats ( 80*CAA ) , induced a mild toxic effect at the latter time-point that did not reach statistical significance ( Figure 1C ) . These results suggest that the toxic effect induced by the expanded polyglutamine tract is specific for expanded CAG . The HTT RNA toxicity was further confirmed with the analysis of early and late apoptotic markers . The results obtained revealed that the expression of CAG-expanded HTT-e1 RNA is sufficient to induce nuclear condensation ( Figure 1D ) and caspase 9 activation ( Figure 1E ) , processes previously reported to occur in HD brain samples [34] , [35] . On the contrary , 80*CAA expressing vector induced milder caspase 9 activation . These data point to a direct link between the toxic effect of expanded HTT RNA and an intrinsic apoptotic process . Transcripts containing long hairpin structures composed of CNG repeats are Dicer targets [22] , [23] . The resultant sRNA products may trigger aberrant gene silencing with putative downstream detrimental effects . To test whether mutant HTT-e1 toxicity was associated to sRNA related mechanisms , we isolated the sRNA fraction ( <100 nt ) from cells expressing the 23*CAG and 80*CAG forms of HTT-e1 , and transfected these sRNA into differentiated human SH-SY5Y neuroblastoma cells . Cell viability assays demonstrated that the sRNA population obtained from cells expressing 80*CAG-PROT and 80*CAG-RNA constructs , induced a remarkable cell death response ( Figure 2A ) , compared to the sRNA population originated from cells expressing the 23*CAG control constructs . These results indicate that the expression of expanded HTT-e1 RNA is sufficient to deregulate the sRNA profile , thereby impairing neuronal viability . Furthermore , the sRNA fraction of cells expressing 80*CAA-PROT failed to induce cell toxicity , suggesting that the sRNA detrimental effect is linked to expanded constructs containing CAG repeats . In agreement with previous studies demonstrating the generation of trinucleotide-repeated sRNA from triplet-expanded transcripts [19] , [23] , the expression of CAG-expanded HTT RNA led to the generation of CAG-repeated sRNAs ( sCAG ) , of around 21 nt long ( Figure 2B ) . The identity of these products was further confirmed by direct sequencing of the PCR products ( Figure S2 ) and northern blotting ( Figure S3 ) . However , cells expressing the CAA expanded construct failed to produce sCAG , suggesting that the production of these species is not an experimental epiphenomenon . Variable penetrance for alleles carrying 36–39 CAG repeats has been noted , but the disease appears fully penetrant when the repeat numbers are above 40 [5] . To confirm the sRNA toxicity in HTT carrying a moderate number or repeats , we generated HTT constructs with 35 , 40 and 48 CAG repeats ( Figure S4 ) . We performed transfection experiments using the sRNA fractions of cells expressing HTT vectors with 23*CAG ( normal ) , 35*CAG ( normal ) , 40*CAG ( pathogenic ) 48*CAG ( pathogenic ) and 80*CAG ( model for juvenile HD ) and subsequently determined cell viability . The sRNA fraction isolated from 40*CAG , 48*CAG and 80*CAG expressing cells induced a significant toxic effect ( Figure 2C ) . Furthermore , the severity of the toxic effect in differentiated SH-SY5Y driven by the sRNA fractions was associated to the length of the CAG stretch , as previously described for the full protein [36] ( Figure 2C ) . In addition , the pools of sRNAs isolated from 40*CAG , 48*CAG and 80*CAG expressing cells contained progressively increasing amounts of sCAGs when compared with that of the 23*CAG and 35*CAG expressing cells ( Figure 2D ) . These results suggest that sRNAs derived from moderately expanded HTT are sufficient to induce a detrimental response and further indicate that expansions above 40*CAG repeats are enough to produce significantly increased amounts of sCAG and a parallel toxic effect . To analyse the role of sCAG products in HTT sRNA toxicity , we then co-transfected sRNAs pools derived from cells expressing 23*CAG or 80*CAG vectors along with either antisense RNA oligonucleotides that specifically block the action of sCAG ( anti-sCAG ) , or scrambled inhibitors as negative controls ( Scrambled sRNA inhibitors ) . The toxic effect of 80*CAG- versus 23*CAG-derived small RNA was not affected in cells transfected with a scramble siRNA . However anti-sCAG significantly decreased the detrimental effect of the HTT-e1 expanded constructs ( Figure 2E ) . We therefore propose that the generation of sCAG is a key element in the toxicity mediated by CAG-expanded HTT-e1 . We next examined whether sCAG were detected in different brain areas of the R6/2 HD mouse model , a transgenic line that over-expresses the exon 1 of human HTT with more than 100 repeats , and recapitulates many of the key features found in patients with HD [37] . R6/2 mice of 8 weeks of age exhibited deficits in coordination and activity , striatal atrophy , HTT-aggregate accumulation and down-regulation of striatal-neuron integrity markers [38] . RT-PCR analysis revealed increased sCAG levels in the cortex and striatum of R6/2 mice with respect to their wild-type littermates ( Figure 3A ) , two brain areas preferentially affected in HD . However , no differences in the expression of sCAG species were detected in the cerebellum and hippocampus of R6/2 mice . These results suggest the existence of region specific mechanisms modulating sCAG biogenesis and/or stability in the R6/2 mouse model . sCAG levels were subsequently analysed in post-mortem brain samples from HD patients and control subjects . RT-PCR analyses confirmed an increase of sCAG in the frontal cortex and caudate regions from HD samples ( Figure 3B and Figure S5 ) . The PCR products were sequenced , and sCAG species of 18 nt length were found in both control and HD brain samples . However , sCAG species of 21 nt-long were only detected in HD human brain samples ( Figure S6 ) . To further validate the pathogenic role of sCAG in human brain samples , we isolated the sRNA fractions from control and HD frontal cortex and caudate , and transfected them into differentiated SH-SY5Y cells . HD-sRNAs reproduced the toxicity exerted by the expanded HTT-e1 sRNAs ( Figure 3C ) . Furthermore , anti-sCAG dramatically diminished the toxic effect of HD-derived sRNA , supporting a pathogenic role of sCAG species produced in HD brains . In CAG-repeat expansion diseases , Dicer-dependent mechanisms result in the formation of sCAG with putative functions in pathogenic gene silencing [19] . We therefore investigated whether RNAi machinery is involved in the generation and function of sCAG in HD . To that end we performed Dicer knockdown experiments in differentiated SH-SY5Y cells that were subsequently transfected with HTTe1-expressing vectors . Dicer depletion prevented the generation of sCAG ( Figure 4A and Figure S7 ) and efficiently mitigated 80*CAG RNA toxicity in SH-SY5Y cells , as indicated by the decrease in LDH release and the inhibition of caspase 9 cleavage ( Figure 4B ) . This result suggest that the toxic effect of the 80*CAG-derived sRNA is caused by a major pathogenic pathway triggered by sCAG . Since the generation of sCAG was largely dependent on Dicer , we determined Dicer levels in several brain areas of control and R6/2 mice ( Figure S8 ) . Dicer expression was significantly decreased in the hippocampus and cerebellum of R6/2 mice while no differences in sCAG levels were detected in these areas ( Figure 3A ) , suggesting that this could be a factor modulating sCAG generation from mutant HTT . To explore the potential mechanisms of HTT sRNA toxicity and sCAG deleterious effect , we next examined the relationship between 80*CAG toxicity and Ago2 activity , a key factor in miRNA/siRNA gene silencing [39] , [40] . Cell viability assays revealed that the toxic effect of sRNA pools originated from cells expressing 80*CAG-PROT and 80*CAG-RNA was diminished in cells depleted of Ago2 ( Figure 4C ) . This result indicates that Ago2 is an important player in the pathogenic effect of 80*CAG-derived sRNA species . The initiation of a sCAG-mediated gene silencing process requires the incorporation of sCAG into RISC . To test whether sCAG could be loaded into the Ago2 silencing complex , we transfected the HTT-expressing constructs into SH-SY5Y cells stably expressing Flag-Ago2 . We performed immunoprecipitation ( IP ) assays using anti-Flag antibodies for Ago2 IP or anti-V5 antibodies as negative control ( Figure S9 ) and RNA bounded to immunoprecipitated Flag-Ago2 was isolated . The analysis of the Ago2-associated sRNA revealed that sCAG generated from mutant HTT RNA efficiently bound to the Ago2 complex ( Figure 4D ) . These results , along with the protective role of anti-sCAG , suggest that sCAG initiate a transcriptome-dependent detrimental response through Ago2-mediated gene silencing mechanisms . To evaluate the direct role of Ago2 in the toxic effect of expanded HTT-e1 we restored Ago2 levels in cells depleted of Ago2 , and determined cell death ( Figure 4E ) . Restoration of Ago2 levels by the co-transfection of a Flag-Ago2 expressing vector significantly re-established HTT toxicity ( Figure 4E ) . In humans , the Ago subfamily consists of Ago1 , Ago2 , Ago3 and Ago4 that guide both siRNAs and miRNAs to complementary sites on target RNAs to modulate their expression [41] . We therefore asked whether Ago2 was the critical mediator on HTT sRNA toxicity or other Ago proteins could be participating as well . Given that Ago3 and Ago4 are not significantly expressed in SH-SY5Y cells ( data not shown ) , we analyzed Ago1 contribution in HTT toxicity . The toxic effect of expanded HTT-e1 was significantly decreased in cells with reduced levels of Ago1 , suggesting that mutant HTT effect is also mediated by Ago1 ( Figure S10 ) . Since Ago2 is the only member of the Ago family with endonucleolytic activity [40] , [42] , the results linking both Ago2 and Ago1 with HTT toxicity suggest that sCAG may be modulating gene expression through target mRNA degradation and/or translational inhibition , as described for miRNAs [43] . To validate the possible detrimental effect of sCAG in human cells , a synthetic 21-nt long CAG-repeated siRNA , [ ( CAG ) 7 siRNA] , was delivered to a panel of primary human cell lines including , breast ( HMEC ) , bladder ( UROTSA ) and pancreatic cells ( HPDE ) . Differentiated SH-SY5Y cells were used as a neuronal model . ( CAG ) 7 siRNA impaired cell viability at variable levels in different cell types . Although these results indicate that ( CAG ) 7 siRNA detrimental effect is not restricted to SH-SY5Y cells , this cell model displayed significant higher sensitivity to ( CAG ) 7 ( Figure 5A and Figure S11 ) . We performed additional experiments in SH-SY5Y cells following several differentiation protocols that result in differential cell morphology ( Figure 5B and Figure S12 ) . These assays demonstrated a correlation between the type of differentiation of SH-SY5Y cells and the sensitivity to ( CAG ) 7 which supports a transcriptome-dependent response in sCAG-mediated toxicity . To validate the gene silencing activity of sCAG , and determine whether a full or partial complementary with the target genes was needed , we generated firefly luciferase-expressing vectors carrying a ( CTG ) 14 stretch in the luciferase 3′UTR . We also developed constructs with the sequence ( CAG ) 14 , which offer an interrupted binding to sCAG . In an attempt to evaluate the consequences of an imperfect matching , we also cloned the sequences 5′-TCCGTGCTGAGCCTGCCTGTCGTCTGTG-3′ and 5′-TGCTAGTATCAGATCTGCTGTGGAATTG-3′ , present in the genes ADORA2A and MEIS2 respectively . These two genes are downregulated in affected brain areas of HD patients and brains from the R6/2 mouse model [44] . Furthermore , in silico analysis of the sCAG and MEIS2 or ADORA2 duplex stability using RNA hybrid suggests that MEIS2 and ADORA2 could be putative targets of sCAG . HeLa cells were co-transfected with the different combinations of HTT-expressing vectors and the luciferase vectors , and luminescence was measured 24 hours after transfection . Expanded HTT RNA was able to moderately silence luciferase expression in a construct containing a CTG14 sequence in the 3′UTR , compared to control luciferase vectors , and the non-expanded forms of HTT-e1 ( Figure 6A ) . These experiments suggest that sCAGs derived from expanded HTT are involved in post-transcriptional silencing of genes containing CTG repeated tracks . In addition , we also detected a moderate reduction of luciferase activity in constructs harbouring the sequence CAG14 , suggesting that , expanded-HTT-e1 targets genes with CAG repeats , although the mechanism related may differ from the canonical miRNA/siRNA silencing pathways . Interestingly , HTT construct expressing 80*CAG moderately decreased the expression ( 10% of reduction ) of the reporters containing ADORA2A and MEIS2 regions ( Figure 6A ) . This result indicates that full complementary between sCAG and its target genes are not needed to induce gene silencing . Therefore , sCAG may behave as siRNA molecules , but also as miRNA-like species , and offer an additional explanation for the broad gene expression deregulation observed in HD [44] . This possibility was further confirmed by RT-PCR quantification of ADORA2 and MEIS2 expression in SH-SY5Y cells transfected with the HTT-e1 expressing vectors . ( Figure 6B and 6C ) . The results obtained reproduced the decrease observed in the luciferase assays . Accordingly , the expression of CAA expanded constructs , which failed to generate sCAG , didn't affect ADORA2A or MEIS2 expression levels . We also evaluated if the 80*CAG construct silenced the expression of genes containing a CUG tract , including DMPK , ASTN2 and ZFR ( Figure S13 ) . The expanded HTT-e1 induced variable silencing of the different genes that did not correlate with the number of CUG repeats . DMPK is the transcript with higher number of CUG repeats in the 3′UTR with 13 CUG repeats; ASTN2 presents a moderate number of consecutive CUG repeats and ZFR transcripts contains a region harboring 4*CAG immediately followed by 5*CTG . The variability in the dowregulation response suggests that the number of CUG repeats it's not a key factor in mutant HTT-e1 silencing activity . We evaluated a possible enrichment in CTG regions ( of a minimal size of 7 ) either in the full transcript or in the 3′-UTR of HD downregulated genes . For this analysis we considered the downregulated genes ( <−1 , 2 downregulation and p<0 , 05 ) , upregulated genes ( >1 , 2 upregulation and p<0 , 05 ) and the group of genes that did not show significant expression deregulation , provided in the study by Hodges et al [44] . No significant enrichment in genes containing CTG regions was detected in the downregulated , upregulated or non-regulated genes ( X-square p>0 , 05 ) , suggesting that the overall mRNA gene expression deregulation was dependent on several pathogenic factors besides sCAG-mediated gene silencing . We next asked whether sCAG could be inducing gene silencing by target mRNA degradation or by translation inhibition . The levels of MEIS2 protein were analyzed in differentiated SH-SY5Y cells transfected with normal and expanded HTT-e1 . Cells were lysated 24 hours after transfection , a time point in which CAG-expanded HTT RNA toxicity was validated . Given that neural cells are more sensitive to HTT-e1 expression and cell death can be detected 21 h after transfection , MEIS2 levels were normalized by Actin and also referenced to GFP expression , which indicates the percentage of transfected living cells at the time of the analysis . Figure 6D shows MEIS2 protein levels after performing this analysis , confirming that CAG-expanded HTT-e1 induce a reduction in MEIS2 levels by 10% , in agreement with the luciferase reporter assays and mRNA quantification . The decrease in MEIS2 protein levels is similar to the reduction in MEIS2 mRNA level , which may suggest that mRNA degradation is the main mechanism in the particular case of MEIS2 post-transcriptional gene silencing . However , an exhaustive study should be performed to fully identify sCAG targets and characterize the mechanisms of gene silencing in each particular case .
The latest evidences suggest that RNA detrimental effects contribute to neurodegeneration in a number of trinucleotide repeat expansion diseases . However , these processes have not been extensively addressed in HD , where pathogenesis has been traditionally thought to involve the mutant HTT protein . Our results suggest an RNA pathogenic mechanism in HD that involves the aberrant generation of sCAG RNA species with an inherent toxic effect in a neuronal cell model . We have shown that the generation of sCAG species from expanded HTT exon 1 is largely dependent on Dicer , in agreement with previous studies showing that triplet repeats formed by CNG units adopt hairpin structures that become sliced to sCNG by dicer [22] , [23] . In addition , it has become apparent that most of the expanded repeat disease loci have transcription occurring from both strands , raising the possibility that the complementary repeat RNAs form double-stranded structures susceptible to be processed by Dicer . Recently , a natural antisense transcript for HTT ( HTTAS ) has been described , covering the exon-1 CAG repeat [45] . Although HTTAS is under the control of a weak promoter , it is expressed throughout the brain and other tissues . Therefore , the production of sCAG in HD brains shown in the present study and in fibroblasts of HD patients [22] may originate both from HTT expanded hairpin structures and HTT/HTTAS double stranded RNAs . Importantly , CAG repeat lengths above the threshold for complete penetrance ( 40 or greater ) generated increased amounts of sCAG compared with non-pathogenic repeat lengths . Furthermore , our data suggest that the generation of sCAG correlated with the length of the repeat , being sCAG levels progressively higher in cells transfected with HTT-exon-1 constructs harboring 40 , 48 and 80 CAG repeats , respectively . This correlated with a gradually increasing detrimental effect driven by the small RNAs fraction of cells expressing HTT-e1 with 40 , 48 or 80 CAG repeats , respectively . These results agree with the increased severity of the disease in HD cases presenting extremely long CAG expansions in the HTT gene [46] . The amount of sCAG products was not equivalent in different brain areas in a HD mouse model , where increased sCAG levels were detected in the more affected areas . Our data suggest that decreased levels of Dicer could contribute to explain the lack of sCAG increase in the hippocampus and cerebellum of R6/2 mice . However , It is worth mentioning that Dicer activity is subject to regulation that affects the accumulation of miRNAs and probably sCAG . Recent work has identified a battery of proteins that regulate processing either interacting with Dicer or with miRNA precursors , being the activity of some regulators restricted to specific miRNA families [47] . In this context , whether Dicer is particularly active in the cortex and the striatum under basal conditions and/or in HD , the possible mechanisms modulating Dicer activity in specific areas and/or diseased state and its relevance to human disease are open questions that deserve specific research . Our data indicate that the toxic effect of the sRNA fraction generated by expanded HTT is dependent on Ago proteins and is abolished by anti-sCAG . Furthermore , increased levels of sCAGs were found in Ago2 immunoprecipitates of cells expressing expanded HTT-e1 , suggesting that sCAG-driven gene silencing may underlie HTT-RNA toxicity . In agreement with RISC-dependent mechanisms , expanded HTT-e1 constructs moderately silenced genes showing pure and interrupted CUG tracks , complementary to sCAG . However , we did not detect a significant enrichment in mRNAs harboring CUG-tracks among those found to be downregulated in human HD brain samples [44] . This suggests that gene expression perturbation in HD brains may reflect primary and secondary pathogenic triggers . In addition , the possibility that sCAG may act through translational repression , a gene silencing mechanism also described for miRNAs [43] , cannot be ruled out . Interestingly , expanded HTT induced similar silencing when using CAG repeats as the target sequence in the luciferase assay . The main structural requirements for gene targeting in miRNA-RISC mediated gene expression regulation are well defined for the most expressed miRNAs , including seed region perfect pairing in the 3′-UTR of the target genes [48] . However , knowledge about the determinants governing gene targeting is far from complete . In fact , targeting can occur through sites other than the 3′-UTR and seed region base pairing is not always required [49] . Whether imperfect base pairing between the CAG tracks in the small RNA and the target genes is compatible with the location and configuration of the sCAG-RISC complex , is an interesting question that should be specifically addressed . In addition , since trinucleotide repeats have been shown to bind proteins , additional functions for sCAGs should be considered . Gene expression modulation by miRNAs recently included a decoy function , where miRNAs bind to proteins that regulate gene expression , thus modulating their activity [50] . The characterization of the sCAG binding proteins that could have consequences in gene expression regulation may shed light to possible additional RNA related pathogenic mechanisms . In summary , we propose a pathogenic RNA dependent mechanism in HD by which sCAG produced over a threshold are neurotoxic . In HD , this mechanism may complement other RNA dependent processes including miRNA deregulation [28]–[32] and possible alterations in alternative splicing driven by MBNL1 sequestration [26] , [51] ( Figure 7 ) . The detrimental effect may depend not only on the amounts of sCAG generated , but also on the target transcriptome and factors modulating RISC function . These aspects may contribute to sCAG variable vulnerability in different human cells observed in the present study . sCAG induced pathogenesis may underlie common phenotypes in triplet repeat diseases showing CAG expansions in different coding RNAs ( leading to polyglutamine expansions in several proteins ) and untranslated RNAs [19] . The identification of the specific sCAG-targeted genes and the cellular processes affected by sCAG should pave the way for the development of new therapeutic approaches for HD and other CAG-repeat expansion diseases .
Human Mammary Epithelial Cells ( HMEC ) were maintained in MEBM medium supplemented with Bullet-kit ( Lonza ) , Human Pancreatic Duct Epithelial Cells ( HPDE ) were cultured in KSFM medium ( Invitrogen ) supplemented with epithelial growth factor ( 0 . 1–0 . 2 ng/mL ) and bovine pituitary extract ( 25 µg/mL ) . UROTSA cells were maintained in RPMI medium ( Invitrogen ) supplemented with 10% FBS ( Fetal Bovine Serum , Invitrogen ) . HeLa cells and SH-SY5Y neuroblastoma cells were maintained in Dulbecco's Modified Eagle's Medium ( DMEM , Invitrogen ) supplemented with 10% FBS , 2 mM L-glutamine , 100 units/ml penicillin and 100 µg/ml Streptomycin ( GIBCO , Invitrogen ) . In the case of SH-SY5Y cells , FBS was heat inactivated for 45 min at 56°C prior to use . Unless otherwise indicated , SH-SY5Y cells differentiation was performed culturing the cells in the standard growing medium containing 10 µM retinoic acid ( RA ) during four days . The media was then replaced by fresh medium containing 80 nM of 12-O-tetradecanoylphorbol-13-acetate ( TPA ) during five additional days [52] Different neuronal differentiation protocols are provided in Figure S8 ) . Different forms of the exon 1 of the HTT gene ( HTT-e1 ) differing in the CAG repeat length ( 23*CAG- , 35*CAG- , 40*CAG- , 80*CAG- or 80*CAA-PROT; and 23*CAG- , 80*CAG-RNA ) were synthesized by Geneart . Flanking EcoRI restriction sites were added during the synthesis that were used to sub-clone the HTT-e1 variants into the pIRES2-EGFP vector ( BD Biosciences , Clontech ) . Not-translatable constructs lack the translation initiation codon ( AUG ) and the second methionine ( AUG ) found in HTT exon1 ( Figure 1A ) . All the transfection experiments were performed using Lipofectamine 2000 ( Invitrogen ) , according to the manufacturer's instruction and at a 60% cell confluence . ( CAG ) 7 ( 5′CAGCAGCAGCAGCAGCAGCAG-3′ ) and control , scrambled siRNA ( 5′-GCGACGUUCCUGAAACCAC-3′ ) were purchased from Dharmacon and were administered at a final concentration of 50 nM , unless otherwise indicated . The anti-sCAG small RNA ( LNA modified 5′- ( CTG ) 7 ) , and scrambled sequences ( LNA modified 5′-GTGTAACACGTCTATACGCCCA-3′ ) were ordered from Exiqon . Both anti-sCAG and the corresponding scrambled inhibitor were transfected at a final concentration of 60 nM . Transfections with sRNA pools were performed using 35 ng of each sRNA pool per well ( quintuplicates , 96wells multiwell ) . Dicer , Ago1 and Ago2 knockdown experiments were performed by a double transfection procedure; consisting in the transfection of the Scrambled , Ago2 or Dicer siRNA in the first assay ( 50 nM ) , and the co-transfection of the siRNA and HTT construct 48 hours later at 75 nM and 400 ng , respectively , in MW6 plates . Dicer siRNA ( 5′- GCUCGAAAUCUUACGCAAAUA-3′ ) , Ago1 siRNA ( 5′- CAUCAGGACUGUUGAGUAA -3′ ) and Ago2 siRNA ( 5′-GCACGGAAGUCCAUCUGAA-3′ ) were purchased from Dharmacon . A siRNA against the 3′UTR of Ago2 ( siAgo2-3′UTR: 5′-GGAAATATGGTTTGCTAAA-3′ ) was used in the HTT toxicity rescue experiments ( Figure 4E ) . Transfection efficiency in experiments using siRNA or sRNA pools was determined at each experimental condition using siGLO transfection indicator ( Dharmacon ) . Transfection conditions were optimized in order to obtain similar transfection efficiencies ( ∼90% ) in all the cell lines analyzed . SH-SY5Y cells were transfected with a Flag/HA-AGO2 expressing vector ( Flag-tagged Ago2 expression vector was kindly provided by Prof . R . Shiekhattar ) . The plasmid encodes for a neo-resistance marker and transfected cells were grown in the presence of 800 µg/ml of Geneticin ( G418 , Gibco Laboratories ) for 10–14 days . Single clones were selected to generate monoclonal cell lines . Expression of Flag/HA-AGO2 protein was checked by western blot and immunofluorescence in several cell clones . For protein extraction , cells were rapidly rinsed with ice-cold PBS and solubilized with a lysis buffer described in [53] . Cells were then scraped off , incubated on ice for 15 min and centrifuged at 14000 rpm for 15 min . Samples were resolved in 10% SDS-PAGE gels and transferred to nitrocellulose membranes using the iBlot Dry Blotting System ( Invitrogen ) . Membranes were blocked for 1 h with 10% skimmed milk in TBS ( Tris-HCl , pH 7 . 5 , 10 mm; NaCl , 100 mm ) containing 0 . 1% Tween-20 ( TBS-T ) . Membranes were incubated at 4°C and overnight with primary antibodies ( diluted in TBS-T ) . After washing with TBS-T , membranes were incubated for 45 min at room temperature with the appropriate secondary antibodies ( diluted in TBS-T ) , and then washed again with TBS-T . Detection was performed by ECL Western blotting detection reagent ( Amersham Bioscience ) . Chemiluminescence was determined with a LAS-3000 image analyzer ( Fuji PhotoFilm Co . , Carrollton , TX , USA ) . Primary antibodies were anti-polyQ ( MAB1574 , 1∶1000 , Millipore ) , anti-GFP ( 1∶2000 , Molecular Probes , rabbit ) , anti-PARP ( 1∶5000 , BD Pharmigen , mouse ) , anti-cleaved caspase 9 ( 1∶1000 , Cell Signaling , rabbit ) , anti-Dicer ( 1∶500 , Abcam , mouse ) , anti-Ago2 , ( 1∶500 , Abnova , clone 2E12-1C9 ) . Anti-Ago1 antibody ( 1∶1000 , rat ) was kindly provided by Dr . G . Meister . Anti-GAPDH ( 1∶4000 , Chemicon , mouse ) , anti-α-Actin ( 1∶5000 , Chemicon , mouse ) and anti-α-Tubulin ( 1∶50000 , Sigma , mouse ) were used as loading controls . Secondary antibodies were HRP-conjugated anti-mouse , anti-rat and anti-rabbit ( 1∶2000 , DAKO ) SH-SY5Y cells grown on coverslides were rinsed several times with PBS and fixed for 20 min at room temperature with 4% paraformaldehyde in PBS . After rinsing , cells were permeabilized for 20 min in 0 . 5% Triton-X-100 in PBS . Non-specific binding sites were then blocked by incubating for 1 h in PBS containing 0 . 2% Triton X-100 and 10% FBS . Incubation with primary mouse anti-PolyQ antibody ( 1∶2000 , Millipore , clone 5TF1-1C21 ) was carried out overnight at 4°C in PBS containing 0 . 2% Triton-X-100 and 1% FBS . After washing , coverslides were incubated with secondary anti-mouse IgG Alexa 555 IgG ( Molecular Probes ) at a dilution of 1∶1000 for 1 h at room temperature . After washing , coverslides were mounted in Vectashield-DAPI solution , and cells visualized under a Leica microscope ( DMR ) . Images were captured using a digital camera ( Leica DC500 ) . Differentiated SH-SY5Y cells were transfected in 96 well plates and cell viability was determined 24 hours post-transfection with the 3- ( 4 , 5- dimethythiazol- . 2-yl ) -2 , 5-diphenyl tetrazolium bromide ( MTT ) assay . MTT was added to cell culture media at 0 . 5 mg/mL final concentration and incubated for 40 minutes at 37°C . Cells were then lysed with 100 µL of DMSO upon medium removal and absorbance was measured at 550 nm . In each experiment , determinations were performed in tetraplicates . Lactose dehydrogenase ( LDH ) released from dying cells was determined using the LDH assay ( Cytotox 96 , Promega ) according to the manufacturer's protocol , at different time-points following transfection ( see figure legends ) . Absorbance was recorded at 490 nm . LDH determinations were performed in quintuplicate . Cell death was also determined with the simultaneous staining of alive and dead cells using fluorescein diacetate ( FDA ) and propidium iodide ( PI ) , respectively in a double staining procedure . Cells were rinsed with PBS 1× and then incubated for 45 s at 22–25°C with 15 mg/ml FDA ( Sigma ) and 4 . 6 mg/ml PI ( Molecular Probes , Inc . , Eugene , OR , USA ) in PBS . The staining solution was replaced by PBS 1× and the stained cells were immediately examined under a Leica microscope . Ago2 Immunoprecipitations assays and the extractions of the Ago2-bounded RNA were carried out as described previously [54] . Flag-Ago2 immunoprecipitation was performed using ANTI-FLAG M2 affinity gel ( Sigma ) . ANTI-V5 affinity gel ( Sigma ) was used as a negative control for Ago2 IP . sCAG levels were determined by polyadenylation and RT-PCR . SH-SY5Y cells transfected with 100 nM ( CAG ) 7 were used as a positive control . Total RNA from cells or brain tissues was extracted using miRNeasy Mini kit ( Qiagen ) . Small RNA species ( <100 nt ) , were fractionated by size-exclusion column chromatography using Microcon Y-10 ( Millipore ) according to the manufacturer's instructions . Total RNA was treated with TURBO DNA-free kit ( Ambion ) . In vitro polyadenylation reactions were carried out using 1 µg of total RNA or 100 ng of sRNA enriched fraction and poly ( A ) polymerase ( Ambion ) for 1 h at 37°C in the presence of ATP ( 1 mM ) . Samples were then annealed with a polyT-adapter primer ( 5′-CGAATTCTAGAGCTCGAGGCAGGCGACATGGCTfGGCTAGTTAAGCTTGGTACCGAGCTCGGATCCACTAGTCCTTTTTTTTTTTTTTTTTTTTTTTTTAC-3′ ) prior to RT reaction . Specific primers recognizing the adapter and sCAG allowed the amplification of specific products by RT-PCR . sCAG expression levels in cells transfected with the non-expanded or expanded HTT-e1 were analyzed by RT-PCR or densitometry of the PCR amplified products . Total RNA , polyadenylated total RNA or sRNA was retrotranscribed using the Superscript III RT kit ( Invitrogen ) . Equal amounts of cDNA were mixed with SYBR Green PCR mix ( Roche ) . Five pmol of the forward primer ( designed on the CAG repeat sequence ) and reverse primer ( based on the adaptor sequence ) were used in each reaction . Amplification was done under the conditions of 15 sec at 95°C and followed by 55 cycles consisting in 1 min at 60°C and 2 min at 72°C in a LightCycler 480 Real-Time PCR System ( Roche ) . The entire experiments were repeated three times on independent RNA preparations . RNU66 expression was used as a reference small RNA . Values were also referenced to the GFP levels , which refers to the number of transfected living cells at a particular time . β-Actin was the endogenous reference gene for GFP normalization . Data are presented as the ratio between the normalized expression of sCAG ( sCAG/RNU66 ) or a particular gene ( gene/β-Actin ) and the normalized expression of GFP ( GFP/β-Actin ) . sRNA qRT-PCRs were performed with the following set of primers: sCAG Forward: 5′-CAGCAGCAGCAGCAGCAG-3′ , sCAG Reverse: complementary to the polyT adapter after polyadenylation ( 5′-CGAATTCTAGAGCTCGAGGCAGG-3′ ) ; RNU66 forward: 5′-GTAACTGTGGTGATGGAAATGTG-3′; RNU66 reverse: 5′- GACTGTACTAGGATAGAAAGAACC-3′; RNU6B forward: 5′-CGCTTCGGCAGCACATATAC-3′; RNU6B reverse: 5′-TTCACGAATTTGCGTGTCAT-3′ . mRNA qRTPCR were performed using the following primer sets: GFP forward: 5′-TGCAGTGCTTCAGCCGCTAC-3′; GFP reverse: 5-TCGCCCTCGAACTTCACCTC-3′; DMPK forward: 5′-TGGGCTACTCCTACTCCTG-3′; DMPK reverse: 5′- AGCTGTTTCATCCTGTGGG-3′; ASTN2 forward: 5′- GACATTCTACACGGAGCAGTAC-3′; ASTN2 reverse: 5′- GTGAGTGGACAAGACATCTGG-3′; ZFR forward: 5′- TGGGACTCAAAATCAGCTACG-3′; ZFR reverse: 5′- TGGTTCTGTTGATGGAATGGG-3′; β-Actin Forward: 5′-CTGGAACGGTGAAGGTGACA-3′; β-Actin Reverse: 5′-GGGAGAGGACTGGGCCATT-3′ . Regular detection of GFP and HTT-e1 expression was performed using the following set of primers GFP Forward/GFP Reverse , and HTT forward//pIRES-GFP reverse ( 5′-GTCCCTCAAGTCCTTCCAGC-3′/5′-GAACTTCAGGGTCAGCTTCG-3′ ) . Gene expression analysis of ADORA2A and MEIS2 genes were carried out using Taqman assays ( assay ID: Hs00169123_m1* for ADORA2A and assay No: Hs00542638_m1* for MEIS2 ) . Data were normalized using MRIP ( assay ID: Hs00819388_m1 ) as an endogenous reference gene . Amplification was done under the conditions: 15 sec at 95°C and followed by 55 cycles consisting in 1 min at 60°C and 2 min at 72°C on the ABI PRISM 7000 Detection system ( Applied Biosystems ) . The entire experiments were repeated four times on independent RNA preparations . qPCR results were analyzed using the 2−delta delta Ct method . The levels of the precursors and mature forms of miR-16 and miR-29 in normal cells and cells with decreased levels of Dicer were determined by polyadenylation and RT-PCR in total RNA , as previously described . The following oligonucleotides were used: for miR-16-1: 5′-TAGCAGCACGTAAATATTGGCG-3′; for miR-29a: 5′- TAGCACCATCTGAAATCGGTT-3′ . CAG PCR products were run on a 15% polyacrylamide gel and visualized by SybrSafe staining ( Invitrogene ) . PCR products were purified and ligated into pGEMT-easy vector . The sequencing reactions of the vectors were carried out using the Big Dye 3 . 1 Termination Cycle Sequencing Kit and DNA Sequencer ( ABI3100 ) from Applied Biosystems . Total RNA ( 30 µg ) or small RNA ( <100 nt long , 4 µg ) were resolved in a 15% acrylamide-7 . 5 M urea gel and transferred to Hybond-N+ membranes ( Amersham Bioscience ) in 0 . 5× Tris-buffered EDTA at 200 mA overnight at 4°C . The membranes were UV cross-linked and heated at 80°C for 1 h . LNA probes ( Exiqon ) and oligoribonucleotide probes against ( CAG ) 7 repeats ( 5′-CTGCTGCTGCTGCTGCTGCTG-3′ ) were labelled with γ-32P-dATP using Optikinase ( USB Corp . ) . LNA probe complementary to RNU6B was used as loading control ( 5′-CACGAATTTGCGTGTCATCCTT-3′ , Exiqon ) and an oligonucleotide probe complementary to GFP ( 5′- GAACTTCAGGGTCAGCTTGC ) was used to detect the expression of the different pIRES-HTT-e1-GFP vectors ( HTT-e1-IRES-GFP transcripts with a length of around 1 . 5 Kb ) . Oligonucleotide probes hybridisation and washings were performed at 50°C using PerfectHyb Plus buffer ( Sigma ) . The membranes were exposed to Fuji Imaging plates , scanned with a FLA-5000 PhosphorImager ( Fuji PhotoFilm Co . ) and quantified with ImageJ software . A series of firefly luciferase-based reporter constructs were used for quantitative measurement of sCAG-mediated post-transcriptional gene silencing in genes containing ( CUG ) 7/ ( CAG ) 7 stretches . The putative target sequences were obtained by the annealing of oligonucleotides with the desired sequence , containing an XbaI restriction site at their 5′ end . The resulting double stranded DNA fragments were cloned downstream of the firefly luciferase reporter gene in the pGL4 . 13 vector ( Promega ) using XbaI restriction site . The oligonucleotides used were: 5′-CTAG ( CTG ) 14-3′ and reverse 5′-CTAG ( CAG ) 14 for genes containing ( CUG ) n repeats; forward 5′-CTAG ( CAG ) 14-3′ and reverse 5′-CTAG ( CTG ) 14-3′ for genes containing ( CAG ) n repeats; forward 5′-CTAGTCCGTGCTGAGCCTGCCTGTCGTCTGTG-3′ and reverse 5′- CTAGCACAGACGACAGGCAGGCTCAGCACGGA-3′ mimicking a CUG rich region located in ADORA2A gene; forward 5′-CTAGTGCTAGTATCAGATCTGCTGTGGAATTG-3′ and reverse 5′-CTAGCAATTCCACAGCAGATCTGATACTAGCA-3′for a CTG containing region of MEIS2 gene . HeLa cells were seeded at 1 . 3×104 cells/well in 96-well plates and 24 h later they were co-transfected with the following set of vectors: HTT-e1 constructs ( 40 ng ) , Firefly reporter constructs ( 24 ng ) and Renilla reporter plasmid pGL4 . 75 ( 3 ng ) . The pGL4 . 13 vector without 3′UTR insertion was used as negative control for gene silencing . The ( CAG ) 7 mimic was used as a positive control for silencing effect of CUG enriched stretches . The activity of Firefly and Renilla luciferases was determined 24 h after transfection using the Dual-Glo™ Luciferase Assay System ( Promega ) . Relative reporter activity was obtained by normalization to the Renilla luciferase activity . Each experiment was done in triplicate , and at least three independent experiments were performed for each construct . Hemizygous male mice transgenic for exon 1 of the human Huntingtin gene with a greatly expanded CAG repeat ( R6/2 mice ) [37] were purchased from The Jackson Laboratory ( Bar Harbor , code B6CBA-Tg ( HDexon1 ) 62Gpb/1J; 155–175 CAG repeats ) . The colony was maintained by back-crossing R6/2 males with ( CBA×C57BL/6J ) F1 females . Mice were sacrificed at 8 weeks of age , and brain samples were snap-frozen and subsequently stored at −80°C until use . Those 8 week-old R6/2 mice exhibited various hallmarks of HD-like disease , such as motor symptoms ( deficits in coordination and activity ) , neuropathological deficits ( striatal atrophy and huntingtin-aggregate accumulation ) and molecular-pathology alterations ( down-regulation of striatal-neuron integrity markers ) [38] . Brain samples corresponding to the frontal cortex ( FC ) and the striatum ( dorsal caudate , CA ) of HD patients and controls were obtained from the Institute of Neuropathology and the University of Barcelona Brain Bank . CAG expansions ranged from 41 CAG repeats to 62 CAG repeats in the HD samples ( control samples harbored less than 23 CAG repeats ) . The neuropathological examination in HD cases revealed severe atrophy of the caudate and putamen , cerebral cortical atrophy . This was accompanied by marked neuronal loss and astrocytic gliosis . Individual neurons in the cerebral cortex and striatum exhibited ubiquitin-positive intranuclear inclusions typical of diseases with CAG triplet expansions . HD cases were categorized as stage 4 following Vonsattel classification . Animal handling procedures was conducted in accordance with Directive 86/609/EU of the European Commission . Brain samples of HD patients and controls were obtained from the Institute of Neuropathology and the University of Barcelona Brain Bank , after the informed consent of the patients or their relatives and the approval of the local ethics committee . Ethical issues and legislation as defined by the European Union and national law . All activities were conducted with the approval of responsible ethical committees . The following general guidelines apply:- The Charter of Fundamental Rights of the EU; - Directive 2004/23/EC of the European Parliament and of the Council of 31 March 2004 on setting standards of quality and safety for the donation , procurement , testing , processing , preservation , storage and distribution of human tissues and cells; - Directive 95/46/EC of the European Parliament and of the Council of 24 October 1995 on the protection of individuals with regard to the processing of personal data and on the free movement of such data . In each experiment “n” refers to completely independent experiments . Statistical analyses were performed using the two-tailed unpaired t-student's test for single comparisons ( p<0 , 05 ) and the analysis of variance ( ANOVA ) when multiple pair-wise conditions were compared , where ad-hoc tests were addressed with the Bonferroni correction . The ANOVA test included an interaction term in the cases were the aim was to evaluate whether specific proteins modulate HTT-e1 response . Unless specifically indicated , p-values withstand Bonferroni correction . | Huntington's disease ( HD ) is a neurodegenerative disorder caused by an abnormal CAG expansion in the Huntingtin gene ( HTT ) , resulting in an expanded polyglutamine track in the HTT protein . Longer CAG expansions correlate with an earlier more severe manifestation of the disease that produces choreic movement , behavioural and psychiatric disturbances , and dementia . Although the causative gene is widely expressed , neuropathology is characterized by striatal and cortical atrophy . HTT interacts with proteins involved in transcription , cell signaling , and transport . The pathogenic role of mutant HTT is not fully understood . This study shows that CAG expanded HTT RNA also contributes to neuronal toxicity . Mutant HTT RNA gives rise to small CAG-repeated RNAs ( sCAGs ) with neurotoxic activity . These short RNAs interfere with cell functions by silencing the expression of genes that are fully or partially complementary , through a mechanism similar to that of microRNAs . These findings suggest that a small RNA–dependent mechanism may contribute to HD neuronal cell loss . The exhaustive identification of the target genes modulated by sCAGs may lead to a better understanding of HD pathology , allowing the development of new therapeutic strategies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"rna",
"nucleic",
"acids",
"gene",
"expression",
"biology",
"genomics",
"molecular",
"cell",
"biology",
"molecular",
"biology",
"genetics",
"and",
"genomics"
]
| 2012 | A Pathogenic Mechanism in Huntington's Disease Involves Small CAG-Repeated RNAs with Neurotoxic Activity |
The pathogenesis of Lassa fever ( LF ) , a hemorrhagic fever endemic to West Africa , remains unclear . We previously compared Lassa virus ( LASV ) with its genetically close , but nonpathogenic homolog Mopeia virus ( MOPV ) and demonstrated that the strong activation of antigen-presenting cells ( APC ) , including type I IFN production , observed in response to MOPV probably plays a crucial role in controlling infection . We show here that human macrophages ( MP ) produce large amounts of CC and CXC chemokines in response to MOPV infection , whereas dendritic cells ( DC ) release only moderate amounts of CXC chemokines . However , in the presence of autologous T cells , DCs produced CC and CXC chemokines . Chemokines were produced in response to type I IFN synthesis , as the levels of both mediators were strongly correlated and the neutralization of type I IFN resulted in an inhibition of chemokine production . By contrast , LASV induced only low levels of CXCL-10 and CXCL-11 production . These differences in chemokine production may profoundly affect the generation of virus-specific T-cell responses and may therefore contribute to the difference of pathogenicity between these two viruses . In addition , a recombinant LASV ( rLASV ) harboring the NP-D389A/G392A mutations , which abolish the inhibition of type I IFN response by nucleoprotein ( NP ) , induced the massive synthesis of CC and CXC chemokines in both DC and MP , confirming the crucial role of arenavirus NP in immunosuppression and pathogenicity . Finally , we confirmed , using PBMC samples and lymph nodes obtained from LASV-infected cynomolgus monkeys , that LF was associated with high levels of CXC chemokine mRNA synthesis , suggesting that the very early synthesis of these mediators may be correlated with a favourable outcome .
Lassa virus ( LASV ) is the causal agent of Lassa fever ( LF ) , a hemorrhagic fever endemic to West Africa [1] . The virus is transmitted to humans through contact with infected Mastomys sp . , rodents living close to housing and constituting a natural reservoir of LASV . Human-to-human transmission then occurs through mucosal/cutaneous contact . LF affects about 300 , 000 people each year , resulting in 5 , 000–6 , 000 deaths . There is no approved vaccine against the disease , and the only treatment available , ribavirin , is neither fully effective nor useful in the field , due to its limited availability and the need to initiate treatment soon after infection [2] . LF is therefore a major public health concern in the countries in which it is endemic , and this problem is exacerbated by the tendency of the zone of endemicity to expand [3] . LASV is an Old World arenavirus from the Arenaviridae family . It is an enveloped bisegmented RNA virus . Its small segment ( S ) encodes the nucleoprotein ( NP ) and the glycoprotein precursor ( GPC ) and is cleaved by the subtilase SKI-1/S1P to generate GP1 and GP2 , mediating viral entry by binding to α-dystroglycan [4] , [5] . The large segment ( L ) encodes the RNA-dependent RNA polymerase and the Z protein , a small zinc-binding protein important for replication , transcription and viral budding [6] , [7] , [8] , [9] . The pathogenesis of LF is poorly understood . Antigen-presenting cells ( APC ) , dendritic cells ( DC ) and macrophages ( MP ) are the principal initial targets of LASV [10] , [11] , [12] . The first few cycles of viral replication occur in these cells and the tropism of LASV then widens , such that viral replication also occurs in hepatocytes , endothelial cells , fibroblasts and some epithelial cells [13] , [14] . However , changes to the liver , endothelium and other organs are not severe enough to account for death , which occurs in a context of hypoxic , hypotensive and hypovolemic shock . Little is known about the immune responses associated with survival or death after LF . The production of specific antibodies ( Ab ) is not correlated with survival , as such Ab are detected in all patients , regardless of outcome [15] . Moreover , LASV does not induce the production of neutralizing Ab [16] , [17] . Instead , protection seems to depend on the induction of specific T-cell responses [17] , [18] , [19] . Mopeia virus ( MOPV ) is an Old World arenavirus closely related to LASV . Indeed , MOPV is 60 to 80% identical to LASV in terms of its nucleotide and amino-acid sequences and has also been isolated from Mastomys natalensis [20] , [21] , [22] . However , MOPV is nonpathogenic in non human primates ( NHP ) , and probably also in humans , and it can even immunize monkeys against LASV [23] . We investigated the immune responses associated with LF , by comparing the responses induced by these two viruses in human in vitro models . The infection of DC leads to the release of large amounts of LASV and MOPV , without significant cell activation or cytokine production [10] , [24] , [25] . MP were not activated by LASV infection , but strong activation and significant amounts of type I IFN production were observed in response to MOPV infection . We showed that viral tropism for APC probably played a crucial role in pathogenesis . Indeed , MOPV- , but not LASV- , infected DC induce robust primary human T-cell responses in vitro [26] . The lack of induction of type I IFN production by LASV is due to the presence in the C-terminal part of the NP of a dsRNA-specific 3′ to 5′ exonuclease related to the enzymes of the DEDDh family [27] , [28] . By digesting viral dsRNA , LASV escapes recognition by the RIG-I and MDA-5 helicases , which have been implicated in the sensing of arenavirus RNA and the induction of type I IFN production [29] . The amino-acid residues involved in this activity have been described and their mutation abolishes the anti-IFN properties of NP [30] . The differences in pathogenicity and type I IFN production between LASV and MOPV cannot be due to exonuclease activity alone , as the DEDDh motif is also present in the MOPV NP . The MOPV NP therefore probably also inhibits type I IFN production , albeit less efficiently than the LASV NP . We therefore used reverse genetics to generate a rLASV containing mutations affecting the DEDDh motif . This virus induced a much stronger type I IFN response than MOPV [31] . In addition , it has recently been shown that the arenavirus NP inhibits IRF3 phosphorylation and subsequent type I IFN synthesis by sequestering the IκB kinase-related kinase IKKε in an inactive form [32] . Finally , arenavirus NP has also been reported to prevent the nuclear translocation of NFκB , consistent with the lack of proinflammatory cytokine release observed in response to LASV infection [33] . Inflammatory chemokines are crucial mediators in the development of innate and adaptive immune responses and viral control , but they may also participate in pathogenesis [34] , [35] , [36] , [37] , [38] . The role of chemokines in LF is poorly understood . The release of CXCL10 ( IP-10 ) and IL-8 into plasma has been correlated with survival in LASV-infected patients [39] . Consistent with this finding , LASV-infected human monocytes/MP , unlike their MOPV-infected counterparts , fail to produce IL-8 [40] . In addition , no increase in plasma concentrations of CCL5 ( RANTES ) , CXCL9 ( MIG ) , CXCL-10 , and IL-8 has been detected in moribund LASV-infected cynomolgus macaques; CCL2 ( MCP-1 ) was the only chemokine circulating in significant quantities in these animals [11] . In another study in cynomolgus macaques , increases in the synthesis of CXCL-10 and CXCL11 ( I-TAC ) mRNA were reported in PBMC and lymph nodes , regardless of the outcome of infection [17] . Finally , rhesus macaques intravenously infected with LCMV as a surrogate for LF display high plasma concentrations of IL-8 and CXCL10 , together with strong synthesis of CXCL10 mRNA in PBMC [41] . We compared the production of CC and CXC chemokines by human DC and MP in response to infection with LASV and MOPV , but also in response to infection with LASV-NP D389A/G392A , a rLASV harboring mutations affecting the NP exonuclease , preventing it from inhibiting type I IFN production . We also confirmed that cynomolgus monkeys produced large amounts of CXC chemokines in vivo during LF .
The AV strain of LASV , isolated from the serum of a patient [42] , and MOPV , strain AN 23166 , isolated from Mastomys natalensis [22] , were subjected to four passages on Vero E6 cells at 37°C , under an atmosphere containing 5% CO2 , in Dulbecco's modified Eagle medium supplemented with 50 IU/ml penicillin-streptomycin , 1% non essential amino acids ( all from Invitrogen , Cergy-Pontoise , France ) and AB+ human serum ( Etablissement Français du Sang [EFS] , Lyon , France ) . Cell-free supernatants were harvested after three days for LASV infection and after four days for MOPV infection , and were used as infectious virus stocks , with a viral titer of 2 . 5×107 FFU/ml . The rLASV NP-D389A-G392A and wild-type rLASV were generated by reverse genetics techniques , as previously described [31] , and were passaged twice on Vero E6 cells to obtain the viral stocks ( 1 . 2×107 FFU/ml ) . BSL-4 facilities ( Laboratoire P4-INSERM Jean Mérieux , Lyon ) were used for all experiments with LASV , whereas MOPV was manipulated in BSL-2 facilities . Vero E6 cells and virus stocks were not contaminated with mycoplasma . Monocytes and lymphocytes were isolated from the blood of healthy human donors ( EFS ) , as previously described [10] , [26] . Briefly , peripheral blood mononuclear cells ( PBMC ) were isolated by density gradient centrifugation on Ficoll-Paque ( GE-Healthcare BioSciences AB ) . Autologous plasma ( AP ) was harvested , heated for 30 min at 56°C and centrifuged for T cell experiments . Monocytes were then separated from peripheral blood lymphocytes ( PBL ) by centrifugation on 50% Percoll ( GE-Healthcare ) in PBS and purified by immunomagnetic depletion . PBL were frozen in RPMI 1640-Glutamax I with 50 IU/ml penicillin-streptomycin , 1% non essential amino acids , 10 mM HEPES ( C-RPMI ) supplemented with 20% AP and 10% dimethylsulfoxide ( DMSO ) ( Sigma-Aldrich , Saint-Quentin Fallavier , France ) and 20% AP and stored in liquid nitrogen . Cells were cultured in C-RPMI supplemented with either 10% FCS for APC cultured alone or 10% AP for cultures containing autologous T cells in order to prevent non specific stimulations ( full-RPMI ) . iDC and MP were differentiated from monocytes in full-RPMI supplemented with 2000 IU/ml recombinant human ( rh ) granulocyte-macrophage colony-stimulating factor plus 1000 IU/ml rh interleukin ( IL ) -4 or with 10 ng/ml rh macrophage colony-stimulating factor , respectively ( all from PeproTech , Rocky Hill , NJ , USA ) . Half the cytokine content and 40% of the culture medium were replaced every 48 h and cells were harvested six days later . In experiments using T cells , two vials of DC were frozen in AP containing 10% DMSO for subsequent re-stimulation . DC and MP cell pellets were incubated with virus-free Vero E6 cell supernatant ( mock infection ) or infectious LASV or MOPV at a multiplicity of infection of 2 , for 1 h at 37°C , with regular shaking . DC and MP were then washed and cultured ( 106 cells/ml ) in full-RPMI . In some experiments , the type I IFN receptor was neutralized by adding antagonistic mAbs directed against CD118 , the beta chain of this receptor ( 5 µg/ml , PBL Biomedical Laboratories , Piscataway , NJ ) . We previously described an in vitro model of DC and T-cell co-culture , with three rounds of stimulation [26] . The first round of stimulation involved the infection of DC with infectious LASV or MOPV . DC were then cocultured with autologous T cells at a ratio of 1 DC to 10 T lymphocytes , at a density of 2×105 and 2×106 cells/ml , respectively . PBL were thawed and depleted of B and NK cells as previously described [26] . The second and third rounds of stimulation were carried out 9 and 19 days after the first round of stimulation by culturing thawed iDC stimulated by inactivated viruses or culture medium ( mock ) and T cells harvested from the previous stimulation . We replaced 30% of the culture medium with fresh medium every two or three days . On day 2 of each round of stimulation , 10 IU/ml ( for the first stimulation ) or 5 IU/ml ( second and third stimulations ) of rhIL-2 and rhIL-7 ( both from Peprotech ) were added to the culture medium . In some experiments , the type I IFN receptor was neutralized by adding 5 µg/ml antagonistic mAbs directed against CD118 ( PBL Biomedical Laboratories ) every two days during the first round of stimulation . Control cells were treated similarly , with an irrelevant mouse IgG2a ( R&D Systems , Lille , France ) . Cellular RNAs were isolated from human cells and cynomolgus monkey PBMC or lymph node cells and used for the synthesis of first-strand cDNA as previously described [10] , [17] . In human samples , mRNAs encoding cytokines were quantified by real-time RT-PCR ( RT-qPCR ) with commercially available primers and probes and TaqMan Universal Master Mix on an ABI PRISM 7000 real-time thermocycler and cDNA were amplified in duplex with β-actin or GAPDH ( all from Applied Biosystems ) . Other TaqMan assays were carried out with the following primers and probes: IFNβ: 5′-TCTCCACGACAGCTCTTTCCA-3′ and 5′-ACACTGACAATTGCTGCTTCTTTG-3′ , probe: 5′-AACTTGCTTGGATTCCT-3′ ; IFNα1: 5′-GTGGTGCTCAGCTGCAAGTC-3′ and 5′-TGTGGGTCTCAGGGAGATCAC-3′ , probe: 5′-AGCTGCTCTCTGGGC-3′ ; IFNα2: 5′-CAGTCTAGCAGCATCTGCAACAT-3′ and 5′-GGAGGGCCACCAGTAAAGC-3′ , probe: 5′-ACAATGGCCTTGACCTT-3′ . CCL2 , CCL3 , CCL5 , CXCL9 , CXCL10 , and CXCL11 mRNAs were quantified in lymph node samples by RT-qPCR in a LightCycler 480 thermocycler ( Roche Diagnostic , Meylan , France ) , using TaqMan Universal Master Mix and commercially available primers and probes ( all from Applied Biosystems ) , as well as the following primers and probe for β-actin: 5′-GCGCGGCTACAGCTTCA-3′ and 5′-CTTAATGTCACGCACGATTTCC-3′ , probe: 5′-CACCACGGCCGAGC-3′ . The cDNA obtained from cynomolgus monkey PBMC mRNA was amplified on an ABI Prism 7000 real-time thermocycler ( Applied Biosystems ) with SYBR green PCR master mix ( Applied Biosystems ) and the following primers: β-actin: 5′-TGAACCCCAAGGCCAACC-3′ and 5′-GCCAGCCAGGTCCAGACG-3′; CXCL9: 5′-GGAACCCCAGTAATGAGGAAGG-3′ and 5′-GCAGGAAAGGTTTGGAGCAA-3′ . CXCL10: 5′-TGAAAAAGAAGGGTGAGAAGAGGT-3′ and 5′-TGATGGCCTTAGATTCTGGATTC-3′; CXCL11: 5′-TACGGTTGTTCAAGGTTTCCC-3′ and 5′-TGGAGGCTTTCTCAATATCTGC-3′ . The specificity of the amplicons was checked by determining their melting temperatures . The levels of cytokine mRNA relative to β-actin or GAPDH mRNA levels for each sample ( relative mRNA levels ) were calculated as follows: Δ Cycle threshold ( Ct ) = Ct gene X - Ct β-actin or GAPDH The ratio ( mRNA of interest/β actin or GAPDH mRNA ) = 2 −ΔCt Supernatants from cultures of DC and MP were harvested , centrifuged and stored at −80°C . Commercial ELISA kits were used for CCL2 , CCL3 , CCL4 , CCL5 , CCL7 , CXCL9 , CXCL10 , and CXCL11 detection , according to the manufacturer's instructions ( Bender MedSystems , Vienna , Austria; R&D Systems; BD Biosciences; eBioscience , Paris , France ) . IFNα was assayed with a human-specific ELISA set ( Bender MedSystems ) , in serial plasma samples obtained from LASV-infected cynomolgus monkeys and control animals [17] . Animal experiments were performed in the Jean Mérieux – INSERM BSL4 animal facilities using eight healthy male cynomolgus monkeys ( Macaca fascicularis ) as previously described [17] . All the procedures for animal handling were approved by the “ethics and animal use committee of the Région Rhône-Alpes” ( record N°007 ) and performed in accordance with the regulations of the “European Convention for the Protection of Vertebrate Animals used for Experimental and Other Scientific Purposes” ( ETS n°123 , French decree 2001/131 ) . Briefly , six monkeys were infected by low or high doses of LASV as specified and the other two were mock infected and used as negative controls . PBMC were isolated by density gradient centrifugation from blood or inguinal lymph nodes , washed twice and cultured in full-RPMI . Student's t-test and Mann-Whitney were used to compare dataset means . Differences were considered to be significant if the p value was less than 0 . 05 . Correlations were assessed by Spearman's rank correlation test . All statistical tests were performed with SigmaPlot software ( SyStat Software Inc , San Jose , CA , USA ) .
Our findings confirm previous reports [24] , [25] that MP and , to a lesser extent DC , produce significant amounts of IFNβ , IFNα1 and IFNα2 mRNA in response to MOPV infection . By contrast , LASV-infected DC did not synthesize large amounts of these mRNAs and in LASV-infected MP , only moderate levels are produced ( figure 1A ) . We then quantified the mRNAs encoding CC and CXC chemokines in infected DC and MP . The infection of DC with LASV or MOPV did not lead to synthesis of the CCL2 , 3 , 4 , 5 , 7 , 13 , and 17 mRNAs ( data not shown ) , whereas an increase in the levels of the mRNAs encoding CXCL9 , 10 , and 11 was observed in MOPV-infected DC , but not in LASV-infected DC ( figure 1B ) . The infection of MP with MOPV induced robust expression of the CCL2 , 3 , 4 , 5 , 7 , and 13 genes , and of the CXCL9 , 10 , and 11 genes , in particular ( figure 1C ) . By contrast , the infection of MP with LASV did not result in the transcription of chemokine genes and only CXCL10 and 11 mRNA are significantly produced . Thus , MP and , to a lesser extent , DC , produce type I IFN and chemokines in response to infection with MOPV , but not LASV . The amounts of chemokines released into the culture supernatants were then assessed by ELISA . The results were consistent with mRNA findings , as large amounts of the CC chemokines CCL3 , 4 , 5 , and 7 and , particularly , of the CXC chemokines CXCL9 , 10 , and 11 were observed 24 and/or 72 h after the infection of MP with MOPV ( figure 2A ) . By contrast , MOPV-infected DC produced only moderate amounts ( although significantly higher than those for mock-infected cells ) of CXCL10 ( data not shown and figure 2B ) . LASV-infected DC and MP produced only modest amounts of CXCL10 , and LASV-infected MP produced only small amounts of CCL4 . We recently described an in vitro model of the induction of primary human T-cell responses by LASV- and MOPV-infected DC [26] . Using this model , we demonstrated that MOPV-infected DC induced robust CD4+ and CD8+ T-cell responses involving memory and cytotoxic T cells . By contrast , LASV-infected DC failed to induce significant T-cell responses . We also observed that the presence of T cells in MOPV-infected DC cultures significantly increased the amount of type I IFN produced . We therefore quantified chemokine production in this model . Two days after the first round of stimulation , moderate but significant levels of CCL2 , 3 , 4 , and 5 mRNA were detected in MOPV-infected DC cultured with T cells , but not in LASV-infected DC , other than for CCL5 mRNA synthesis , which was slightly stronger in these cells than in mock-infected cells ( figure 3A ) . CCL7 mRNA levels in MOPV-infected DC were about 10 times higher than those in mock- and LASV-infected DC . Infection with MOPV induced the massive synthesis of CXCL10 and 11 mRNA in DC , whereas only modest levels of CXCL11 mRNA were detected in response to LASV stimulation . By contrast , CXCL9 mRNA levels were not significantly affected by infection with either virus ( data not shown ) . No further significant increase in mRNA synthesis was observed after the second and third rounds of DC stimulation with inactivated viruses ( data not shown ) , other than for CXCL10 mRNA synthesis , which persisted in MOPV-stimulated DC and was significantly induced after the second stimulation with LASV ( figure 3A ) . The levels of CCL4 , CCL5 , CCL7 , CXCL10 , and CXCL11 in supernatants were significantly higher in MOPV-infected DC than in mock- or LASV-infected DC cultured with T cells ( figure 3B ) . In contrast , the increase in CCL2 levels in MOPV-infected DC/T cell cocultures was not significant and no difference was observed for CCL3 release in supernatants between mock- , MOPV- and LASV-infected DC/T cell cocultures ( data not shown ) . Finally , the increase in CCL4 , CCL7 , CXCL10 , and CXCL11 expression was strongly correlated with the intensity of IFNβ , α1 , and α2 mRNA synthesis in LASV- and MOPV-infected DC ( figure 3C ) . For confirmation of the involvement of type I IFN in the observed production of chemokines by MOPV-infected DC cultured in the presence of autologous T cells , we neutralized the type I IFN response by adding a neutralizing antibody against CD118 , the β chain of the type I IFN receptor , during the first round of T-cell stimulation with MOPV-infected DC and analyzing the synthesis of chemokine mRNA after the first and second stimulations . Neutralization of the type I IFN response abolished the synthesis of the CCL4 , CCL7 , CXCL10 , and CXCL11 mRNAs induced by the first round of stimulation with MOPV-infected DC ( figure 4A ) , and that of the CCL2 , CCL3 , and CCL5 mRNAs ( data not shown ) . The increase in the synthesis of the CCL4 , CCL7 , CXCL10 and CXCL11 mRNAs induced by the second round of stimulation with MOPV was not significant , and neutralization of the type I IFN response during the first round of stimulation clearly abolished this overexpression ( figure 4B ) . Similar results were obtained for the CCL2 , CCL3 , and CCL5 mRNAs ( data not shown ) . We recently described a rLASV harboring mutations affecting the exonuclease site of the NP and abolishing the inhibition of the type I IFN response by LASV [31] . We investigated the involvement of the immunosuppressive properties of the LASV NP in the defective production of chemokines by infected APC , by evaluating chemokine production in NP-D389A/G392A ( rNP ) LASV-infected APC and comparing the results obtained with those for recombinant wild-type ( rWT ) LASV . Consistent with the results obtained with LASV ( figure 1B and data not shown ) , no significant increase in the production of CC and CXC chemokine mRNA and proteins was observed in rWT LASV-infected DC ( figure 5A and B ) . By sharp contrast , a robust increase in the amounts of CCL2 , CCL4 , CCL5 , CXCL9 , CXCL10 , CXCL11 mRNAs was detected in rNP LASV-infected DC and these chemokines were significantly released in large amounts in the culture supernatants . Interestingly , mRNA levels were even higher than those obtained in MOPV-infected DC . The amounts of CC and CXC chemokines produced were similar in rWT LASV-infected MP ( figure 5C and D ) and in LASV-infected MP ( figure 1C ) . The infection of MP with rNP LASV led to a robust and significant increase in the synthesis of CCL2 , 4 , 5 , and 7 , and of CXCL9 , 10 , and 11 mRNA and to a strong release of these chemokines in the culture supernatants , as shown by comparison with mock-infected cells . Levels of mRNA and proteins were also significantly higher than those observed in rWT-infected MP , except for CCL2 . We investigated whether chemokines were also produced in vivo during LF , by evaluating the synthesis of mRNAs encoding CC and CXC chemokines in samples obtained from cynomolgus monkeys infected with LASV . We used mRNA extracted from PBMC obtained from serial blood samples collected during the course of infection and from lymph nodes removed nine days after infection , from three monkeys infected with 103 FFU and three monkeys infected with 107 FFU of LASV . As previously reported , two of the three animals infected with low doses died , 16 and 21 days after infection , whereas all the monkeys infected with high doses survived acute LF [17] . We compared the results obtained for both PBMC and lymph nodes with those obtained for these samples from the two mock-infected monkeys . Levels of CCL2 , CCL3 , CCL4 , CCL5 and CCL7 mRNA synthesis in the infected monkeys were no higher than those in the mock-infected monkeys ( data not shown ) . By contrast , increases in the amounts of mRNA encoding CXCL9 , CXCL10 and CXCL11 were detected in PBMC during the course of the disease , and in the lymph nodes of all infected monkeys ( figure 6A and B and [17] ) . CXCL10 and CXCL11 mRNA synthesis in PBMC was most marked six days after infection , in all infected monkeys ( figure 6A and [17] ) . However , three days after infection , these mRNAs were considerably more abundant in survivors than in the monkeys that subsequently died . These outcome-associated differences were no longer evident six days after infection and were not detected at any other point in the course of the disease . Finally , the increase in mRNA levels for CXCL10 and CXCL11 and , to a lesser extent , CXCL9 , in PBMC was strongly correlated with the levels of IFNα release into the plasma of the animals ( as reported in [17] ) during the course of LASV infection ( Figure 6C ) .
Little is known about the pathogenesis of LF and the immune responses that allow a substantial number of patients to recover from acute hemorrhagic fever or even to control viral replication without developing symptoms . We have developed in vitro models for studying the response of human immune cells to infection with the closely related arenaviruses LASV and MOPV , for the identification of host parameters associated with virulence ( LASV ) or an absence of pathogenicity ( MOPV ) [10] , [24] , [25] , [26] , [43] . We have also recently developed reverse genetics techniques for characterizing the role of viral factors in the pathogenicity of LASV [31] . Here , we compare the abilities of LASV and MOPV to induce the production of CC and CXC chemokines in human APC , cultured alone or in the presence of T cells . Our findings provide confirmation in vivo , in NHP , that LF is associated with CXC chemokine release . We found that MOPV induced a massive release of CC and , more particularly , of CXC chemokines in MP . By contrast , significant , but modest , increases in mRNA production in response to infection were observed only for CXCL10 and CXCL11 in LASV-infected MP . This striking difference in the response of MP to infections with LASV and MOPV is consistent with our previous finding that these cells are strongly activated by MOPV , but not by LASV [24] , [25] . The cells potentially targeted by these chemokines express CCR1 , CCR2 , CCR5 and CXCR3; they are thus mostly monocytes , MP , iDC , activated T cells , Th1 cells , B cells , and NK cells [44] . The release of substantial amounts of chemokines by MOPV-infected MP may play a crucial role in the lack of pathogenicity associated with this virus . Indeed , these mediators may act at several levels in the control of viral infection . First , most of the CC chemokines described here are involved in the induction of innate responses through the recruitment of inflammatory cells , mostly iDC and MP , to the site of infection for the establishment of a local inflammatory response [45] , [46] , [47] . CC chemokines , such as CCL2 , 3 , 4 , 5 and 7 , are also involved in attracting naïve and activated CD4+ and CD8+ T cells and bringing them into contact with antigen-presenting DC [34] , [48] , [49] , [50] , [51] . CXCL9 , CXCL10 and CXCL11 are also crucial for the induction of immune responses . Indeed , these CXC chemokines are recognized by CXCR3 , a receptor specifically expressed by activated T lymphocytes and , to a lesser extent , NK cells , but not resting T cells [52] , [53] . CXCL9 , 10 and 11 are therefore major attractors of activated T cells , mostly Th1 and CTL , and they mediate the recruitment of effector T cells and NK cells in secondary lymphoid organs and inflamed tissues [54] , [55] , [56] , [57] . These chemokines have also been reported to stimulate T-lymphocyte proliferation and effector cytokine production [58] . Overall , these results suggest that the production of CXC chemokines may be beneficial , helping to combat viral infection [59] . However , an involvement of these chemokines in immunopathological events has also been reported in some viral infections [38] , [60] , and further investigations are required to clarify the role of CXC chemokines during LF , as we found that LASV also induced the release of significant quantities of CXCL10 and 11 in vitro in MP and in vivo in NHP . The production of these mediators in lymph nodes seemed to be more elevated 9 days after infection in fatally-infected animals than in survivors . Although this difference was noticeable in only one fatality , it would be interesting to investigate whether there are higher amounts of chemokines released in secondary lymphoid organs during severe LF in comparison with non fatal infection and whether these mediators play a deleterious role in this case . The role of chemokines in LF in humans and relevant NHP models remains unclear . Serum concentrations of IL-8 ( CXCL8 ) and CXCL10 have been shown to be higher in patients surviving acute LF than in those who die , whereas CCL5 concentrations are high in both groups [39] , suggesting a possible beneficial effect of CXC chemokines on outcome . Fatal LASV infection in cynomolgus monkeys has been associated with high plasma concentrations of CCL2 and eotaxin in the absence of CCL5 , CXCL9 and CXCL10 release [11] . A transcriptomic analysis on PBMC from cynomolgus monkeys fatally infected with LASV by aerosol recently reported the upregulation of CCL23 , CCRL2 , IL-8 and CXCL12 mRNA synthesis [61] . We did not detect the synthesis of CC chemokines in our cynomolgus monkeys infected with LASV , which is not consistent with the results obtained with MOPV or rLASV NP389A/D392A . This discrepancy is unclear but may be linked to the lack of MP in PBMC and to their limited proportion among splenocytes , as these cells were the main source of CC chemokines in vitro . In any case , the putative protective role of CC chemokines in LF should be interpreted according to this apparent lack of in vivo production . In contrast , we did observe an increase in mRNA levels for CXCL9 , 10 and 11 , in both PBMC and lymph nodes ( this report and [17] ) . No significant difference in mRNA levels during acute disease was detected between lymph nodes ( 9 days after infection ) and PBMC ( from 6 days after infection ) . However , very shortly after infection ( 3 days after infection ) , during the incubation period , CXCL10 and CXCL11 mRNA levels were found to be higher in the PBMC of the monkeys that subsequently survived the acute infection than in the PBMC of monkeys that died . These results have to be confirmed , as the number of animals included here was very low and because the chemokine levels have not been evaluated in lymph nodes at this early time . Nevertheless , these results are consistent with the higher levels of IL-8 ( CXCL8 ) and CXCL10 detected in serum samples from patients surviving acute LF than in those from patients who die [39] , suggesting that CXC chemokines may be beneficial for outcome , particularly during early stages of the disease . However , the evaluation of the overall production of chemokines in patients would be required to determine whether there is a correlation between the severity of LF and the production of chemokines . If it be so , it would be important to understand why different chemokine responses are induced by the same virus in patients . The mechanisms that lead to the diversity of clinical presentation of LF in humans ranging from subclinical infection to catastrophic illness and death , are not clarified . These different outcomes could be due to the type of cells early targeted by the virus , the route of infection , the inoculum dose , preexisting immunity against LASV or heterologous immunity , a different immunological status at the time of infection or a genetic background ( HLA ) . LF was characterized by a massive , generalized infiltration of mononuclear cells , mostly macrophages , into the tissues and organs of cynomolgus monkeys [17] . These cells could potentially be the source of CXC chemokines , attracting activated T cells and NK cells to sites of inflammation . It is possible that the rapidity with which this innate response is established determines whether the disease is fatal or has a favorable outcome . However , the mechanisms leading to a rapid response in some animals and a slower response in others remain unclear , and further investigations of this aspect are required . The production of CCL4 , CCL7 , CXCL10 and CXCL11 in LASV- and MOPV-infected DC cultured in the presence of autologous T cells was strongly correlated with type I IFN synthesis , and neutralization of the type I IFN receptor completely abolished the release of these chemokines . Chemokine production was inhibited during the first coculture in the presence of type I IFN receptor-neutralizing antibodies , but also after the second stimulation of T cells with MOPV-pulsed DC . The lack of type I IFN and/or chemokines during the first round of T-cell stimulation with MOPV-infected DC therefore had major consequences for the priming of T cells , altering their ability to activate iDC during the second round of stimulation . These results suggest that these mediators play a crucial role in the induction of adaptive T-cell responses to MOPV and provide additional evidence in favor of a beneficial role of chemokines during arenavirus infection . The ability of type I IFN to stimulate the production of CXC chemokines , such as CXCL10 and CXCL11 in particular , is well documented [62] . This close link between type I IFN and CXC chemokine production was also observed in vivo , in our primate model . In particular , the early release of IFNα observed specifically in the plasma of the monkeys that survived and not in those that died [17] , is consistent with the increase in CXCL10 and CXCL11 mRNA levels in PBMC . The different production of type I IFN observed between surviving and fatally-infected monkeys despite the presence in animals of a virus presenting similar ability to inhibit this response is unclear . One hypothesis could be that the early type I IFN release observed in surviving primates was due to plasmacytoid DC ( pDC ) , as pDC activation following viral infection is mediated by different TLR than myeloid DC or MP . Indeed , the ability of mouse pDC to bind LASV GP and to release type I IFN after infection with LCMV has been recently reported [63] , suggesting that these cells could play a crucial role in the induction of immune responses during LF . Further investigations are required to confirm this hypothesis in vivo and , if it occurs , to explain the activation of these cells in some animals only . In addition , it would be important to evaluate the ability of pDC to produce chemokines after arenavirus infection to better understand the role of these cells during infection . Nevertheless , these results suggest that the different responses of APC to MOPV and LASV make a major contribution to the difference in pathogenicity between these viruses and to the crucial role of an early activation of innate immunity to the induction of a successful adaptive response and survival in LF . Although the rLASV NP-D389A/G392A is severely attenuated in APC [31] , LASV and MOPV replicate at similar levels in APC [10] , [25] , suggesting that the different chemokine responses are not related to different amounts of viral particles released . Further evidence of the correlation between type I IFN and chemokine responses was provided by the analysis of APC responses to infection with our rLASV NP-D389A/G392A . This LASV bears mutations abolishing the function of the exonuclease domain of the NP; it therefore strongly induces type I IFN production [31] . Consequently , the levels of production of CC and CXC chemokines by MP infected with this virus reached those observed with MOPV . More interestingly , whereas only modest levels of CXC chemokines and no significant synthesis of CC chemokines were induced in MOPV-infected DC , large amounts of mRNAs encoding all these chemokines were observed in response to the rLASV . Despite the presence of the exonuclease site , which is conserved among arenavirus [64] , in MOPV NP , this virus is however able to induce substantial type I IFN responses in APC . Several hypotheses can explain this observation . Other inhibitory properties have been linked to arenavirus NP , such as the ability to sequester IKKε in an inactive form or to inhibit NFκB activation [32] , [33] . It is therefore possible that the efficiency of the exonucleases to digest RNA or of these other inhibitory properties of the NP is different between both viruses . In addition , the role of other viral proteins such as the Z protein in immunosuppression cannot be excluded . These results confirm the central role of NP in the pathogenicity and lack of immunogenicity of LASV , and demonstrate that the ability of LASV NP to inhibit type I IFN induction has profound consequences for the innate , and probably also adaptive , immune responses directed against LASV . In summary , we show here , in both in vitro and in vivo models , that CC and CXC chemokines , including CXCL10 and CXCL11 in particular , are probably key mediators during LF . Our results suggest that the production of these molecules is associated with the lack of pathogenicity of MOPV and with the immunogenicity of LASV NP-D389A/G392A , and that their early release during the incubation period is associated with the control of LASV infection in NHP . Finally , we show that APC must produce type I IFN if they are to release chemokines , and this requirement for type I IFN seems to hold for cynomolgus monkeys . Further investigations in NHP will nevertheless be required , to confirm the protective role of these mediators and to elucidate their mode of action . | Lassa virus ( LASV ) causes a viral hemorrhagic fever that affects about 300 , 000 people and leads to 5 , 000 deaths annually . Lassa fever ( LF ) is a public health problem in West Africa , where it is endemic , because of the number of cases , deaths and disabling effects . There is no vaccine against LASV and the only treatment , ribavirin , is not useful in the field . Little is known about the pathogenesis and immune responses associated with LF . Chemokines are involved in the induction of immunity and attraction of immune cells to inflamed sites . We compared the ability of antigen-presenting cells to produce chemokines in response to infection with LASV , the closely related but nonpathogenic Mopeia virus ( MOPV ) and a LASV unable to inhibit the type I IFN response due to mutations in its nucleoprotein gene . We found that MOPV and the mutant LASV , but not wild-type LASV , strongly induced CC and CXC chemokine production by dendritic cells and macrophages , in a type I IFN-dependent manner . We confirmed in cynomolgus monkeys that these mediators probably play a role during LF . These results highlight the role of innate immunity in LF control and provide insight into the mechanisms leading to survival or death after infection . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"animal",
"models",
"of",
"infection",
"immune",
"cells",
"emerging",
"infectious",
"diseases",
"immunity",
"virology",
"innate",
"immunity",
"emerging",
"viral",
"diseases",
"antigen-presenting",
"cells",
"immunity",
"to",
"infections",
"immunology",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"immune",
"response"
]
| 2014 | Production of CXC and CC Chemokines by Human Antigen-Presenting Cells in Response to Lassa Virus or Closely Related Immunogenic Viruses, and in Cynomolgus Monkeys with Lassa Fever |
Chromosomal rearrangements are a major driver of eukaryotic genome evolution , affecting speciation , pathogenicity and cancer progression . Changes in chromosome structure are often initiated by mis-repair of double-strand breaks in the DNA . Mis-repair is particularly likely when telomeres are lost or when dispersed repeats misalign during crossing-over . Fungi carry highly polymorphic chromosomal complements showing substantial variation in chromosome length and number . The mechanisms driving chromosome polymorphism in fungi are poorly understood . We aimed to identify mechanisms of chromosomal rearrangements in the fungal wheat pathogen Zymoseptoria tritici . We combined population genomic resequencing and chromosomal segment PCR assays with electrophoretic karyotyping and resequencing of parents and offspring from experimental crosses to show that this pathogen harbors a highly diverse complement of accessory chromosomes that exhibits strong global geographic differentiation in numbers and lengths of chromosomes . Homologous chromosomes carried highly differentiated gene contents due to numerous insertions and deletions . The largest accessory chromosome recently doubled in length through insertions totaling 380 kb . Based on comparative genomics , we identified the precise breakpoint locations of these insertions . Nondisjunction during meiosis led to chromosome losses in progeny of three different crosses . We showed that a new accessory chromosome emerged in two viable offspring through a fusion between sister chromatids . Such chromosome fusion is likely to initiate a breakage-fusion-bridge ( BFB ) cycle that can rapidly degenerate chromosomal structure . We suggest that the accessory chromosomes of Z . tritici originated mainly from ancient core chromosomes through a degeneration process that included BFB cycles , nondisjunction and mutational decay of duplicated sequences . The rapidly evolving accessory chromosome complement may serve as a cradle for adaptive evolution in this and other fungal pathogens .
Chromosomal rearrangements are major drivers of genome evolution . Dobzhansky [1] realized that chromosomal polymorphism would “supply the raw materials for evolution” , providing some of the earliest support for Darwin's theory of evolution . Since Dobzhansky's work on Drosophila , cytogenetic studies have revealed a large number of chromosomal rearrangements in the genomes of plant and animal species [2] , including humans [3] . Chromosomal rearrangements were shown to contribute to sex chromosome differentiation [4] , [5] , reproductive isolation [6] , speciation [7]–[10] and complex adaptive phenotypes [11] . Chromosomal rearrangements involve deletions , duplications , inversions and translocations within and among chromosomes . In most cases , the molecular mechanisms that generated the observed rearrangements are not known , but a common explanation is mis-repair of double-stranded DNA breaks [12] , [13] . Repetitive DNA has been strongly associated with chromosome rearrangements in plant and animal genomes and is thought to promote non-allelic homologous recombination during meiosis due to the misalignment of dispersed repeats [14]–[16] . Telomeres play a major role in maintaining chromosome stability [17] , [18] . Although chromosomes lacking a telomere are particularly susceptible to chromosomal fusion , subtelomeric double-strand breaks may also cause chromosomal fusion [19] . McClintock's classic cytogenetic work on maize in the 1930s and 1940s showed that mis-repair of damaged chromosomal ends could generate cycles of chromosomal degeneration termed breakage-fusion-bridge ( BFB ) cycles [20] , [21] . BFB cycles begin when a telomere breaks off a chromosome . When the damaged chromosome replicates , its sister chromatids fuse and form a bridge during anaphase , with the two centromeres of the fused sister chromatids pulled into opposite poles of the dividing cell . After the bridge breaks , the resulting daughter cells receive defective chromosomes that lack telomeres and can initiate new BFB cycles . BFB cycles have also been identified in animals [22] , [23] and yeast [24] , [25] . In humans , BFB cycles play a significant role in cancer progression [18] , [26] , [27] . Fungal chromosomes are generally too small for traditional cytogenetic analyses based on chromosome staining and microscopic examination . But fungi were found to show extensive chromosomal polymorphisms following the invention of pulsed-field gel electrophoresis ( PFGE ) . Application of PFGE revealed that many fungal species exhibit a high variability in chromosome number and size , even among individuals drawn from the same random mating population [28]–[30] . Mechanisms generating the differences in chromosome length and number remained largely elusive , although chromosome breakage and non-allelic homologous recombination among repetitive elements during meiosis were suggested to play a role [28] , [31] . High chromosomal variability in pathogenic fungi may play an important adaptive role [32] . For example , dramatic changes in copy numbers of an arsenite efflux transporter in Cryptococcus neoformans occurred during experimental evolution favoring arsenite tolerance [33] . Chromosomal disomy was associated with increased antifungal drug resistance in several human pathogens including C . neoformans and Candida albicans [34] , [35] . Copy-number variation and aneuploidy were frequently found in clinical and environmental isolates of the same species [36]–[39] . Some of the most polymorphic chromosomal complements were found in plant pathogenic fungi . Several species carry chromosomes that are not shared among all members of the species [40] . Chromosomes exhibiting a presence/absence polymorphism within a species have been referred to as B , dispensable , supernumerary or accessory chromosomes to differentiate them from the “core” chromosomes that are shared among all members of a species [29] , [32] , [40] . We refer to the chromosomes not shared among all individuals as accessory chromosomes because many of these chromosomes play an adaptive role in pathogen evolution , hence these chromosomes are not truly dispensable [32] . Nor do they fit the classic definition of B chromosomes , because they can carry many coding genes and may be necessary for survival in some environments . One of the best studied fungal accessory chromosomes was found in isolates of the pathogen Nectria haematococca and contains a gene cluster important for virulence on peas [41] , [42] . The tomato pathogen Fusarium oxysporum f . sp . lycopersici contains several accessory chromosomes that carry a series of genes important for virulence [43] . In the rice blast fungus Magnaporthe oryzae and related species , a major effector called AVR-Pita that confers virulence on rice was frequently translocated between subtelomeric regions of different chromosomes including accessory chromosomes [44] . Flanking retrotransposons likely contributed to the extreme mobility of the AVR-Pita gene within and among closely related species . The largest known complement of accessory chromosomes is found in the wheat pathogen Zymoseptoria tritici ( syn . Mycosphaerella graminicola [45] ) . The eight smallest chromosomes of the reference genome of Z . tritici , ranging in size from 409–773 kb , were identified as accessory chromosomes [46] . The core chromosomes of the reference genome range in size from 1 , 186–6 , 089 kb [46] . In contrast to accessory chromosomes found in other pathogenic fungi , Z . tritici accessory chromosomes contain over six hundred annotated genes , however the function of these genes is poorly understood [46] . The fungus shows extensive chromosomal length and number polymorphisms within random mating field populations [30] , [47] . Some of the chromosomal diversity appears to be generated through meiosis because progeny populations exhibited frequent chromosome loss and disomy as a result of nondisjunction of accessory chromosomes [48] . The origin of the accessory chromosomes of Z . tritici is not known , though both horizontal chromosome transfer from an unknown donor and degeneration of core chromosomes have been proposed [46] . Comparative genomics of closely related species suggested that several accessory chromosomes originated prior to the emergence of Z . tritici [49] . Several lines of evidence suggest that accessory chromosomes may be important for virulence , including the finding that genes on accessory chromosomes are under accelerated evolution and that these are more likely to show a protein signature consistent with a role in pathogenicity [46] , [49] . The large set of accessory chromosomes in Z . tritici and its close relatives provides a powerful model system to elucidate the mechanisms underlying fungal chromosomal polymorphisms and the origins of accessory chromosomes . We combined population genomic resequencing and PCR-based chromosome segment genotyping to measure the diversity in chromosomal structure at a global scale . We then performed controlled sexual crosses to trace the fate of accessory chromosomes through meiosis and to identify structural rearrangements in chromosomes among the progeny . We confirmed the findings from our resequencing data with electrophoretic karyotyping that enabled chromosomal separation , isolation and visualization by Southern blotting . Our study provides the most comprehensive view to date of mechanisms underlying chromosomal polymorphisms in evolving fungal populations .
Z . tritici is distributed globally and exhibits high genetic diversity for neutral markers [50]–[52] as well as high phenotypic diversity for quantitative traits , including virulence and thermal adaptation [51]–[53] . To assess the composition and frequency of accessory chromosomes across global populations , we designed 57 PCR assays covering all 8 known accessory chromosomes found in the reference strain IPO323 ( chromosomes 14–21; [46] ) . Amplicons ranging in size from 400–600 bp were targeted to coding regions and primer sites were chosen in conserved regions of each gene ( Table S1 ) . The genes comprised in the PCR assay were evenly distributed along the accessory chromosomes and were located mostly in GC-rich regions interspersed by regions of higher repeat content ( Figure 1C and 1D ) . Gene density varies along accessory chromosomes and the PCR assays covered the entire range of known gene locations for each chromosome ( Figure 1E ) . The function of most genes included in the PCR assay is unknown and only 7 out of 57 genes were characterized by gene ontology ( Figure 1F; [46] ) . As a control we designed 15 additional PCR assays covering core chromosomes 10 and 13 . Known microsatellite loci were included in each PCR as a positive control . In total , we surveyed 98 isolates sampled from a global collection of four wheat fields at 72 evenly spaced chromosome positions ( Table S2 ) : Oregon , United States ( n = 19 ) , Israel ( n = 23 ) , Australia ( n = 30 ) and Switzerland ( n = 26 ) . The PCR assays on the core chromosomes 10 and 13 showed that 10 chromosomal segments were present in all 98 isolates ( Figure 1A ) . Three chromosomal segments were missing in 1–3 isolates distributed at random across the populations . One segment on each chromosome was missing in a large fraction of the isolates , but was at approximately the same frequency across all populations ( Figure 1A ) . None of the isolates was missing an entire core chromosome . In contrast , chromosomal segments on accessory chromosomes showed large frequency variations among populations and different accessory chromosomes showed different patterns of segmental presence/absence ( Figure 1A ) . With the exception of chromosome 18 , all accessory chromosomes were found at a frequency higher than 50% in the four field populations . Chromosome 16 was present at the highest frequency with several chromosomal segments being fixed within populations . Individual accessory chromosomes showed substantial differences compared to the chromosomes of the Dutch reference strain IPO323 . Central chromosomal segments located on chromosome 14 were almost entirely missing in isolates from Australia , the United States and Israel . Swiss isolates showed a central chromosomal segment at approximately half the frequency as segments closer to the telomeric ends of the chromosome . The haplotypic diversity for the presence or absence of individual chromosomal segments was substantial among isolates ( Figure S1 ) . Nearly every isolate showed a unique combination of presence or absence of individual accessory chromosome segments . We assessed the population differentiation for presence or absence of chromosomal segments among populations using Wright's FST statistic . Frequencies of several accessory chromosome segments were strongly differentiated among populations . The central segments of chromosome 14 showed FST ranging from 0 . 15–0 . 55 ( Figure 1B ) . High levels of differentiation were also found for the second segment of chromosome 15 and the first segment of chromosome 17 . Chromosome 18 showed elevated levels of differentiation across the chromosome , largely because this chromosome was almost entirely missing from the Australian and USA populations ( Figure 1A ) . In contrast , previous data on neutral genetic markers on core chromosomes showed little differentiation among these and other populations [50] . We found substantial karyotypic diversity in accessory chromosomes among isolates from Switzerland ( Figure 2 ) . In order to obtain a fine-scale map of structural variation in accessory chromosomes among isolates , we performed Illumina resequencing on 9 of the Swiss isolates ( mapping coverage 10–23×; Table 1 ) . We identified genomic divergence between the reference isolate IPO323 and the resequenced isolates by mapping all sequence reads to the finished reference genome . To avoid spurious read mapping in repetitive regions of the chromosomes , we restricted our comparison to the coding regions of the accessory chromosomes . Furthermore , we considered exons of multi-exon genes separately to avoid biases introduced by gene length . In summary , we mapped reads to 1763 exons corresponding to 654 unique genes . The average exon length on accessory chromosomes is 314 bp compared to 517 bp on core chromosomes . The read depth from the resequencing data of 9 Swiss isolates mapped against the reference genome did not suggest any disomic chromosomes ( i . e . doubled read depth for a particular chromosome ) . However , the different isolates varied greatly in gene content on accessory chromosomes . Four isolates ( 3C4 , 3D1 , 3F5 and 1A5; Figure 3 ) showed a nearly complete set of coding sequences compared to the reference genome , with a substantial number of coding sequences present on all 8 accessory chromosomes . Isolate 3D7 contained the smallest complement of accessory chromosome genes , as only four chromosomes showed a substantial proportion of coding sequences to be present . The read mapping to coding regions indicated that accessory chromosomes 14 , 19 and 21 likely differ in length among homologous chromosomes ( Figure 3 ) . Chromosome 16 was found in all isolates except one . However , chromosome 16 likely differs substantially among isolates due to a large number of deletions compared to the chromosome 16 of the reference genome . Nearly all surveyed 20 kb segments along chromosome 16 showed missing genes in at least some of the resequenced isolates . The strongest variation in coding sequence complements was found among variants of chromosome 14 . Isolate 3D1 lacked 149 out of 292 coding sequences , while smaller segments of missing coding sequences were found in six isolates ( 9G4C , 3B8 , 3C4 , 3F5 , 1E4 and 1A5 ) . The number of missing coding sequences ranged from 18–45 among these six isolates . At one end of chromosome 19 , isolate 3B8 showed 46 missing coding sequences out of 220 coding sequences . Similarly , isolates 1A5 and 1E4 showed 31 missing coding sequences out of 155 coding sequences on chromosome 21 . To investigate the nature of large missing chromosomal segments , we performed chromosome-length dotplots of the reference strain chromosome sequence against assemblies of the resequenced isolates . In particular we were interested in whether the large missing segments of chromosome 14 found in isolate 3D1 were due to a single deletion event . The comparison of the reference chromosome 14 of IPO323 with genomic scaffolds of resequenced isolates showed that both the Swiss isolate 3D1 and a previously sequenced Iranian isolate A26b carried one large deletion spanning nearly 400 kb ( Figure 4 ) . In addition , we identified two shorter deletions at homologous locations in both isolates ( at 210–250 kb and 690–720 kb ) compared to the reference chromosome 14 . Interestingly , isolate 9G4C was lacking the large central deletion , however , this isolate shared the two peripheral deletions with isolates 3D1 and A26b ( Figure 4 ) . A fourth isolate ( 1E4 ) shared only the 690–720 kb deletion ( Figure 4 ) . In order to determine the sequence of events leading to the large length polymorphism of chromosome 14 segregating within Z . tritici populations , we performed dotplots with genomic assemblies of three closely related species . We identified significant matching scaffold sequences from isolates of the closest relative Z . pseudotritici spanning the central deletions found in 3D1 and A26b ( Figure 4 ) . In the more ancestral species Z . ardabiliae we did not identify any significant matches for chromosome 14 . However , in the more distantly related species Z . passerinii , a genomic scaffold spanned the entire central region . The deletion matched the regions identified in 3D1 and A26b , as well as Z . pseudotritici ( Figure 4 ) . This suggests that the ancestral chromosome 14 was significantly shorter than the chromosome 14 found in the reference strain IPO323 . Furthermore , this finding indicates that the missing sequences in 3D1 and A26b actually represent large insertions into chromosome 14 of the reference strain . We aimed to ascertain whether the predicted length variants of chromosome 14 are reflected in the karyotypic profiles of the different isolates . For this , we used chromosome-specific probes to identify chromosome 14 in different Z . tritici isolates and Z . passerinii . Hybridization with two chromosome-specific probes ( see Table 2 ) located at opposite ends of the chromosomes showed that the reference isolate IPO323 carried a chromosome 14 in the size range of 780 kb ( Figures 5A and 5B; data shown for probe 2 ) as expected for the isolate [46] . Isolates 3D1 and A26b both carried a substantially shorter chromosome 14 in the range of 400–450 kb , as predicted from the genomic scaffold alignments . The outgroup species Z . passerinii also carried a chromosome 14 that is substantially shorter than in IPO323 ( Figures 5A and 5B ) . Isolate 9G4C was predicted to be of intermediate size between the variants found in IPO323 and 3D1 and A26b . Hybridization with chromosome-specific probes indeed identified a chromosome 14 variant of about 530 kb ( Figures 5A and 5B; data shown for probe 2 ) . To better understand mechanisms leading to the sequence insertions , we identified the precise locations of the breakpoints by performing multiple sequence alignments of the reference chromosome 14 and the scaffold sequences of 3D1 , A26b , 9G4C and Z . passerinii . Interestingly , the four sequence breakpoints characterizing the central section of chromosome 14 are at exactly homologous positions in Z . tritici and Z . passerinii ( Figure 5D ) . The first set of breakpoints is located at 213 , 639 bp and 250 , 917 bp ( breakpoints A and B on Figure 5D ) in the IPO323 genome . The second set of breakpoints is located at 256 , 832 bp and 609 , 754 bp ( breakpoints C and D on Figure 5D ) . We aimed to identify the nature of the novel sequences inserted into chromosome 14 of the reference strain . The overall GC-content of chromosome 14 was 48 . 5% and corresponds to the lowest chromosomal GC content of the Z . tritici reference genome [46] . The two sequences located between breakpoints A-B and E-F showed a consistently lower GC-content than neighboring sequences ( Figure 6 ) . The largest sequence , located between breakpoints C and D , showed a heterogeneous GC-content . The density of repeat sequences increased sharply near the breakpoints of the shorter sequences located between breakpoints A-B and E-F ( Figure 6 ) . Furthermore , no genes were located between breakpoints A-B and only a single gene was found between breakpoints E-F ( Figure 6 ) . The large sequence inserted between breakpoints C-D contained several gene-poor regions . However , the overall gene density of this large sequence is similar to other regions of chromosome 14 ( Figure 6 ) . The large inserted sequence contained 16 genes with predicted functions related to a wide variety of metabolic , signaling and transcription factor activities ( Figure 7B ) . By performing a self-alignment of the reference strain chromosome 14 sequence , we identified a substantial number of repeated sequences distributed along the chromosome . In particular , we found a large palindromic sequence located between 500–550 kb that showed high sequence similarity on both sequence strands ( Figure 5C ) . Chromosome 14 of the reference strain contains a series of transposable element ( TE ) remnants distributed along the chromosome ( Figure 7A ) . Several of the inserted sequences contain TE remnants near the flanking regions . In particular , a non-long terminal repeat ( non-LTR ) element is found near both flanking regions of the insertion between alignment breakpoints E and F . The same element is found at flanking regions of the two other insertions ( alignment breakpoints A and D ) . The large palindromic sequence is flanked by outwards facing LTR Copia element remnants . A major contribution to polymorphisms in accessory chromosomes may arise through meiotic recombination [48] . We performed controlled crosses involving three pairs of isolates from the Swiss population and analyzed 48 progeny from each cross . We applied the same PCR assays targeting 15 chromosomal segments on two core chromosomes and 57 chromosomal segments on the accessory chromosomes . Chromosomal segments on core chromosomes that were missing in either of the two parents were found to be segregating in approximately equal proportions in all three progeny sets ( Figure 8 ) . Patterns of segregation were different for several accessory chromosomes . In Cross 1 ( 9B8B×9G4C ) , we found a loss of chromosome 16 in one offspring despite the fact that both parental isolates were carrying a near full-length chromosome 16 ( Figure 8E ) . In Cross 2 ( 1A5×1E4 ) , we found that 8 progeny were missing all chromosome 14 segments , although both parental isolates carried the corresponding chromosome segments ( Figure 8C ) . Similarly , chromosomes 16 , 18 , 20 and 21 were entirely missing from one offspring though both parents carried these chromosomes . Cross 3 ( 1A5×3D7 ) showed the strongest segregation distortions . Parental isolate 3D7 was missing four accessory chromosomes ( Figure 8A; chromosomes 14 , 15 , 18 and 21 ) . Two of these four chromosomes ( 15 and 21 ) were inherited in significantly higher proportion than expected under random segregation ( X2 test , p<0 . 0007 multiple comparisons corrected , Figure 8B ) . Interestingly , in Cross 1 the parental strains similarly differed in their presence of chromosomes 15 and 21 , however we did not detect any significant segregation distortion in this cross ( Figure 8E ) . Furthermore , two progeny of Cross 3 lost accessory chromosomes 17 , 19 and 20 entirely , although both parental strains carried these chromosomes . We randomly selected 24 and 34 offspring from Cross 1 and Cross 2 , respectively , in order to identify changes in electrophoretic karyotype profiles among progeny . Progeny of both crosses showed substantial karyotypic diversity . Through hybridization with chromosome-specific probes , we found that parental isolates of Cross 2 showed length variation for chromosome 19 of approximately 0 . 3 Mb ( data not shown ) . Chromosomes 15 and 21 showed nearly identical chromosome lengths among the parental isolates . Progeny of Cross 2 segregated the two length variants of chromosome 19 in approximately equal proportions ( data not shown ) . Larger chromosomes ( 1 . 0–3 . 0 Mb ) of parents and progeny of Cross 2 showed similarly diverse electrophoretic karyotypes ( Figure 9A ) . In Cross 2 , we identified two progeny ( A2 . 2 and A66 . 2 ) out of 34 tested with a chromosomal band estimated to be around 0 . 9 Mb . However , neither of the two parents were found to have a chromosomal band in the range of 0 . 7–1 . 2 Mb , as shown by different PFGE gels optimized to separate either the smallest ( <1 . 0 Mb ) or medium-sized chromosomal bands ( 1 . 0–3 . 0 Mb ) ( Figures 9A and B ) . In order to elucidate the origin of the novel chromosome found in two offspring of isolates 1A5 and 1E4 , we performed whole genome resequencing of these progeny . The sequencing reads were mapped to all coding sequences of the reference genome , identically to the procedure used for the resequencing of the Swiss population . The parental isolates 1A5 and 1E4 both carried an almost complete set of accessory chromosomes except that 1E4 lacked chromosome 17 ( Figure 3 ) . Progeny A2 . 2 and A66 . 2 both showed a complete set of accessory chromosomes . However , in contrast to parental isolate 1A5 , we did not find any mapping reads for coding sequences spanning the terminal portion of chromosome 17 ( ranging from 481–558 kb on the reference chromosome 17; Figure 10A ) . This missing chromosome segment would result in a reduced length of approximately 100 kb compared to the length of chromosome 17 in the reference strain ( full length 584 kb; [46] ) . To test for potential duplication events , we used read depth on chromosome 17 as a proxy for duplicated sequences . The parental isolate 1A5 showed a homogeneous distribution of read depth along the chromosome . The parental isolate is suggested to be missing a large chromosomal segment between 1–85 kb compared to the reference genome ( Figure 10B ) . The two progeny A2 . 2 and A66 . 2 also lacked the region between 1–85 kb compared to the reference genome ( Figure 10B ) . The central region of chromosome 17 was divided into two sharply distinct regions based on read depth . A region of high read depth between 85–350 kb and a region of low read depth between 350–481 kb ( Figure 10B ) . We tested whether the increased read depth on chromosome 17 was distinct from the read depth on other chromosomes of the progeny . We calculated the average read density on coding sequences across all 13 core chromosomes as a reference baseline . The average read densities of the parental isolate 1A5 and the two progeny A2 . 2 and A66 . 2 were respectively 11 . 96 , 30 . 93 and 36 . 34 reads per base pair of coding sequence . We compared these average values to read densities on accessory chromosomes ( Figure 10C ) . In order to mitigate biases introduced by large missing segments on the various accessory chromosomes , we calculated the average read density using only mapped positions for each isolate . Accessory chromosomes of the parental isolate 1A5 showed read densities ranging from 69 . 4–94 . 4% of the average read density on core chromosomes , with chromosome 17 showing the lowest read density . Both progeny showed on average slightly higher read densities ranging from 69 . 1–122% and 76 . 1–133% for A2 . 2 and A66 . 2 , respectively . Among all accessory chromosomes , chromosome 17 showed the largest increase ( 1 . 77–1 . 92 fold ) in relative read density compared to the parental isolate 1A5 in both progeny ( Figure 10C ) . We hypothesized that this nearly two-fold increase in read density reflected a large duplication event occurring on chromosome 17 . To determine the genomic content of the novel chromosomal band found in the two offspring , we excised the new chromosomal band found at 0 . 9 Mb from the PFGE gel of progeny A2 . 2 . After purification and whole-genome amplification , we performed Illumina sequencing on the resulting amplified DNA . The sequencing reads were mapped to all coding sequences of the reference genome . The average read density per chromosome was highly variable , with most chromosomes showing an average read density of 0 . 78–17 . 4 reads per base pair ( Figure 10D ) . By far the highest read density was found for coding sequences on chromosome 17 with 175 reads per base pair . We designed two genomic probes specific to chromosome 17 and hybridized the probes to chromosomal bands separated by PFGE . The probes showed that parental isolate 1A5 was carrying a chromosome 17 of the expected length as compared to the reference isolate IPO323 ( Figure 9B ) . We found no hybridization signal on any chromosomal band for parental isolate 1E4 . Both progeny A2 . 2 and A66 . 2 showed a specific hybridization signal for chromosome 17 on the novel chromosomal band at 0 . 9 Mb ( Figures 9B and 9C; probe 4 see Table 2 ) . A second chromosome-specific probe for chromosome 17 gave identical results ( data not shown ) . Taken together , this strongly suggests that the novel chromosome band is either entirely or almost entirely constituted by sequences belonging to chromosome 17 .
The global survey of chromosomal segments revealed highly diverse accessory chromosome complements . We found that isolates not only differed in the number of accessory chromosomes as expected , but that homologous chromosomes showed markedly different gene contents due to numerous insertions and deletions . Several accessory chromosomes such as chromosome 16 were found near fixation in some populations , such as Australia and Israel . In contrast we found that chromosome 18 was almost entirely missing from the sampled Australian population . The near fixation or losses of accessory chromosomes in some populations may be due to stochastic processes such as founder events during the establishment of the pathogen in previously unaffected geographical regions . Populations also differed strongly in the diversity of chromosomal haplotypes detected by the PCR assays . The Swiss population had a much higher number of unique haplotypes for chromosomes 14 , 16 and 18 than the Australian population . Founder effects were hypothesized to explain the low genetic diversity found for neutral markers in Australian Z . tritici populations that were introduced along with wheat not later than ∼200 years ago [50] . In agreement with this earlier finding , accessory chromosomal segments of the Australian population showed the strongest deviation from global frequencies . However , large variations in accessory chromosome complements were also found in other populations . Hence , the diversity in chromosomal complements reflects a previously uncharacterized form of genetic differentiation in this pathogen . Frequency differences in accessory chromosomes among populations may also result from selection operating on chromosomes carrying genes that confer a selective advantage or disadvantage in particular environments . For example , gene products such as effectors that contribute to host virulence in a gene-for-gene interaction may be strongly disfavored in some wheat fields due to the presence of matching resistance genes [54] . If virulence factors such as effectors are located on accessory chromosomes , this may enable rapid adaptation in an arms race to overcome detection by the host immune system . The rapid loss of non-essential virulence factors located on accessory chromosomes may provide a significant selective advantage to a fungal pathogen [32] . The resequencing of Swiss isolates revealed extensive variation in gene content among homologous accessory chromosomes . In comparison to the chromosome sequence of the reference strain , accessory chromosomes of the resequenced isolates carried deletions ranging from a few genes to large sections affecting several dozens of genes . Surprisingly , missing segments were rarely contiguous as would be expected from single deletion and insertion events generating a chromosomal length polymorphism . Accessory chromosome 16 showed numerous short deletions spanning only a few coding sequences in the Swiss population compared to the reference chromosome . Our resequencing analysis ( Figure 3 ) suggests that several chromosomes may be missing chromosomal ends including telomeres . However , our resequencing data was not informative on the integrity of telomeric repeats and we could not be certain that telomeres were missing in these isolates . In the reference genome , one telomere sequence on chromosome 21 could not be sequenced and may be missing [46] . Intact telomeres play a crucial role in chromosomal stability by ensuring homologous chromosomal pairing and disjunction during meiosis [17] . Defective telomeres are thought to initiate the development of breakage-fusion-bridge cycles leading to major chromosomal anomalies [17] , [18] . If some accessory chromosomes are indeed defective for telomeres , this may play a major role in generating the observed chromosome polymorphisms . The most dramatic chromosomal length polymorphism segregating within a population was found for chromosome 14 , with the shortest identified chromosome variant approximately half the length of the longest known chromosome variant . In a related pathogen found on barley ( Z . passerinii ) that is ancestral to the more closely related pathogens found on wild grasses [55] we identified a homologous chromosome . The ancestral form of chromosome 14 is largely identical to the shortest variant found in Iranian and Swiss populations . Several lines of evidence suggest that the large insertions leading to the chromosome 14 variant found in the reference strain occurred recently . First , the longest chromosome variant is found almost exclusively in the Swiss population , which is closest to the location where the Dutch reference isolate IPO323 was isolated . The A26b isolate from Iran was sampled close to the center of origin of Z . tritici and this isolate carried the shortest known chromosomal variant . Second , sequences immediately adjacent to the insertion breakpoint locations showed only a single nucleotide polymorphism compared to the reference chromosome . Third , sequences near the breakpoint location were highly similar even when compared with the phylogenetically distant Z . passerinii . A major open question is the source of the inserted sequences . We did not find closely related sequences either at a different location in the reference genome or in any resequenced strain . The largest insertion in chromosome 14 contains several dozen genes and may have functional consequences for isolates carrying the large chromosome variant , because several of these genes were predicted to encode transcription factors or other functions . All three inserted sequences are flanked on at least one end by remnants of the same class of transposable elements . The presence of these elements near the flanking regions suggests that non-allelic homologous recombination with an unknown chromosome may have played a role in the insertion of these sequences into chromosome 14 . Interestingly , the two shorter insertions in chromosome 14 showed a markedly lower GC-content than surrounding regions and these inserts were virtually devoid of genes . These isochores may be regions of reduced recombination , as the inserted regions may lack homologous sequences necessary for meiotic crossing-over . The largest inserted sequence also contains a very large palindromic sequence similar in extent to palindromes on the human Y chromosome [56] , [57] . The palindrome is flanked by the remnants of two copies of a transposable element , similar to the inserted sequences . In yeast , palindromes were shown to mediate gene amplification and intra-chromosomal recombination and may lead to genomic instability [58] , [59] . Goodwin et al . [46] hypothesized that accessory chromosomes originated through an ancient horizontal transfer from an unknown donor species . Our analyses show that the accessory chromosome 14 was maintained through multiple speciation events and hence may be a remnant of an ancient core chromosome . The large insertions observed in extant Z . tritici populations suggest that chromosome 14 is undergoing a degeneration process . The insertions do not seem to have a severe effect on the fitness of the organism , as many different length variants of chromosome 14 were found segregating within the Swiss population . The tolerance to large sequence rearrangements may be a hallmark of the degeneration process affecting accessory chromosomes . Meiosis is thought to play a major role in genomic instability in fungi [28] , [48] , [60] . Non-allelic homologous recombination among dispersed repeats was hypothesized to be the main source of chromosomal length polymorphism in fungi [28] , [60] . In Z . tritici , aberrations during meiosis were suggested to lead to the loss of accessory chromosomes and hence contribute to chromosomal number polymorphism among isolates [48] . Our analyses of progeny from three different crosses showed that chromosomal loss affected nearly all accessory chromosomes . We detected low levels of chromosomal losses for all accessory chromosomes except chromosome 15 . The loss of a chromosome may be due to nondisjunction of sister chromatids during meiosis . This defect during meiosis would create progeny carrying a duplicated ( i . e . disomic ) chromosome . Frequent loss of accessory chromosomes during meiosis poses an apparent paradox in Z . tritici . Populations would be expected to gradually lose all accessory chromosomes over generations in the absence of mechanisms to maintain the accessory chromosomes . Interestingly , our data on inheritance of accessory chromosomes revealed a mechanism that may maintain accessory chromosome complements in populations . Analyses of segregation frequencies revealed that chromosomes 15 and 21 , if present in only one of the two parental strains , were inherited significantly more frequently than expected under random segregation of the chromosomes . Distorted segregation was restricted to one cross and no distortion was detected in the second cross differing in the presence of chromosomes 15 and 21 . Segregation distortion was found to be a key characteristic of numerous animal B chromosomes [61] . We hypothesize that segregation distortion is one of the mechanisms that maintains accessory chromosomes in Z . tritici populations . The most striking example of chromosomal plasticity was the fusion of sister chromatids to generate a much longer chromosome 17 in two progeny of Cross 2 ( Figure 10 ) . This meiotic abnormality occurred in a cross between a strain carrying a chromosome 17 similar to the reference isolate and a strain lacking the entire chromosome . One mechanism for creating the new chromosome is non-allelic homologous recombination between inverted repeats on sister chromatids ( [62] , Figure 10E ) . The recombination event could create an isodicentric chromosome 17 carrying duplicated and non-duplicated regions , consistent with the striking difference in read depth observed across chromosome 17 ( Figure 10B ) . If the fused chromosome contains two centromeres , it is expected to form a bridge at anaphase and undergo BFB cycles [20] , [21] , [26] . The rejoining of broken ends during new rounds of cell division will create new chromosomal arrangements including deletions and duplications . The lack of a homologous chromosome 17 during meiosis may have contributed to the initiation of a BFB cycle . The fate of the novel chromosome over subsequent generations is currently under investigation . The meiotic pairing of the large duplicated chromosome 17 with the parental chromosome variant is likely to generate further highly unstable chromosomal variants . Z . tritici possesses a genomic defense mechanism known as RIP [46] that is common to a large number of ascomycete fungi [63] . RIP rapidly degenerates highly similar genomic regions through the introduction of point mutations . We predict that the novel chromosome 17 variant generated by duplicating a large fraction of the original chromosome would be subjected to rapid degeneration as a result of RIP . We propose that BFB cycles coupled with RIP played a major role in creating the degenerated accessory chromosomes of Z . tritici . Our study revealed extensive yet viable chromosomal rearrangements generated by meiosis . Genomic instability and insertion of exogenous sequences led to highly diversified sets of homologous chromosomes affecting hundreds of genes . The large number of insertions and deletions found among accessory chromosomes suggests that these chromosomes underwent an extensive degeneration process . The chromosomal degeneration process may well have been initiated in an ancestor of Z . tritici . The shorter gene length and lower gene density on accessory chromosomes compared to core chromosomes suggests that degeneration processes affected accessory chromosomes over long evolutionary time scales . We identified large insertions and the initiation of breakage-fusion-bridge cycles as two major contributors to chromosomal abnormalities . Surprisingly , isolates of Z . tritici appear to be highly tolerant of these abnormalities , which may contribute to the maintenance of extensive karyotypic diversity in populations . The extensive degeneration , distorted segregation and frequent loss of accessory chromosomes highlight a central question surrounding fungal accessory chromosomes: How and when do these chromosomes originate ? We showed that chromosome 14 is ancient , as its origin predates several speciation events prior to the emergence of Z . tritici in the Fertile Crescent [55] . We postulate that the accessory chromosomes found in extant Z . tritici populations likely originated from the core chromosomes through a degeneration process . The initiation of chromosome degeneration is particularly likely in isolates that carry disomic chromosomes due to nondisjunction . Disomy would provide redundancy in gene content and , hence , relax selection pressure to maintain chromosomal integrity . Chromosomal degeneration may then proceed rapidly through BFB cycles , nondisjunction and RIP of duplicated regions . The emergence of a highly diverse and rapidly evolving set of accessory chromosomes in Z . tritici illustrates how an accessory genome can be created to serve as a cradle for adaptive evolution in this and other fungal pathogens .
We assessed the diversity in chromosomal structure in a global collection of Z . tritici . We included field populations from Israel ( n = 23 ) , Oregon , USA ( n = 19 ) , Switzerland ( n = 26 ) and Australia ( n = 30 ) ( Table S2 ) . These populations were previously assayed for neutral genetic diversity and variation in quantitative traits [53] . These isolates showed substantial variation for several quantitative characters , including virulence , fungicide resistance and thermal adaptation , among and within populations [51]–[53] , [64] . Sexual crosses were performed between three pairs of isolates from the Swiss population ( see Table S2 ) using the established protocol for Z . tritici [65] . The crosses were between isolates ST99CH9B8B and ST99CH9G4C ( Cross 1 ) , ST99CH1A5 and ST99CH1E4 ( Cross 2 ) and ST99CH1A5 and ST99CH3D7 ( Cross 3 ) . In order to survey presence-absence polymorphism among accessory chromosomes , we designed PCR assays to amplify approximately 500 bp of coding sequences at regular intervals of approximately 100 kb along the chromosomes of reference strain IPO323 . For detailed information on the targeted genes and chromosomal locations see Table S1 . Primers for PCR amplification were designed on conserved sections of the targeted coding sequence . Sequence conservation was assessed using the reference assembly of nine resequenced Swiss isolates and two resequenced Iranian isolates ( for details see below ) . We used Primer 3 . 0 for primer design [66] . In order to control for successful PCR , we included a primer pair of a microsatellite locus in each PCR mix [67] . Successful PCRs produced a band at approximately 250 bp that was clearly distinguishable from the PCR product associated with each chromosomal segment . PCR reactions were performed in 20 µl volumes containing approximately 5–10 ng genomic DNA , 0 . 5 µM of each primer , 0 . 25 mM dNTP , 0 . 6 U Taq polymerase ( DreamTaq , Thermo Fisher , Inc . ) and the corresponding PCR buffer . PCR products were visualized on agarose gels . We used the R graphics package ggplot2 to plot the raw datasets and analyses [68] , [69] . Measures of genetic differentiation among populations ( FST ) were calculated with the function var . comp in the R package hierfstat [70] . The presence-absence data generated by the PCR assays were considered as two possible alleles at haploid loci . We tested for segregation distortion of chromosomal segments among progeny by testing for deviations from the expected 1∶1 segregation ratio of presence-absence polymorphism among progeny with a χ2 contingency table . We accounted for non-independence of chromosomal segments and multiple testing with a conservative Bonferroni correction . We calculated the repeat content on accessory chromosomes by identifying direct repeats with a repeat motif between 2–50 bp [71] . For each repeat , we calculated the total length of the repeat and subtracted the number of mismatches in the repeat motif , as a proxy for the extent and purity of the repeat element . We used the previously published genome assemblies of two Iranian isolates of Z . tritici ST01IRA26b and ST01IRA48b . In addition , we included five genomes of Z . pseudotritici ( STIR04_3 . 11 . 1 , STIR04_2 . 2 . 1 , STIR04_4 . 3 . 1 , STIR04_5 . 3 , STIR04_5 . 9 . 1 ) four genomes of Z . ardabiliae ( STIR04_3 . 3 . 2 , STIR04_3 . 13 . 1 , STIR04_1 . 1 . 1 , STIR04_1 . 1 . 2 ) and one genome of the outgroup species Z . passerinii ( P63 ) [72] , [73] . All genome assemblies are available under the NCBI BioProject [PRJNA63131] on GenBank . We resequenced nine Z . tritici isolates from Switzerland ( full isolate names: ST99CH1A5 , ST99CH1E4 , ST99CH3B8 , ST99CH3C4 , ST99CH3D1 , ST99CH3D7 , ST99CH3F5 , ST99CH9B8B and ST99CH9G4C ) and two progeny from Cross 2 ( A2 . 2 and A66 . 2 ) . We performed Illumina paired-end sequencing on 500–700 bp insert libraries to generate between 1–2 Gb of quality-trimmed sequence data per isolate ( theoretical coverage of 25–50× ) . The read length was either 82 bp or 90 bp . Illumina sequence data are available from the NCBI Short Read Archive ( see Table 1 for accessions ) . We used SOAPdenovo v . 1 . 5 [74] to generate de novo assemblies , including scaffolding and gap closing . De novo assemblies yielded a scaffold N50 ranging from 79 , 920–121 , 161 bp depending on the resequenced isolate . Total assembly space ( scaffolds and singletons ) ranged from 35 . 57–38 . 33 Mb ( see Table 1 ) . All genome assemblies are available on GenBank under BioProject [PRJNA178194] ( see Table 1 ) . The comparison with the total finished genome size for the reference isolate IPO323 ( 39 . 7 Mb ) shows that the genomic assemblies account for a very large proportion of the genome of the sequenced isolates . The assembly statistics were similar to the assemblies reported earlier for the same species [72] . We mapped the Illumina reads of each resequenced isolate and offspring to the finished genome of IPO323 [46] . We used Bowtie 2 . 1 . 0 [75] to perform the mapping , allowing only reads that were mapped as paired-ends . We assessed the read coverage on the reference genome by filtering all reads based on their mapping quality ( minimum mapping quality of 20 ) with GATK version 2 . 3-9-ge5ebf34 [76] . Coverage of coding sequences was extracted using the BEDtools utilities [77] . We scored the absence of coding sequences conservatively , requiring that less than 10 bp of a coding sequence should be covered and that the average read density on the coding sequence would be below 2× . Structural changes among chromosomes of different isolates were analyzed using Nucmer [78] . We used the –mum option requiring unique anchor matches that are unique in both the query and the reference genome . Genome assemblies were compared in pairwise comparisons between the finished reference genome of IPO323 and the draft assemblies of the different isolates of Z . tritici , Z . pseudotritici , Z . ardabiliae and Z . passerinii . In order to visualize synteny among different variants of chromosome 14 , we extracted all scaffolds matching the reference chromosome 14 . We discarded scaffolds that were shorter than 10 kb and that showed a match identity with the reference chromosome of less than 80% . Scaffold alignments were plotted with the R package ggplot2 [69] . Repetitive and palindromic sequences of the reference chromosome 14 of IPO323 were visualized by performing a self-alignment with LASTZ ( http://www . bx . psu . edu/~rsharris/lastz ) . The finished chromosome 14 sequences were analyzed for short and medium length tandem repeats with the software Tandem Repeat Finder v . 4 . 04 [79] . We set the matching weight to 2 , the mismatching and indel penalty to 10 and the match and indel probability to 80 and 10 , respectively . The minimum alignment score was required to be 10 and the maximum period size of repeats was set to 50 bp . The occurrence of repeats was visualized along a 5 kb sliding window ( with increments of 1 kb ) . The gene density on each chromosome was reported as the occurrence of start codons according to the latest annotation [46] . GC content of each chromosome was reported in 5 kb sliding windows with increments of 1 kb . We identified transposable element remnants on chromosome 14 by querying the annotated repeat libraries provided by Repbase Update [80] . High molecular weight chromosomal DNA ( Ch-DNA ) was prepared by in situ digestion of cell walls of agarose-embedded conidia . We used a slightly modified non-protoplasting method according to McCluskey et al . [81] . The following Z . tritici isolates were used: ST01IRA26b , ST99CH9B8B ( parental isolate of Cross 1 ) , ST99CH9G4C ( parental isolate of Cross 1 ) , ST99CH1A5 ( parental isolate of Cross 2 and 3 ) , ST99CH1E4 ( parental isolate of Cross 2 ) , ST99CH3B8 , ST99CH3C4 , ST99CH3D1 , ST99CH3D7 , ST99CH3F5 and IPO323 . In addition , we included the isolate P63 of Z . passerinii [55] . To screen progeny of sexual crosses , we randomly selected 24 and 34 confirmed progeny from Cross 1 and Cross 2 , respectively . All isolates were transferred from stocks maintained in glycerol at −80°C to Yeast Malt Agar ( YMA ) plates and were grown for 3 to 4 days in the dark at 18°C . After incubation , conidia were washed off the plates with sterile water and 600–800 µl of suspended conidia were transferred to 2 to 3 fresh YMA plates . The plates were incubated for 2 to 3 days as described above . Conidia were harvested using sterile distilled water and filtered through sterile Miracloth ( Calbiochem , La Jolla CA , USA ) into 50 ml screw-cap Falcon tubes . The tubes were filled with distilled water up to 50 ml total volume . The suspension was centrifuged at 3750 rpm at room temperature for 15 min with a clinical centrifuge ( Allegra X-12R , Beckman Coulter , Brea CA , USA ) . The resulting pellets were resuspended in 1–3 ml TE buffer ( 10 mM Tris-HCL , pH 7 . 5; 1 mM EDTA , pH 8 . 0 ) and gently vortexed . The spore concentration of the solution was determined using a Thoma haematocytometer cell counter . An aliquot of 1 . 5 ml spore suspension with a concentration between 8×107 to 2×108 spores/ml was transferred to a fresh 50 ml screw-cap tube and incubated at 55°C in a water bath for several minutes . To each tube , 1 . 5 ml pre-warmed ( 55°C ) low-melting-point agarose prepared in TE Buffer was added ( 2% w/v; molecular biology grade , Biofinex , Switzerland ) . The solution was thoroughly mixed by gentle pipetting . An aliquot of 500 µl was solidified on ice for approximately 10 min in a precooled plug casting mold ( BioRad Laboratories , Switzerland ) . A total of five agarose plugs per isolate were incubated in 15 ml screw-top tubes containing 5 ml of a lysing solution containing 0 . 25 M EDTA , pH 8 . 0 , 1 . 5 mg/mL protease XIV ( Sigma , St . Louis MO , USA ) , 1 . 0% sodium dodecyl sulfate ( Fluka , Switzerland ) . The incubation was performed for 28 h at 55°C . During the incubation the lysing solution was changed once after 18 h and gently mixed every 2–3 h . Chromosomal plugs were washed three times for 15–20 min in 5–6 ml of a 0 . 1 M EDTA ( pH 9 . 0 ) solution and then stored in the same solution at 4°C until they were used . Pulsed-field gel electrophoresis ( PFGE ) was carried out using a BioRad CHEF II apparatus ( BioRad Laboratories , Hercules CA , USA ) . Chromosomal plugs were inserted into the wells of a 1 . 2% and 1 . 0% ( wt/vol ) agarose gel ( Invitrogen , Switzerland ) to separate small chromosome ( <1 Mb ) and medium-sized chromosomes ( 1 . 0 Mb–3 . 0 Mb ) , respectively . Small chromosomes ( i . e . accessory chromosomes ) were separated at 13°C in 0 . 5× Tris-borate-EDTA Buffer ( Sambrook & Russell 2001 ) at 200 V with a 60–120 s pulse time gradient for 24–26 h . Medium-sized chromosomes were separated at 100 V with a 250–900 s pulse time gradient for 48–50 h using the same buffer and running temperature as above . Gels were stained in ethidium bromide ( 0 . 5 µg/ml ) for 30 min immediately after the run . Destaining was performed in water for 5–10 min . Photographs were taken under ultraviolet light with a Molecular Imager ( Gel Doc XR+ , BioRad , Switzerland ) . As size standards , we used chromosome preparations of Saccharomyces cerevisiae ( BioRad , Switzerland ) and Hansenula wingei ( BioRad , Switzerland ) . Southern blotting and hybridization were performed according to standard protocols [82] . In summary , hydrolysis was performed in 0 . 25 M HCl for 30 min and DNA was blotted onto Amersham HybondTM-N+ membranes ( GE Healthcare , Switzerland ) overnight under alkaline conditions [82] . DNA was fixed onto the membranes at 80°C for 2 h . Membranes were prehybridized overnight with 25 ml of a buffer containing 20% ( w/v ) SDS , 10% BSA , 0 . 5 M EDTA ( pH 8 . 0 ) , 1 M sodium phosphate ( pH 7 . 2 ) and 0 . 5 ml of sonicated fish sperm solution ( Roche Diagnostics , Switzerland ) . Probes were labeled with 32P by nick translation ( New England Biolabs , Inc . ) following the manufacturer's instructions . Hybridization was performed overnight at 65°C . Blots were subjected to stringent wash conditions with a first wash in 1× SSC and 0 . 1% SDS and a second wash with 0 . 2× SSC and 0 . 1% SDS . Both washes were performed at 60°C . Membranes were exposed to X-ray film ( Kodak BioMax MS ) for 2 to 3 days at −80°C . All hybridization probes used to identify specific chromosomes are listed in Table 2 . Chromosomal DNA was separated with CHEF gel electrophoresis as previously described for the separation of small chromosomes except that a 1 . 0% agarose gel was used . The novel 0 . 9 Mb chromosomal band from isolate A2 . 2 was excised and DNA was recovered using the Wizard SV Gel and PCR Clean-up System kit ( Promega , Switzerland ) with the following modifications to the manufacturer's recommendations: during the incubation at 65°C the gel slice was vortexed two times for 5 minutes , sonication was for 3 min and followed by a final incubation for 1 min . The resulting purified DNA was amplified using a whole genome amplification kit ( REPLI-g Mini Kit , Qiagen , Germany ) . Amplified DNA was subjected to whole genome sequencing with an Illumina HiSeq 2000 as described above . | Chromosomal rearrangements are a hallmark of genetic differences between species . But changes in chromosome structure can also occur spontaneously within species , within populations , or even within individuals . The causes and consequences of chromosomal rearrangements affecting natural populations are poorly understood . We investigated a class of fungal chromosomes called accessory chromosomes that are not shared among all individuals within a species . Using a fungal pathogen possessing numerous accessory chromosomes as a model , we assessed chromosome diversity based on whole-genome sequencing and a PCR assay of chromosomal segments that included a global collection of isolates . We show that the accessory chromosomes are highly variable in their gene content and that geographic differences correlate with the number and the structure of the chromosomes . We applied the same approach to document chromosomal rearrangements occurring during sexual reproduction . We identified viable offspring carrying a novel chromosome that originated from a large duplication affecting the majority of the chromosome . Our study showed that chromosomal structure can evolve rapidly within a species to generate a highly diverse set of accessory chromosomes . This chromosomal diversity may contribute significantly to the adaptive potential of fungal pathogens . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genome",
"sequencing",
"mycology",
"plant",
"science",
"genome",
"evolution",
"chromosome",
"biology",
"fungal",
"evolution",
"plant",
"pathogens",
"microbial",
"pathogens",
"plant",
"pathology",
"population",
"genetics",
"comparative",
"genomics",
"biology",
"genomics",
"evolutionary",
"biology",
"genomic",
"evolution",
"microbiology",
"computational",
"biology"
]
| 2013 | Breakage-fusion-bridge Cycles and Large Insertions Contribute to the Rapid Evolution of Accessory Chromosomes in a Fungal Pathogen |
Inflammation , which is directly regulated by interleukin-6 ( IL-6 ) signaling , is implicated in the etiology of several chronic diseases . Although a common , non-synonymous variant in the IL-6 receptor gene ( IL6R Asp358Ala; rs2228145 A>C ) is associated with the risk of several common diseases , with the 358Ala allele conferring protection from coronary heart disease ( CHD ) , rheumatoid arthritis ( RA ) , atrial fibrillation ( AF ) , abdominal aortic aneurysm ( AAA ) , and increased susceptibility to asthma , the variant's effect on IL-6 signaling is not known . Here we provide evidence for the association of this non-synonymous variant with the risk of type 1 diabetes ( T1D ) in two independent populations and confirm that rs2228145 is the major determinant of the concentration of circulating soluble IL-6R ( sIL-6R ) levels ( 34 . 6% increase in sIL-6R per copy of the minor allele 358Ala; rs2228145 [C] ) . To further investigate the molecular mechanism of this variant , we analyzed expression of IL-6R in peripheral blood mononuclear cells ( PBMCs ) in 128 volunteers from the Cambridge BioResource . We demonstrate that , although 358Ala increases transcription of the soluble IL6R isoform ( P = 8 . 3×10−22 ) and not the membrane-bound isoform , 358Ala reduces surface expression of IL-6R on CD4+ T cells and monocytes ( up to 28% reduction per allele; P≤5 . 6×10−22 ) . Importantly , reduced expression of membrane-bound IL-6R resulted in impaired IL-6 responsiveness , as measured by decreased phosphorylation of the transcription factors STAT3 and STAT1 following stimulation with IL-6 ( P≤5 . 2×10−7 ) . Our findings elucidate the regulation of IL-6 signaling by IL-6R , which is causally relevant to several complex diseases , identify mechanisms for new approaches to target the IL-6/IL-6R axis , and anticipate differences in treatment response to IL-6 therapies based on this common IL6R variant .
Originally identified as a B-cell differentiation factor , interleukin-6 ( IL-6 ) is now recognized as one of the most pleiotropic cytokines in humans . IL-6 can activate a wide-range of cell types and is recognized as a critical regulator of acute inflammatory reactions [1] . In addition , IL-6 plays a key role in controlling the activation and differentiation of T-cell responses , promoting a pro-inflammatory environment , which has been associated with the pathogenesis of several autoimmune and inflammatory diseases in humans [2] . Binding of IL-6 to the membrane-bound IL-6 receptor ( IL-6R ) induces homodimerization with its co-receptor gp130 , resulting in the phosphorylation of the transcription factors STAT3 and STAT1 ( classical signaling ) [3] . Alternatively , a circulating soluble form of IL-6R ( sIL-6R ) , if bound to IL-6 , is able to stimulate cells expressing gp130 ( “trans-signaling” ) , even in the absence of membrane-bound IL-6R [4] . The role of genetic variation in IL6R in the etiology of human disease has been highlighted by genetic studies reporting the association of variants in the gene with the risk of several diseases with an inflammatory component , including coronary heart disease ( CHD ) [5]–[7] , rheumatoid arthritis ( RA ) [8] , atrial fibrillation ( AF ) [9] , abdominal aortic aneurysm ( AAA ) [10] and asthma [11] . A common ( MAF 30–40% in European and Asian HapMap populations ) non-synonymous variant Asp358Ala in IL6R ( rs2228145 A>C , previously rs8192284 ) has been suggested to be the causal variant at this locus , because of its strong correlation with circulating concentrations of sIL-6R [12] . However , the effect of this variant on classical IL-6R signaling remains unclear . Here we demonstrate that the 358Ala allele regulates IL-6R surface expression at the protein level in specific immune cell subsets , resulting in altered IL-6 signaling . These findings clearly demonstrate the effect of Asp358Ala in the regulation of classical IL-6 signaling and provide further insight into the functional mechanism underpinning the association of this genetic variant with human diseases .
Given the association of rs2228145 with a variety of human inflammatory diseases , we assessed its association with type 1 diabetes ( T1D ) . We found evidence for a protective effect of 358Ala in 8 , 371 T1D patients and 10 , 092 unrelated healthy controls ( P = 0 . 0092 , OR [95% CI] = 0 . 94 [0 . 91–0 . 99]; Figure S1 ) . We replicated these findings in an independent collection of 3 , 771 T1D families ( P = 0 . 0035 , OR [95% CI] = 0 . 92 [0 . 87–0 . 97]; Figure S1 ) . These data implicate the IL6R locus in the etiology of T1D , with a consistent protective effect as reported for CHD , RA , AF and AAA ( Figure S1 ) . The minor allele of rs2228145 ( 358Ala ) has been shown to be strongly associated with increased concentrations of circulating sIL-6R [5] , [12] and has , therefore , been assumed to be the causal allele in the IL6R locus . To confirm that rs2228145 is the major determinant of circulating sIL-6R levels in IL6R , and to identify potential additional genetic determinants of sIL-6R levels at this locus , we correlated circulating sIL-6R concentrations measured in 3 , 605 individuals with 45 SNPs genotyped at the IL6R locus using the Illumina ImmunoChip [13] ( Figure 1A and Table S1 ) . Twenty four SNPs were associated with sIL-6R at P<5×10−8 , including rs2228145 , which had the strongest evidence of association ( chi2 ( 2df ) = 2296 , P<10−300 ) . These differences equated to an increase of approximately 35% in the concentrations of sIL-6R for each copy of the 358Ala allele ( Figure 1B and Table S2 ) . Using a forward stepwise regression analysis , two additional intronic SNPs , rs4329505 ( r2 = 0 . 113; D′ = 1 with rs2228145 in CEU ) and rs1386821 ( r2 = 0 . 001; D′ = 0 . 064 with rs2228145 in CEU ) were both found to be independently associated with circulating sIL-6R ( P = 7 . 4×10−29 and P = 5 . 1×10−11 ) . However , their effects were substantially smaller ( R2 = 1 . 1% and 0 . 4% , respectively ) compared to R2 = 29 . 3% for rs2228145 ( Figure 1B and Table S2 ) . Previous studies support two independent processes for the generation of circulating sIL-6R: i ) transcriptional regulation of a differentially spliced isoform ( ds-IL6R; Figure 1A ) [14] and ii ) increased proteolytic cleavage of the membrane-bound receptor [15] , [16] . While it has been speculated that 358Ala affects circulating sIL-6R through both of these processes , there is no comprehensive evidence on the mechanism of the variant , especially with regards to surface IL-6R expression and signaling . To investigate the effect of rs2228145 on the transcriptional regulation of IL6R in human primary cells , we designed a qPCR assay to measure the relative expression levels of both the full length IL6R isoform ( fl-IL6R ) , encoding the membrane-bound receptor , and the ds-IL6R isoform , which lacks exon 9 containing the IL-6R trans-membrane domain , and hence encodes sIL-6R . Importantly , in PBMCs from 88 healthy CBR donors ( Table S3 ) , the expression of fl-IL6R was not significantly different according to rs2228145 genotype ( P = 0 . 8 , Figure 2A ) . Consistent with a previous report on multiple myeloma plasma cells [17] , there was a significant increase of ds-IL6R expression in carriers of the 358Ala allele ( P = 8 . 3×10−22 , Figure 2B ) . The ratio of ds-IL6R transcript to the fl-IL6R isoform was small ( approximately 0 . 03 in Asp/Asp homozygotes ) , as determined by normalizing the expression of the ds-IL6R to the fl-IL6R isoform ( data not shown ) . These data indicate that the skipping of exon 9 is a rare event , which explains why the expression of the fl-IL6R is not affected by genotype at rs2228145 . Neither of the other two SNPs in IL6R that showed an independent association with circulating sIL-6R were associated with ds-IL6R expression , after accounting for rs2228145 ( Table S4 ) . However , considering the relatively small effect sizes for the association of these two SNPs with circulating sIL-6R , our study is likely to be underpowered to detect mRNA expression effects secondary to rs2228145 . The strong effect of rs2228145 in the regulation of circulating sIL-6R has led to the hypothesis that this is the functional mechanism underlying the disease association of this variant . However , very little is known about the potential role of this genetic variant in the regulation of surface expression of IL-6R at the protein level , particularly in immune cell subsets expressing high levels of IL-6R , and whether this could affect classical IL-6R signaling . As 358Ala did not affect the expression of membrane-bound IL-6R at the mRNA level , we next assessed the effect of this variant on the expression of surface IL-6R in individual cells . We used polychromatic flow cytometry to measure the surface expression of IL-6R in PBMCs from 128 Cambridge BioResource donors ( Table S5 ) on the four immune cell subsets of PBMCs which we found to express IL-6R ( Figure S2 ) : CD4+ naïve and memory T cells , regulatory T cells ( Tregs ) and monocytes . The 358Ala allele was strongly associated with reduced surface levels of IL-6R on all four immune cell subsets ( P≤5 . 6×10−22; Figure 3 ) . The effect sizes across the different cell types equated to a per-allele reduction of approximately 20 . 6–27 . 8% ( Figure 3 ) . To exclude the possibility that the observed results arose from a technical artifact caused by an altered epitope induced by or in linkage disequilibrium ( LD ) with rs2228145 , we confirmed IL-6R measurements using a different anti-IL-6R antibody clone ( Figures S3 and S4 and Methods ) . As expected , we found no evidence of genotype-specific differences on the surface expression of the gp130 co-receptor ( Figure S5 ) . Given the reduction of membrane-bound IL-6R expression in 358Ala carriers , we hypothesized that 358Ala would impair classical IL-6 signaling . We quantified the proportion of cells phosphorylating STAT3 and STAT1 in response to IL-6 stimulation in a subset of 14 Asp/Asp and 14 Ala/Ala homozygous donors ( Table S6 ) . While there was no significant difference between genotype groups in the absence of IL-6 , carriers of 358Ala showed a significantly lower frequency of pSTAT activation upon IL-6 stimulation in the three assessed cell types: CD4+ naïve T cells ( PgXd ( pSTAT3 ) = 8 . 6×10−40 , PgXd ( pSTAT1 ) = 5 . 9×10−10 ) , CD4+ memory T cells ( PgXd ( pSTAT3 ) = 5 . 1×10−15 , PgXd ( pSTAT1 ) = 5 . 2×10−7 ) and monocytes ( PgXd ( pSTAT3 ) = 3 . 4×10−15 ) ( Figure 4; for modeled mean differences see Figure S6 ) . The intracellular immunostaining method did not allow Treg discrimination ( see Methods ) . Monocytes were somewhat less sensitive to IL-6 stimulation than the other two cell types , and did not show any noticeable dose response effect with pSTAT1 or genotype-dependent differences ( P = 0 . 06 ) . Qualitatively similar results were observed when assessing the MEF values of pSTAT3 and pSTAT1 induced in the three cell populations ( Figure S7 ) . In addition , we found a strong correlation between the surface levels of IL-6R and pSTAT3 activation at the dose found to have the strongest genotype-dependent differences in pSTAT3 activation in response to IL-6 stimulation ( R2>0 . 7 in CD4+ naïve and memory T cells and R2 = 0 . 6 in monocytes; Figure S8 ) . Further supporting the specificity of surface levels of IL-6R on the observed pSTAT signaling differences , we found no genotype-specific effect on the activation of pSTAT3 or pSTAT1 in response to IL-27 , which shares the gp130 co-receptor or on the activation of pSTAT3 in response to IL-10 , which signals through a different receptor ( data not shown ) . These data are in agreement with a recent report showing that IL-6R-mediated pSTAT3 or pSTAT1 signaling differences in patients with relapsing-remitting multiple sclerosis were not recapitulated following stimulation with IL-10 or IL-27 [18] . Thereby , our study corroborates and improves on recent indirect evidence , showing that the expression of IL-6 target genes after stimulation with IL-6 is associated with a proxy variant in high LD with rs2228145 [10] . Interestingly , IL-6 is known to induce the differentiation of the STAT3-dependent Th17 lineage from naïve T cells [19]–[21] and inhibit FOXP3 expression [22] , [23] . In our study we did not find any evidence for an effect of rs2228145 on the homeostatic frequencies of the four cell types studied ( CD4+ naïve and memory T cells , Tregs and monocytes; Table S7 ) . However , an effect on differentiation of naïve T cells under specific Th17 polarizing and Treg inducing conditions cannot be excluded . The contrasting role of Th17 cells and Tregs in the regulation of inflammation is becoming increasingly apparent [24] , and provides a hypothesis for the protective role of 358Ala in the pathogenesis of human inflammatory diseases .
The accumulating evidence linking the Asp358Ala non-synonymous variant with risk of multiple human diseases underscores the relevance of IL-6 signaling in the etiology of inflammatory diseases . In this study , we provide evidence for the association of the minor 358Ala allele of this variant with protection from T1D . While the association was not at a GWAS level of significance ( for which 22 , 000 cases and controls would be required to have 80% power; Table S8 ) the association was replicated in two independent sample sets with consistent effect estimates as reported for RA , CHD , AF , and AAA . Furthermore , we have confirmed that the 358Ala allele is associated with an increase ( of approximately 35% ) in the concentrations of circulating sIL-6R and demonstrated that while rs2228145 is the major genetic determinant for this trait in the IL6R locus , there are two additional variants with small independent effects . Most importantly , we provided comprehensive evidence for the molecular mechanism of the 358Ala variant . The lack of an association between 358Ala and expression of the mRNA encoding the membrane-bound form of IL-6R , combined with the allele's strong association with reduced surface IL-6R expression at the protein level , suggests that 358Ala exerts an effect on proteolytic cleavage of the membrane-bound receptor independent of its effect on alternative splicing . Increased shedding of surface IL-6R in carriers of the 358Ala allele could be interpreted as a mechanism to dampen chronic classical IL-6 signaling and prevent exacerbated IL-6-driven inflammation , which is concordant with our results . Our findings that 358Ala regulates surface expression of IL-6R were in contrast to a previous study using lymphoblastoid cell lines [12] , which did not find differences in surface IL-6R expression according to rs2228145 . There are , however , several potential reasons for this discrepancy . Firstly , while we determined IL-6R surface expression in specific cell subtypes expressing high levels of IL-6R , the previous study was performed on EBV-transformed B-cell lymphoblastoid lines , with very low or no expression of IL-6R , which are therefore unlikely to show any genotype dependent differences . Secondly , whilst our study used primary cells , which were frozen directly after collection , the previous study employed immortalized cell lines . Cell culturing and immortalization processes are likely to have influenced expression of the IL-6R and potentially also enzymes required for proteolytic cleavage of IL-6R . This was supported by our observation that culturing PBMCs , even for a very short period reduces surface expression of IL-6R ( data not shown ) . Thus , we have provided the first clear evidence that the common non-synonymous variant rs2228145 regulates the balance of surface and sIL-6R , and also affects the responsiveness of immune cells to IL-6 stimulation . This mechanism underpins the effect of 358Ala on the IL-6/IL-6R pathway and has implications for our understanding of the role of IL-6 signaling in the etiology of human disease and therapeutic targeting of this signaling pathway . Inflammation has been implicated in the etiology of RA , AF , AAA , T1D and CHD , including reports of higher circulating levels of IL-6 preceding the onset of some of these conditions [25] , [26] . 358Ala is associated with higher circulating IL-6 levels [5] , [6] , [12] , [27] , but yet protective in these diseases . This apparent paradox may be explained in two ways . Firstly , given our finding of decreased IL-6R surface expression and signaling with carriage of 358Ala , it is likely that the increase in circulating IL-6 is an indirect effect resulting from reduced IL-6 clearance through membrane-bound IL-6R in the liver [28] . This interpretation is concordant with the observation of lower circulating levels of C-reactive protein [5] , [6] , [29] and fibrinogen [5] , [6] , [30] , [31] , liver-derived markers of systemic inflammation , in carriers of 358Ala . Secondly , in the circulation , sIL-6R and sgp130 are in molar excess over IL-6 and any secreted IL-6 will therefore bind to sIL-6R , which will subsequently bind to sgp130 , rendering IL-6 inactive [32] . The 358Ala associated increase in sIL-6R will therefore lead to an increased IL-6 buffering capacity of the circulation , thus explaining the increase in ( inactive ) IL-6 levels , as observed for 358Ala . Therefore , it is likely that the observed reductions in IL-6 classic signaling in carriers of 358Ala in this study may be even more pronounced under physiological conditions , where IL-6 can be inactivated by the sIL-6R/gp130 buffer system . Interestingly , it has been shown that monocytes isolated from long-term T1D patients spontaneously produce increased concentrations of IL-6 , which has been linked to the differentiation of pathogenic Th17 cells [33] . Under these conditions , reduced classical IL-6 signaling could protect from chronic inflammation and differentiation of Th17 cells , consistent our finding of reduced risk of T1D in carriers of the 358Ala allele . In contrast to the strong effect of 358Ala on classical IL-6 signaling , the effect of this variant on trans-signaling is less clear , since the nature of the samples available for this study did not enable us to investigate trans-signaling . Although we confirmed that 358Ala increases the levels of the trans-signaling mediator sIL-6R in the circulation , a material effect on systemic trans-signaling activity is unlikely , since the vast majority of the IL-6/sIL-6R complexes will be bound to the natural inhibitor sgp130 [3] , [32] and , therefore , be inactive . This is consistent with the protective effect of 358Ala in RA , which mirrors the clinical effect of tocilizumab , an anti-IL-6R agent that blocks both classic and trans-signaling [34] . However , the effects of 358Ala on sIL-6R generation may be relevant in the local context of specific tissues . In the lung , for example , experimental evidence in mice suggests distinct roles for classic and trans-IL-6 signaling in the progression of airway inflammation and asthma [35] . While classic signaling appears responsible for regulatory T-cell suppression , trans-signaling seems to promote T helper 2 cell polarization in the lung . IL-6R has been shown to be expressed in the epithelium , smooth muscle and vascular endothelium of human airways , and in macrophages and granulocytes of bronchoalveolar lavage fluid ( BALF ) [36] . Soluble IL-6R levels in BALF are elevated in asthmatic patients compared to controls , and are elevated further upon allergen challenge [35] , indicating an important role of sIL-6R in the context of the lung and associated tissues . Consistent with these findings , 358Ala is also associated with an increased severity of asthma [36] . Future studies examining the effects of this variant on the lung and associated tissues are therefore warranted . More generally , as our findings support a causal role for classical IL-6 signaling in CHD , they intensify interest in recently-launched phase III trials of anti-inflammatory agents in the secondary prevention of CHD [37]–[40] . Furthermore , our findings , in conjunction with genetic associations with RA and CHD , and now T1D at the same locus , suggest IL-6 signaling pathways as a potential mechanistic link between these conditions , with immediate clinical relevance given the increased risk of CHD in RA [41] , [42] and T1D patients [43] , [44] . Our findings have established that a common variant in IL6R , previously known to be associated with several common complex diseases , has specific and important effects on IL-6 signaling . As the IL-6 pathway is a major therapeutic target for several human diseases [45] , [46] , our findings should inform the clinical development of IL-6 inhibitors and encourage exploration of pharmacogenetic or stratified medicine approaches that exploit common functional genetic variation in IL6R to optimize targeting and dosing of such agents in people with different genetic profiles .
All T1D patients were under 17 years of age at diagnosis and recruited from across Great Britain and have been described previously [47] . Controls were obtained from the British 1958 Birth Cohort and the Wellcome Trust Case-Control Consortium UK Blood service Common Control ( UKBS ) sample collection [48] , [49] both of which were recruited from across Great Britain and geographically matched to cases in tests for association with T1D . All 3 , 771 T1D families were of white European decent . 403 multiplex ( affected sibling pair ) families were from the Diabetes UK Warren I collection; 43 simplex families from Yorkshire , Great Britain; 211 multiplex/simplex families from Northern Ireland; 275 multiplex families were from the Human Biological Data Interchange; 956 multiplex/simplex families from Finland; 215 simplex families from Romania; the remaining 1 , 668 multiplex families were made available through the Type 1 Diabetes Genetics Consortium . In addition to 1 , 761 T1D cases and 1 , 030 British Birth cohort controls , 2 , 221 plasma samples were randomly selected from blood donors who joined the Cambridge BioResource during their local blood donation sessions at NHS Blood and Transplant , for measurement of circulating sIL-6R levels ( 1 , 034 recruited in 2007 , BR4000 and 1 , 187 recruited in 2009 , BR8000 ) . All subjects were of white European ancestry . All samples and information were collected with written and signed informed consent . The study was approved by the local Peterborough and Fenland research ethics committee for the project entitled: ‘An investigation into genes and mechanisms based on genotype-phenotype correlations in type 1 diabetes and related diseases using peripheral blood mononuclear cells from volunteers that are part of the Cambridge BioResource’ ( 05/Q0106/20 ) . Samples for sIL-6R analysis were genotyped using the Illumina ImmunoChip platform [13] . Individuals for the T1D association analysis ( 8 , 371 T1D patients and 10 , 092 healthy controls ) , T1D family collections and all Cambridge BioResource volunteers selected for cell-based experiments were genotyped at rs2228145 using TaqMan ( Applied Biosystems ) . Circulating sIL-6R concentrations were measured using a highly sensitive non-isotopic time-resolved fluorescence ELISA assay based on the dissociation-enhanced lanthanide fluorescent immunoassay technology ( DELFIA; PerkinElmer ) . Test plasma samples diluted 1∶20 in PBS+10% FBS were measured in duplicate on 384-well MaxiSorp microtiter plates ( Nunc ) , coated with 1 µg/ml monoclonal anti-human IL-6R antibody ( clone 17506; RD Systems ) . Detection was performed using a biotinylated mouse anti-CD126 monoclonal antibody ( clone M182 , BD Biosciences ) diluted to a final concentration of 100 ng/ml in PBS+10% FBS and a Europium-Streptavidin detection solution ( PerkinElmer ) , diluted in PBS+0 . 05% tween , 1% BSA , 7 µg/ml DTPA to a final concentration of 0 . 05 µg/ml . Quantification of test samples was obtained by fitting the readings to a human recombinant IL-6Rα ( RD systems ) serial dilution standard curve plated in quadruplicate on each plate . mRNA expression was measured in 88 healthy CBR donors . cDNA was generated from total RNA isolated from peripheral blood mononuclear cells ( PBMCs ) as described previously [50] . Relative expression of the fl-IL6R and ds-IL6R isoforms was determined by real-time quantitative PCR ( TaqMan , Applied Biosystems ) . We designed two specific forward primers to hybridize to unique exon boundaries of IL6R ( fl-IL6R: exons 9/10; ds-IL6R: exons 8/10 ) and a common reverse primer and probe to amplify the two IL6R isoforms . Primers and probe sequences are summarized in Table S9 . Full-length IL6R and ds-IL6R expression was normalized to the β2 microglobulin housekeeping gene . In addition , ds-IL6R was normalized to fl-IL6R to allow approximation of the ds-IL6R/fl-IL6R ratio . All probes were labeled with FAM and a non-fluorescent quencher ( BHQ1; Sigma ) . PCR efficiency and the amplification factor for each reaction were calculated using a 1∶2 serial dilution curve of 8 random cDNA samples . The PCR efficiency was 100 . 2% , 94 . 2% and 97 . 3% and the resulting amplification factor was 2 . 0024 , 1 . 9423 and 1 . 9735 for the B2M , fl-IL6R and ds-IL6R reactions , respectively [51] . Relative gene expression levels were calculated using the formula af∧ΔCT , with af representing the amplification factor of the respective PCR reaction and ΔCT the difference in cycle threshold between the target and control transcripts . 128 samples for flow cytometry experiments were selected according to rs2228145 genotype from a CBR subset with available cryopreserved PBMCs ( Table S5 ) . To reduce the effects of day-to-day variation , where possible , rare homozygotes ( Ala/Ala ) were matched to one or more common homozygotes and up to three heterozygotes , from the same 10-year age band and sex . Batches of up to ten samples were assembled from the matched groups , maximizing diversity of the age , sex and T1D distribution within each batch . Since there were no differences in IL-6R surface expression between T1D patients and controls , we included T1D cases in the primary analyses to increase power . Primary analyses were adjusted for T1D status , because limited availability of Asp/Asp and especially Ala/Ala T1D cases resulted in oversampling of T1D cases in the heterozygote group ( Table S5 ) . All assays were performed blinded to sample genotype and disease status . PBMCs were isolated and cryopreserved ( 10×106 cells/aliquot ) as described previously [52] . Cryopreserved PBMCs were thawed in a 37°C water bath and resuspended in X-Vivo ( Lonza ) +10% heat-inactivated , filtered human AB serum ( Sigma ) in a drop-wise fashion and then washed in X-Vivo+1% AB serum . 5×105 cells were stained for 1 h at 4°C ( using the surface immunostaining Panel 1 detailed in Table S10 ) , washed with BD CellWash ( BD Biosciences ) and then fixed with freshly prepared BD CellFix ( BD Biosciences ) . To exclude the possibility that an altered IL-6R epitope induced by the Asp358Ala polymorphism could affect the binding affinity of the anti-IL6R antibody used in this study ( clone UV4 ) , we designed a second surface immunostaining panel ( Panel 2; Table S10 ) . There was a near-perfect correlation between the two anti-IL-6R clones used in this study in the three immune cell subsets that were directly comparable ( Spearman rho >0 . 93; Figure S3 ) . We found that measurements of IL-6R expression were reproducible between two independent measurements ( Spearman rho >0 . 71; Figure S9 ) . We also performed surface staining of random PBMC donors using global lineage discrimination markers , including α-CD19 , α-CD8 and α-CD56 to assess surface IL-6R and gp130 expression levels on the main immune cell subsets in addition to CD4 T cells and monocytes , which we have focused on this study . IL-6 signaling experiments employed a subset of the matched groups for the surface immunostainings , consisting of one Asp/Asp homozygote and one Ala/Ala homozygote pair from each matched group , for those where additional cryopreserved PBMC aliquots were available . All assays were performed blinded to sample genotype and disease status . To assess IL-6 responsiveness , cryopreserved PBMCs were thawed as described above and aliquots of 5×105 cells resuspended in 100 µl X-Vivo+1% AB serum , plated in a U-bottom 96-well cell culture plate ( Cellstar ) and rested for 10 min at 37°C before stimulation . Following stimulation with 0 , 0 . 1 , 1 or 10 ng/ml of IL-6 for 10 min ( 37°C , 5% CO2 ) , cells were immediately fixed with BD phosphoflow lyse/fix buffer ( BD Biosciences ) to maintain their phosphorylation state and incubated for 10 min ( 37°C , 5% CO2 ) . After washing with PBS+0 . 2% BSA , cells were permeabilized with 100% , ice cold , methanol and incubated at 4°C for 30 min . Cells were then washed with PBS+0 . 2% , and blocked for 15 min with PBS+1% BSA . Staining was performed as described above using the intracellular immunostaining Panel 1 ( Table S10 ) and cells resuspended in PBS+0 . 2% BSA . The anti-CD127 and anti-CD25 antibodies required for Treg discrimination were not compatible with the methanol fixation and , therefore , not included in the intracellular immunostaining panel . Immunostained samples were analyzed using a BD Fortessa ( BD Biosciences ) flow cytometer with FACSDiva software ( BD Biosciences ) . Flow cytometry data were exported in the format 3 . 0 and analyzed using FlowJo ( Tree Star , Inc . ) . Gating strategy was performed as depicted in Figure S2 . Doublet exclusion was performed for both CD4+ T cell and monocyte populations . Cyto-Cal calibration beads ( Thermo Scientific ) were used to assess instrument stability and to convert individual mean fluorescence intensity ( MFI ) values into normalized molecules of equivalent fluorochrome ( MEF ) values as described previously [53] . Distribution of the unstained ( FMO; fluorescence minus one ) , isotype control and stained test samples using the anti-IL-6R UV4 clone ( surface immunostaining Panel 1 ) or anti-gp130 antibodies ( surface immunostaining Panel 2 ) are depicted in Figure S10 . Statistical analyses were performed using Stata ( www . stata . com ) , regional association plots were produced by LocusZoom [54] . For the sIL-6R assay , the average CV between duplicate samples was relatively consistent for each batch , and averaged 8 . 58% in the entire study ( samples with CV>15% were excluded from the analyses ) . Technical repeatability was examined by measuring the same 16 samples within each of the 6 batches . The average CV for the 16 repeats performed in each batch was 8 . 1% ( range 6 . 86% to 20 . 81% ) , which suggests these results were as consistent as duplicate samples within the same plate ( hence no obvious batch effects ) . An average R2 of 0 . 9995 ( with a range of 0 . 9975 to 1 ) was determined for the standard curves across all batches . The minimum and maximum detectable concentrations within the linear range were given as 4 . 69 ng/ml and 300 ng/ml respectively . All measured plasma samples were within these values , with a minimum of 8 . 3 ng/ml and a maximum of 133 . 4 ng/ml . For qPCR measurements , test samples were measured in duplicate , with an average intra-assay CV% of 0 . 97% , 0 . 67% and 0 . 78% between replicates for the B2M , fl-IL6R and ds-IL6R reactions , respectively ( data not shown ) . The flow cytometer was found to be very stable during the entire experimental procedure for both the PE and APC quantification channels ( average CV% for the five fluorescent bead populations = 1 . 77% and 4 . 63% , respectively ) . For the anti-IL-6R competition assays , aliquots of 5×105 PBMCs from a donor with high levels of surface IL-6R expression were resuspended in 100 µl X-Vivo+1% AB serum and plated in a U-bottom 96-well cell culture plate ( Cellstar ) . Cells were incubated with 0 , 0 . 01 , 0 . 1 or 1 µg of unconjugated anti-IL-6R UV4 antibody , unspecific mouse IgG1κ control , or with 1 µg of the unconjugated 17506 anti-IL-6R clone used in the ELISA assays for 1 h at 4°C . Cells were then washed with PBS+0 . 2% BSA ( Sigma ) and stained using either the surface immunostaining Panel 1 or Panel 2 as described above . The unconjugated UV4 monoclonal antibody ( mAb ) was unable to block binding of labeled BL-126 . Conversely , the anti-IL-6R clone ( 17506 ) , used to measure sIL-6R , only inhibited BL-126 staining ( Figure S3 ) , showing that both clones bind to the same or adjacent epitopes . Since Ala358 was associated with increased concentrations of sIL-6R measured by ELISA using the 17506 mAb , these data demonstrate that , not only the UV4 and BL-126 clones are recognizing different epitopes , but also that the genotype-specific differences in IL-6R expression could be not be due to structural differences affecting antibody affinity caused by Asp358Ala . | Interleukin-6 ( IL-6 ) is a complex cytokine , which plays a critical role in the regulation of inflammatory responses . Genetic variation in the IL-6 receptor gene is associated with the risk of several human diseases with an inflammatory component , including coronary heart disease , rheumatoid arthritis , and asthma . A common non-synonymous single nucleotide polymorphism in this gene ( Asp358Ala ) has been suggested to be the causal variant in this region by affecting the circulatory concentrations of soluble IL-6R ( sIL-6R ) . In this study we extend the genetic association of this variant to type 1 diabetes and provide evidence that this variant exerts its functional mechanism by regulating the balance between sIL-6R ( generated through cleavage of the surface receptor and by alternative splicing of a soluble IL6R isoform ) and membrane-bound IL-6R . These data show for the first time that the minor allele of this non-synonymous variant ( Ala358 ) directly controls the surface levels of IL-6R on individual immune cells and that these differences in protein levels translate into a functional impairment in IL-6R signaling . These findings may have implications for clinical trials targeting inflammatory mechanisms involving IL-6R signaling and may provide tools for identifying patients with specific benefit from therapeutic intervention in the IL-6R signaling pathway . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"genomics",
"genetic",
"association",
"studies",
"personalized",
"medicine",
"genetics",
"immunology",
"biology",
"human",
"genetics",
"genetics",
"of",
"disease",
"genomic",
"medicine",
"pharmacogenomics",
"autoimmunity"
]
| 2013 | Functional IL6R 358Ala Allele Impairs Classical IL-6 Receptor Signaling and Influences Risk of Diverse Inflammatory Diseases |
Five X-linked severe combined immunodeficiency patients ( SCID-X1 ) successfully treated with autologous bone marrow stem cells infected ex vivo with an IL2RG-containing retrovirus subsequently developed T-cell leukemia and four contained insertional mutations at LMO2 . Genetic evidence also suggests a role for IL2RG in tumor formation , although this remains controversial . Here , we show that the genes and signaling pathways deregulated in murine leukemias with retroviral insertions at Lmo2 are similar to those deregulated in human leukemias with high LMO2 expression and are highly predictive of the leukemias induced in SCID-X1 patients . We also provide additional evidence supporting the notion that IL2RG and LMO2 cooperate in leukemia induction but are not sufficient and require additional cooperating mutations . The highly concordant nature of the genetic events giving rise to mouse and human leukemias with mutations at Lmo2 are an encouraging sign to those wanting to use mice to model human cancer and may help in designing safer methods for retroviral gene therapy .
SCID-X1 patients are deficient in the common γ chain of the interleukin-2 receptor ( IL2RG ) [1] . In three SCID-X1 trials , CD34+ hematopoietic stem cells were cultured ex vivo and transduced with a defective Moloney murine leukemia virus ( MuLV ) expressing IL2RG and then transplanted back to the patients . The possibility that retroviral gene therapy could induce cancer through insertional mutagenesis had been widely discussed but until these trials no cases had been reported . Among the five leukemias that have occurred ( four in the French and one in the UK trial ) , four had insertional mutations at LIM domain Only 2 ( LMO2 ) [2] , [3] , [4] , [5] . LMO2 is a T-cell oncogene [6] , suggesting that these leukemias resulted from insertional mutagenesis . SCID-X1 is caused , in part , by a failure of T-cell production , and the infusion of immature gene-corrected CD34+ cells into SCID-X1 patients favors the engraftment of the T-cell lineage over other lineages . This could explain why T-cell leukemias predominate in these patients . LMO2 is also expressed early in hematopoiesis [6] , [7] which makes it a good target for insertional mutagenesis since retroviruses like to preferentially integrate near the 5′ end of actively transcribed genes [8] . LMO2 is not the only T-cell oncogene expressed during early hematopoiesis , however , indicating there must be other reasons that favor mutagenesis of LMO2 . A murine T-cell leukemia with insertional mutations at Lmo2 and Il2rg has also been reported [9] . The probability of finding insertional mutations in both genes by chance in the same leukemia is exceedingly small and has led to the suggestion that Il2rg is a leukemia gene that cooperates with Lmo2 . While IL2RG is not overexpressed in SCID-X1 leukemias [3] or in the mouse leukemia with an Il2rg insertion , subtle effects on its expression , such as an inability to downregulate its expression during T cell development , could be oncogenic [10] . Given the large number of IL2RG-infected cells transplanted into each patient , there is ample chance that a patient would receive a cell that contains an insertional mutation at LMO2 [9] . Why then don't all patients develop leukemia ? The most likely explanation is that other cooperating mutations are needed for leukemia to occur . Consistent with this , leukemias take several years to develop and contain mutations in other T-cell oncogenes [3] , [11] . To provide further insights into this problem we cloned and sequenced the retroviral integrations from five murine leukemias containing insertional mutations at Lmo2 using high-throughput ligation-mediated PCR ( LM-PCR ) /sequencing method that makes it possible to identify most of the insertionally mutated genes in these leukemias . We then compared the microarray data from human leukemias with upregulated LMO2 expression and murine leukemias with insertional mutations at Lmo2 . Our studies show that murine leukemias are highly predictive of the leukemias induced in SCID-X1 patients . They also support a model whereby deregulated IL2RG and LMO2 expression cooperate to induce leukemia but are not sufficient and require other cooperating mutations .
The mouse Retroviral Tagged Cancer Gene Database ( RTCGD ) ( http://rtcgd . abcc . ncifcrf . gov ) lists four AKXD murine leukemias with insertional mutations at Lmo2 [9] , in addition to the AKXD leukemia analyzed previously which contains insertional mutations at Lmo2 and Il2rg . All of the Lmo2 insertions are located 5′ of coding exons 4–6 and are in the same general location as the insertions identified at LMO2 in SCID-X1 patient leukemias and the chromosomal breakpoints at LMO2 identified in sporadic human T-ALL [3] , [12] ( Figure 1A ) . Southern analysis showed these insertions are clonal ( Figure 1A , bottom left panel , data not shown ) and contain 2–4 clonal retroviral insertions each ( Figure 1A , bottom right panel ) , as would be expected if they harbor mutations in Lmo2-cooperating genes . Tumors 98-031 and 86-277 are T cell in origin [9] . Tumors 7105 and 7107 are also likely to be T cell as they have clonal Igh and T-cell receptor Jβ1 rearrangements ( Figure 1B ) . The origin of tumor 3095 is unclear . It was isolated from a mouse with thymomegaly and lymphadenopathy and contains Igh but no T-cell receptor gene rearrangements ( Figure 1B , Jβ2 not shown ) . Tumors 3095 and 7105 showed aberrant B220 staining ( Figure 1C ) , which has been seen before in T-cell tumors from Lmo2 transgenic mice [13] and in some human T-ALLs expressing B-cell markers [14] . To identify genes that might cooperate with Lmo2 in tumor induction we adapted an LM-PCR method to amplify the other viral insertions present in the five AKXD leukemias [8] , [15] . Following nested PCR using virus-specific and adaptor-specific primers , the amplified products were shotgun cloned . The inserts were then sequenced and BLAST-searched against the mouse genome and nearby candidate cancer genes identified ( Table 1 ) . These data were then combined with a less robust data set generated using inverse PCR ( IPCR ) [16] . This analysis identified 84 insertions in the five tumors ( see Dataset S1 ) . The percentage of cells in a tumor that harbor an insertion can be estimated from the number of shotgun clones isolated ( Freq % in Dataset S1 ) . Insertions present in all tumor cells will be enriched during PCR relative to insertions present in only a fraction of tumor cells and will thus be overrepresented in the shotgun library . Remarkably , 35 insertions are located at common insertion sites ( CISs ) , a highly significant result ( Fisher's exact test p = 7 . 4×10−39 ) ( Table 1 ) . CISs are regions in the genome that are mutated by viral insertion at a rate higher than predicted by random chance and are thus likely to harbor a cancer gene [16] . Thirty-one of these genes are also mutated in human cancer , another highly significant result ( Fisher's exact test p = 5 . 7×10−37 ) ( Table S1 , Dataset S1 ) [17] . This strongly suggests that some of these genes are Lmo2-cooperating genes . Surprisingly , tumor 7107 contains two Il2rg insertions ( Table 1 ) , which was confirmed by conventional cloning and sequencing ( Figure 2A ) . Thus , 2 of the AKXD murine leukemias with insertions at Lmo2 also contain insertions at Il2rg , a highly significant result ( p = 1 . 34×10−9 , see Text S1 for calculation ) . Likewise , tumor 7107 has two insertions at Irs2 ( Figure 2B ) , another highly significant result ( p = 1 . 16×10−5 , see Text S1 for calculation ) . Il2rg and Irs2 are functionally linked in lymphoid cells where it has been shown that Il2rg can promote the phosphorylation of Irs2 by binding to and activating the tyrosine kinase Jak3 [18] , which may explain the co-selection for mutations in these genes . To quantitate the Irs2 and Il2rg insertions , we amplified one Lmo2 and one Il2rg insertion using insertion-site-specific primers and real time PCR . The Irs2 insertion was present at about one copy per cell ( Figure 2B ) , indicating it is present in every tumor cell . The other Irs2 insertion must therefore be in the same tumor cell , either on the same or different chromosome . In contrast , the Il2rg insertion was present at 0 . 5 copies per cell ( Figure 2B ) , suggesting it is present in only half the tumor cells of this male mouse . We ruled out hyperploidy for the Irs2 locus and also confirmed that the Il2rg gene was present at one copy per cell ( see Figure S4 ) . The two Irs2 insertions must therefore have occurred first and in the same tumor cell followed by the two Il2rg insertions in different subpopulations of tumor cells . Similar to what was reported previously [9] , Il2rg is not misexpressed in tumor 7107 ( see Figure S3 ) . Most of the genes that were insertionally mutated in this tumor were also highly overexpressed when compared to normal thymus control ( Figure 2C ) . Since our prior study , exons for the Med12 gene were annotated and are shown to direct transcription in the opposite orientation to Il2rg ( see Figure 2A ) . All the tumors have insertions in the same orientation to Med12 except tumor 98-031 and one of the insertions disrupts the second coding exon of Med12 . Med12 was not found to be up-regulated in the tumors and no spliced fusion transcript between provirus and Med12 could be identified ( see Figure S3 ) . It is conceivable that these insertions implicate Med12 in tumorigenesis and not Il2rg , however , the RTCGD contains many common insertion sites in genes of the Il2rg pathway ( e . g . Il2ra , Il4ra , Il7 , Jak1 , Stat5a/5b ) . Several other genes are also mutated more than once in the Lmo2 tumors , indicating that they might also represent Lmo2-cooperating genes . Two tumors have insertions in the leukemia transcription factor oncogene Prdm16 ( Figure 3A ) but only the insertion in tumor 7107 is clonal ( Figure 3A , bottom left panel ) . Quantitative RT-PCR/sequencing revealed marked upregulation of a Prdm16 fusion transcript in tumor 7107 that initiates in the viral 5′LTR and splices to Prdm16 exon 2 ( Figure 3A , bottom right panel ) . This transcript is predicted to encode a truncated protein similar to that expressed in human leukemias with PRDM16 mutations [19] . Remarkably , 3 tumors also have insertions in intron 2 of Mef2c ( Figure 3B , top panel ) , which is a transcription factor oncogene that cooperates with Sox4 in leukemia induction [20] . All insertions are in the same transcriptional direction as Sox4 and are clonal ( Figure 3B , bottom left panel ) . Quantitative RT-PCR/sequencing revealed high expression of Mef2c fusion transcripts in these tumors . The transcripts all initiated in the viral 5′LTR and spliced into the first coding exon of Mef2c ( Figure 3B , bottom right panel ) , similar to what is reported for other leukemias with Mef2c insertions [20] . Tumor 7107 also showed high Mef2c expression ( Figure 3B , bottom right panel ) and has an insertion located 390 kb upstream of Mef2c ( Table 1 ) . There is precedent for long-range enhancer effects exerted by proviruses [21] , [22] , potentially accounting for the high Mef2c expression . Thus , 4 of 5 AKXD tumors have insertions at or near Mef2c , strongly suggesting that Mef2c is an Lmo2-cooperating oncogene . Consistent with this , tumors 98-031 , 7105 and 86-277 also had insertions near the Mef2c-cooperating gene Sox4 , although only the insertion in tumor 7105 was clonal ( data not shown ) . Finally , tumors 98-031 and 7105 have insertions in the 5′ end of the putative acetylglucosaminyltransferase gene , B3gntl1 ( Table 1 ) . Very little is known about this gene so the significance of this result is unclear . We also identified insertions in two E2A-related genes , Tcfe2a ( TCF3 in humans ) and Tcf12 ( Table 1 ) , suggesting a role for E2A-related genes in the Lmo2 tumors ( p = 0 . 008 ) . TCF3 is translocated to a number of fusion partners in pre- and pro-B-ALL [23] , while TCF12 is fused to NR4A3 in extraskeletal mixoid chondrosarcoma [24] . The Tcf12 insertion is located in intron 5 and is in the same transcriptional orientation as Tcf12 ( Figure 4A ) . 5′RACE showed that it induces a fusion transcript , which initiates in the 5′LTR and splices to Tcf12 exon 9 ( Figure 4B ) . The first ATG is located in exon 9 and is in-frame with Tcf12 coding sequences . Assuming this is the preferred translational start site , the predicted polypeptide would lack 219 residues from the amino terminus , which was confirmed by in vitro transcription and translation studies ( Figure 4C , left panel ) . The major polypeptide migrated at 55 kDa , similar to its predicted mass of 52 kDa . A faint 50-kDa band was also observed . Both proteins were immunoprecipitated with a rabbit polyclonal antibody specific to Tcf12 , confirming they are Tcf12-derived ( Figure 4C , right panel ) . The smaller protein could arise from an alternate translational initiation site or be due to protein degradation . Tcf12 and Tcfe2a are class I bHLH transcription factors . The amino terminus of Tcf12 encodes two transactivation domains , AD1 and AD2 . AD1 is similar to AD1 of the related E2A proteins , which has been shown to have greater potency in transactivation than AD2 [25] . While there are no mutagenesis data on the homologous domains of Tcf12 , the truncated Tcf12 transcript expressed in tumor 3095 would lack the AD1 domain but have an intact AD2 domain , which could result in attenuated transactivation . Microarray analysis showed there was considerable differential expression of E2A target transcripts when the five tumors were compared to normal thymus [26] ( Figure 4D ) . Similar , but somewhat less , differential expression , was observed when the tumors were compared with tumor 7065 , which has an insertion at Notch1 ( Figure 4D ) . With the exception of Gfi1b , targets activated by E2A are poorly expressed as expected if E2A signaling is attenuated . Conversely , Gm2a , which is repressed by E2A , was highly expressed . These results provide further evidence that the E2A pathway is attenuated in tumors with mutations at Lmo2 . Microarray expression analysis also showed that these tumors express Tal1 as well as the other experimentally verified Lmo2 binding partners , consistent with the involvement of Lmo2 in these tumors ( Figure S1 ) . To provide additional evidence that the genes insertionally mutated in murine Lmo2 tumors are causally associated with human T-ALL , we examined the raw data from three large human T-ALL microarray data sets ( total of 118 cases ) [27] , [28] , [29] and asked whether any of the murine genes are deregulated in human T-ALL with high LMO2 expression . We separated the human T-ALL cases into LMO2-high and LMO2-low expression classes and performed clustering analysis to identify the genes differentially expressed in the LMO2-high class ( p<0 . 001 significance level ) . The data sets were remarkably consistent in the genes that were upregulated in the LMO2-high class , despite different array platforms and patient heterogeneity . When genes that were consistently deregulated in the LMO2-high class were queried against the RTCGD , a statistically significant number were located at CISs or are insertionally mutated in murine tumors with Lmo2 insertions ( Figure S2 ) , and the number of such genes is much higher than expected by chance ( Figure S2; Ferrando et al , p = 1 . 1×10−3 and Yeoh et al , p = 1 . 1×10−3; Chiaretti et al , p = 4 . 1×10−6 ) . Remarkably , MEF2C was overexpressed in the LMO2-high group in all three T-ALL data sets ( Figure 5A ) . This is extremely unlikely to have occurred by chance and further confirms the role of this gene in human T-ALL . Likewise , LAPTM5 , which is the site of an insertion in tumor 7107 , was identified in all three data sets , while NMYC , an insertion site in tumor 7105 , was identified in two microarray data sets ( Figure 5A ) . STAT5A was found in two data sets . Stat5a is not insertionally mutated in the murine tumors but it is a validated leukemia gene and functions in the same signaling pathway as IL2RG . Most of the genes that clustered with the LMO2-high class were also highly expressed in the mouse tumors and some showed differential expression in comparison to normal thymus or tumor 7065 ( Figure 5B ) . Multiple RAP1 pathway genes also clustered with the LMO2-high group , including RAP1B , RAP1A and RAPGEF5 . None are insertionally mutated in AKXD tumors , however , Rap1gds1 is insertionally mutated in tumor 98-031 and is the site of a recurrent chromosomal translocation in human T-ALL [30] . Rap1gds1 stimulates the exchange of GTP for GDP on Rap1 and is an activator of the Rap1 pathway . Upregulation of Rap1gds1 could therefore produce the same effect as upregulating RAP1B , RAP1A or RAPGEF5 . Pathway analysis on genes shared in at least two microarray datasets ( mouse and human combined ) also suggested a role for cytokine signaling in tumor development ( see Figure S5 ) .
To provide a better mechanistic understanding of SCID-X1 patient leukemias , we have analyzed five murine T-ALLs with insertional mutations at Lmo2 . In each case these insertions are clonal . Therefore , they must have occurred early in tumor induction , similar to four of the five cases of gene therapy-induced leukemias . Transcriptional profiles of these T-ALLs showed high expression of the Tal1 and Lyl1 oncogenes as well as other Lmo2-binding partners such as Gata1 and Gata2 . TAL1 and LYL1 are class II bHLH transcription factors that are frequently overexpressed along with LMO2 in human T-ALLs [27] , [31] . Consistent with this , patient #4 , who developed leukemia in the French trial , and patient #8 , who developed leukemia in the English trial , showed rearrangement of the SIL-TAL1 loci along with insertional activation of LMO2 [3] , [5] , [32] . Murine T-ALLs with Lmo2 insertions also showed consistent downregulation of genes activated by E2A and upregulation of genes repressed by E2A , consistent with previous studies suggesting that Lmo2 redirects E2A activity by binding it through its partner proteins , Tal1 or Lyl1 [6] . Genes that are activated by the Lmo2/Tal1/E2A/Gata1-containing complex in normal erythrocyte development were also overexpressed in one murine T-ALL with high Gata1 expression . These findings suggest that murine T-ALLs with Lmo2 insertions resemble human T-ALLs that are initiated by LMO2 deregulation . Our high throughput insertion site analysis identified additional disease-related genes that are likely to cooperate with Lmo2 in leukemia induction . We identified a second tumor ( 7107 ) , in addition to tumor ( 98-031 ) described previously [9] that has an Il2rg insertion . Remarkably , this tumor has two insertions at Il2rg that occurred independently suggesting strong selection for Il2rg deregulation in Lmo2-initiated T-ALL . In addition , tumor ( 7065 ) has insertions in Notch1 and Il2rg , suggesting that Il2rg might be able to cooperate with other T-cell oncogenes ( i . e . Notch1 ) in leukemia induction . The newly identified insertions are also close to or within the Med12 gene . Similar to Il2rg , this gene is not activated or downregulated by the insertions . We favor Il2rg as the retroviral target since the RTCGD is replete with common insertions in cytokine receptors that act in the exact same pathway as Il2rg . More recently , we have found Il2rb is a common insertion site in a retroviral insertional mutagenesis screen in Lmo2 transgenic mice ( U . P . Davé , unpublished observation ) . Other Il2rg binding partners , Il7ra and Jak1 , were insertionally activated in the same screen which has not reached saturation . While still somewhat controversial , two studies provide additional evidence that Il2rg can function as an oncogene under some circumstances [33] , [34] . A statistically significant enrichment of Il2rg-dependent cytokine pathways in murine T-ALLs containing Lmo2 insertions and human LMO2-overexpressing T-ALLs was also observed by Ingenuity pathway analysis ( see Figure S5 ) . A role for these pathways in T-ALL is further supported by studies showing that cytokines , which depend on Il2rg for signaling can induce T-ALL in transgenic mice , and by the murine T-ALL with two Il2rg and two Irs2 insertions . This murine T-ALL is significant since Irs2 encodes a protein adaptor that is phosphorylated by JAK3 tyrosine kinase in response to cytokine ligation by Il2rg and its heterodimeric partners such as Il7ra , Il4ra and Il9ra [18] , [35] . The occurrence of leukemias in the SCID-X1 gene therapy trials also lends compelling support to the notion that IL2RG is oncogenic and a cooperating “hit” with LMO2 . Il2rg expression was not upregulated in any of these three tumors but there is precedent for cancer genes being dysregulated without gross overexpression or at specific developmental stages [36] , [37] , [38] . Similarly , the IL2RG transgene was not overexpressed in patients receiving gene therapy or in those developing T-ALL [5] , [11] . We theorize that a gene is less able to be silenced if there is a nearby retroviral insertion . Enforced expression of Il2rg has been shown to cause T-cell leukemias without gross overexpression [33] , [39] . In the study by Woods et al , they specifically remark that transgene levels in thymic lymphomas were comparable to Il2rg expression levels seen in developing thymi [34] . Another compelling possibility is that Il2rg is misexpressed in a target cell population where Il2rg is not normally found such as a cell more primitive than a thymocyte precursor . The study by Shou et al presents an alternate hypothesis that the SCID-X1 background is required for leukemia development perhaps due to altered numbers of hematopoietic precursors or stem cells [33] . It may be that the lack of Il2rg creates a differentiation block that expands a cell type that is susceptible to transformation . Recurrent insertions in genes that are also CIS in the RTCGD were identified by Shou et al but Lmo2 was not one of them . The tumors were not analyzed for Lmo2 overexpression or for Notch1 activating mutations . Our data suggests that leukemia requires many “hits” and in their model somatic mutation in cooperating oncogenes or tumor suppressor genes may be more likely than insertional mutation since their vector was replication-defective . Retroviral insertion site profiles have recently been published for the French and English cohorts of SCID-X gene therapy patients ( n = 14 ) [4] , [5] , [11] , [40] . Presently , there is no statistical difference between the incidence of leukemia in the French ( 4 of 10 ) and English patients ( 1 of 10 ) and with long term follow-up , even the insertion site profiles may be quite similar in the two studies . Comparison of the insertion sites identified in the French patients with those in RTCGD showed there was a statistically significant number of insertions in these patients , which are located at CISs in murine hematopoietic tumors ( 63/554 , Fisher's exact test , p = 4 . 5×10−22 ) . Remarkably , patient #10 of the French trial showed clonal insertions in LMO2 and BMI1; these two insertions were observed in tumor 98-031 , underscoring the predictive power of our mouse models [11] . In designing safer vectors for transduction , it will be important to consider self-inactivating LTRs and perhaps lentiviral backbones , which have different insertion site preferences than gamma-retroviruses , which may carry a lower risk of insertional mutation [41] . With the high transduction efficiencies achieved in these gene therapy trials , one could also see clonal expansion in cases where the transduced gene is not oncogenic . This , in fact , happened recently in two chronic granulomatous disease patients treated by retroviral gene therapy [42] . In both patients , a clonal expansion of the myeloid compartment was observed that began 3 to 5 months post-transplant . Clonal expansion was associated with insertional mutation of just three genes: MDS/EVI1 , PRDM16 , or SETBP1 . Astonishingly , in the mouse , insertional mutation of these same genes has been associated with immortalization of early hematopoietic progenitor and/or myeloid progenitor cells [19] . Insertional mutations in these genes may have been selected in transplant patients due to their effects on increased self-renewal or engraftment potential . Prdm16 is also insertionally mutated in two of the murine T-ALLs with Lmo2 insertions so this gene's involvement in tumor induction might not be limited to the myeloid lineage .
Mice were aged in SPF facilities according to approved IACUC protocols at NCI-Frederick . Various AKXD strains were developed at the Jackson Laboratory . These are recombinant inbreds that have a high spontaneous rate of leukemia or lymphoma onset due to the presence of an endogenous retrovirus [43] , [44] . The mice are viremic from birth and perhaps in utero and develop disease at various latencies , usually over six months of age . Murine leukemia viruses are not introduced . The mice are simply aged in SPF facility until the onset of disease . At the first appearance of morbidity , mice were sacrificed and gross necropsies performed . Organomegaly and abnormal features were noted and lymphoid and hematopoietic tissue was harvested and flash frozen in liquid nitrogen . Tissue was also fixed in formalin for immunohistochemistry . Frozen tissue was used in the preparation of high molecular weight DNA and whole RNA as previously described [45] . Tumor genomic DNA was restriction digested and loaded on to agarose gels for overnight runs . The gels were transferred as described to Nitrocellulose ( Amersham ) [44] . Membranes were baked and UV-crosslinked . They were hybridized with 32P-labeled probes and exposed to film . For quantification , blots were exposed to phosphorimager plates and a Fuji phosphorimager used for quantitation . Details on probes used are available upon request . Paraffin-embedded tumors were sectioned and probed with a labeled BAC encompassing the mouse Irs2 gene . Labeling and FISH were performed by manufacturer's protocol ( Vysis ) . Inverse PCR was performed as previously described [46] . For LM-PCR , tumor genomic DNA ( 1 µg ) was digested overnight with NlaIII or MseI . The internal amplicon was eliminated by a SpeI digest . DNA was column purified ( Amersham GFX ) and ligated to a double stranded adaptor . This adaptor was made by annealing sense and antisense strands with NlaIII or MseI compatible overhangs as previously described [45] . The ligation mix was column purified ( Qiagen ) and then used for nested PCR . The PCR products were column purified and then ligated into pGEMT-Easy ( Promega ) plasmids and plated . Ninety six colonies were picked and miniprep DNA prepared using Whatman Ultra filters . Miniprep DNA was sequenced in 96 well format ( Functional Biosciences ) and files were analyzed in FASTA format . Valid sequences contained 3′LTR and adaptor at either end were then batch BLASTed against the mouse genome . Annotated insertions are posted at http://rtcgd . abcc . ncifcrf . gov . Insertions were confirmed by either Southern analysis or PCR using gene-specific primers in the vicinity of insertions . For tumor 7107 , gene specific primers corresponding to Il2rg and Irs2 were used in real time PCR ( Biorad MyIQ ) to confirm biclonality . For tumor 3095 , virus-specific and Tcf12-specific primers were used in RT-PCR to generate cDNA for the viral-fusion transcript . 5′RACE was done using SMART Race kit ( BD Biosciences ) to verify that the transcriptional start site was in the virus . The full length cDNA was sequenced from both ends and TA-cloned into pGEMT-Easy ( Promega ) ; NotI digest released this cDNA for subcloning into pcDNA3 . 1 ( Invitrogen ) . Orientation was verified and in vitro transcription and translation was performed using the TnT kit ( Promega ) with 35S-labeled methionine ( Amersham ) . In vitro translated Tcf123095RIS was resuspended in RIPA buffer and immunoprecipitated using antibody against the COOH-terminus of Tcf12 ( sc-357 , Santa Cruz Biotechnology ) . Tumors were homogenized in Trizol reagent ( Sigma ) and whole RNA was isolated by manufacturer's protocol . First strand cDNA was synthesized using oligo-dT primer and Superscript II or III reverse transcriptase enzymes ( Invitrogen ) . Primers used for RT-PCR are available upon request . For real time PCR , we used Biorad's MyIQ or ABI 7700 machines with either Sybr green master mixes ( Biorad ) or Taqman probes ( Applied Biosystems ) . Probe assays used are available upon request and were validated by Applied Biosystems . RNA was processed for hybridization to the Affymetrix chips according to standard protocols ( see Text S1 ) . Microarray analysis was performed with BRB-ArrayTools ( version 3 . 2 . 3 ) developed by Biometric Research Branch at the National Cancer Institute . LMO2- positive and LMO2-negative tumors were compared by using univariate significance tests at the significance level of 0 . 001 . The maximum false discovery proportions were restricted to 0 . 1 using multivariate permutation tests . As for the criteria of LMO2-positive and LMO2-negative tumors , there was no clear cut-off level for classifying a sample as having either high or low expression . We used artificial cut-off values to classify tumors into two groups . At first , we identified the median value of LMO2 expression signals as an initial indicator for the classification , and then we removed the marginal cases from the comparisons . For example , in the case of St Jude's data , we classified tumors with LMO2 signal>2000 as LMO2-positive group ( 13 cases ) , tumors with LMO2 signal<1000 as LMO2-negative group ( 14 cases ) . Statistical comparison between tumor classes was done using the “BRB Array Tools” software ( http://linus . nci . nih . gov/BRB-ArrayTools . html ) . We collated CEL file format of Affymetrix data by using the ‘RMA’ method [47] of the ‘affy’ package from BioConductor ( http://www . bioconductor . org/ ) . To identify genes that were differentially expressed among the two classes , we used a random-variance t-test . The random-variance t-test is an improvement over the standard separate t-test as it permits sharing information among genes about within-class variation without assuming that all genes have the same variance [48] . Genes were considered statistically significant if their p values were less than 0 . 001 . A stringent significance threshold was used to limit the number of false positive findings . We also performed a global test of whether the expression profiles differed between the classes by permuting the labels of which arrays corresponded to which classes . For each permutation , the p values were re-computed and the number of genes significant at the 0 . 001 level was noted . The proportion of the permutations that gave at least as many significant genes as with the actual data was the significance level of the global test . We performed cluster analysis of genes and produced a heat map image to represent the over- and under-expression of each gene in each sample . We used quantile data ranges to ensure the even presence of all available colors on the map . For comparison of mouse and human datasets , we used Fisher's exact test , binomial distribution and Chi-square tests to generate p values . Please see Text S1 regarding specific statistical questions analyzed . | Twenty patients with X-linked severe combined immunodeficiency ( SCID-X1 ) have been successfully treated by gene therapy . Unfortunately , five of these patients have developed T-cell leukemia two or more years after receiving the therapeutic gene IL2RG on a retroviral vector . The leukemias developed because the vector inserted itself near cancer-causing genes and disrupted their normal regulation . Remarkably , in four patients , the vector inserted near a known T-cell oncogene , LMO2 . We have found that in mice , similar retroviruses cause T-cell leukemias by inserting near Lmo2 . We have found two leukemias that have retroviral insertions near Lmo2 and Il2rg in the same cell . The probability of these insertions happening by chance is exceedingly small and these results imply that these two genes are deregulated together to induce leukemia . Our data show that Lmo2 and Il2rg cooperate but may not be sufficient for leukemia development and additional mutations contribute to leukemia development . We have also found cooperating retroviral insertions in genes that are abnormally expressed in human T-cell leukemias . The mouse models provide unique insight into the pathogenesis of T-cell leukemia , and they are highly predictive of the leukemias caused by SCID-X1 gene therapy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"hematology/acute",
"lymphoblastic",
"leukemia",
"genetics",
"and",
"genomics/disease",
"models",
"virology/viruses",
"and",
"cancer",
"oncology/hematological",
"malignancies",
"genetics",
"and",
"genomics/cancer",
"genetics"
]
| 2009 | Murine Leukemias with Retroviral Insertions at Lmo2 Are Predictive of the Leukemias Induced in SCID-X1 Patients Following Retroviral Gene Therapy |
Hepatitis C virus ( HCV ) has infected around 160 million individuals . Current therapies have limited efficacy and are fraught with side effects . To identify cellular HCV dependency factors , possible therapeutic targets , we manipulated signaling cascades with pathway-specific inhibitors . Using this approach we identified the MAPK/ERK regulated , cytosolic , calcium-dependent , group IVA phospholipase A2 ( PLA2G4A ) as a novel HCV dependency factor . Inhibition of PLA2G4A activity reduced core protein abundance at lipid droplets , core envelopment and secretion of particles . Moreover , released particles displayed aberrant protein composition and were 100-fold less infectious . Exogenous addition of arachidonic acid , the cleavage product of PLA2G4A-catalyzed lipolysis , but not other related poly-unsaturated fatty acids restored infectivity . Strikingly , production of infectious Dengue virus , a relative of HCV , was also dependent on PLA2G4A . These results highlight previously unrecognized parallels in the assembly pathways of these human pathogens , and define PLA2G4A-dependent lipolysis as crucial prerequisite for production of highly infectious viral progeny .
Approximately 160 million people are chronically infected with hepatitis C virus ( HCV ) [1] . Without treatment , at least 20% of patients will develop liver cirrhosis and of these , approximately 15% will progress to liver cancer within ten years [2] . HCV is the sole member of the genus Hepacivirus within the family of Flaviviridae . Its plus strand RNA genome encodes a polyprotein that is flanked by non-translated regions . Proteolytic processing releases ten viral proteins which coordinate viral RNA replication and particle assembly . The non-structural proteins NS3 , NS4A , NS4B , NS5A and NS5B in conjunction with cellular co-factors are both necessary and sufficient to catalyze genome replication [3] . Core protein , envelope protein 1 and 2 ( E1 , E2 ) reside in the very N-terminal portion of the viral polyprotein and compose the virus particle encasing the RNA genome . These proteins are essential for virus assembly . Interestingly , the ion channel protein p7 , and the NS2 protease also contribute functions to the production of infectious viral progeny [4] , [5] . Lipid droplets have been recognized as an essential cellular organelle for production of infectious HCV progeny [6] . During virus production core protein resident on lipid droplets recruits viral proteins and RNA , which is an essential prerequisite for virus production [6] . In turn , core protein is loaded onto these cellular organelles through an interaction with diacylglycerol acyltransferase-1 ( DGAT-1 ) [7] , an enzyme which catalyzes the final step in the biosynthesis of triglycerides that is essential for lipid droplet biogenesis [8] . In addition , host factors involved in the biosynthesis and secretion of human lipoproteins have emerged as essential cofactors for virus production . Specifically , apolipoprotein B ( ApoB ) , apolipoprotein E ( ApoE ) and microsomal triglyceride transfer protein ( MTTP ) were shown to contribute to virus production [9] , [10] , [11] . Likely as a consequence , infectious HCV is a “lipo-viro particle” rich in cholesteryl esters and comprising viral proteins , ApoB and ApoE [12] , [13] . Cells constantly respond to environmental changes by sensing these alterations through dedicated receptors and associated signaling cascades that reprogram cellular processes . Such signaling-dependent modifications may also influence important cellular HCV dependency factors regulated by these pathways thus providing a lead for identification of novel and possibly druggable host factors crucial for HCV . Using this approach we show that mitogen-activated protein kinase ( MAP kinase ) regulated enzymatic activity of PLA2G4A is crucial for production of infectious HCV progeny highlighting the intricate involvement of host cell lipids and lipid-modifying enzymes in the replication of this virus .
Cultured cells respond to multiple stimuli by growth factors and hormones present in serum-containing culture media . Therefore , to reduce the complexity of our experimental system we developed a transient virus replication assay and cultured cells in serum-free conditions . When transfecting our infectious firefly luciferase reporter virus genome Luc-Jc1 [14] into highly permissive Huh-7 . 5 human hepatoma cells [15] cultured in the presence or absence of serum , we measured comparable levels of luciferase activity 4 to 48 h after transfection ( Figure S1A ) . Likewise , transduction of luciferase activity upon inoculation of naïve cells with culture fluid from the transfected cells was similar ( Figure S1B ) . Therefore , in this transient time course viral RNA replication and production of infectious progeny particles was comparable in both serum-free and serum-containing conditions . To identify new host factors involved in HCV replication and/or virus production we used pathway-specific inhibitors of central signal transduction cascades including AKT/PKB , mTOR , Wnt , JNK and MAPK/ERK ( Figure 1 and data not shown ) . To reveal rapid , signal-mediated changes of RNA replication and virus production we added the inhibitors 41 h post transfection during the logarithmic growth phase of cells . One hour later culture medium was replaced with fresh medium with or without inhibitors and virus production as well as RNA replication was assessed 6 h later . This procedure which is summarized in Figure 1A ensures that specifically infectivity of particles produced and released during inhibitor treatment – and thus blockade of the respective signaling cascade – is evaluated . Among the inhibitors tested , U0126 a selective inhibitor of the MAPK/ERK pathway , substantially decreased the production of infectious virus particles as is evident from the dose-dependent reduction of luciferase activity in the inoculated cells ( Figure 1B ) . Interestingly , U0126 did not measurably impede RNA replication and only affected HCV particle production when added in serum-free medium . This latter finding may be related to a much more efficient blockade of the MAPK/ERK pathway in serum-free conditions compared to serum-containing medium that is evident from a lower level of phosphorylated ERK1 and ERK2 in the presence of the drug when serum was absent ( Figure 1C ) . Collectively , these results suggested that the applied doses of U0126 efficiently prevented phosphorylation of ERK under serum-free conditions thus impeding production of infectious HCV progeny . Interestingly , PD98059 and Sorafenib which inhibit the MAPK/ERK pathway upstream of U0126 [16] [17] also reduced production of infectious HCV particles ( Figure S2 ) further confirming the role of MAPK/ERK signaling in HCV morphogenesis . Notably , at least under serum free conditions , these inhibitors also reduced RNA replication . This may either be linked to a more potent suppression of MAPK/ERK signaling or due to inhibition of additional signaling events . To confirm these findings , we analyzed the impact of U0126 on the production of infectious wildtype Jc1 particles [18] in transfected cells ( Figure 1D and E ) . Congruent to our findings with reporter viruses , treatment of cells with U0126 reduced both extracellular and intracellular levels of infectious HCV particles ( Figure 1D ) . Although intracellular levels of core and NS5A were moderately reduced in the absence of serum ( Figure 1E ) , addition of U0126 did not further reduce abundance of viral proteins suggesting that the inhibitor did not prevent RNA translation or RNA replication . We also tested if presence of U0126 interferes with HCV cell entry by adding the drug to infectious reporter virus particles that had been produced in the absence of the inhibitor ( Figure S3 ) . Since HCV infection was not decreased , we concluded that addition of U0126 does not prevent HCV cell entry but interferes with production of infectious progeny particles under serum-free conditions . Besides activating transcription factors in the nucleus , ERK1/2 also directly regulate the activity of cytoplasmic enzymes . Therefore , we searched for cellular factors that are regulated by the MAPK/ERK pathway and operate at the ER , the presumed site of HCV particle production . Based on these criteria we focused on PLA2G4A , which is activated by MAPK/ERK-dependent phosphorylation [19] and recruited to the ER by Ca2+ ions [20] , [21] , as a possible candidate host factor that may be responsible for the observed U0126-dependent blockade of HCV particle production . In line with our finding that U0126 only inhibited HCV in the absence of serum , phosphorylation of PLA2G4A was selectively inhibited under these conditions and not affected when serum was present ( Figure 2A ) . To investigate if PLA2G4A activity is relevant for the production of infectious HCV particles we treated Luc-Jc1 transfected cells with pyrrolidine-2 ( Py-2 ) , a specific inhibitor of this type of phospholipase [22] , [23] . Irrespective of culturing these cells in the presence or absence of serum , we observed a strong and dose-dependent inhibition of production of infectious particles resulting in a more than 100-fold reduction of luciferase transduction at 5 and 20 µM of Py-2 , respectively ( Figure 2B ) . Similar to U0126 , RNA replication and cell entry were not affected by Py-2 ( Figure 2B and data not shown ) . Moreover , accumulation of HCV proteins in cells transfected with authentic HCV was not changed by addition of Py-2 ( Figure 2C ) . However , titers of both extracellular as well as intracellular infectivity were strongly impaired by 50- and 10-fold , respectively ( Figure 2D ) . Interestingly , Py-2 also inhibited production of infectious genotype 1A , 3A and 5A HCVcc particles , indicating that PLA2G4A activity is important for HCV virus production across different HCV genotypes ( Figure S4 ) . Intracellular phospholipase A2 enzymes comprise cytosolic , Ca2+-dependent enzymes ( cPLA2 ) as well as the structurally similar Ca2+-independent lipases ( iPLA2 ) [24] . To investigate if iPLA2 activity contributes to HCV particle production we treated Luc-Jc1 transfected cells with bromenol lactone ( BEL ) a specific inhibitor of iPLA2 [25] , [26] . However , BEL neither affected HCV RNA replication nor production of infectious particles ( Figure 2E ) , supporting the notion that specifically the PLA2G4A is involved in the HCV life cycle . To determine whether utilization of PLA2G4A activity is unique to HCV or common to other enveloped viruses , we analyzed production of infectious VSV and DENV in the presence of Py-2 . In case of VSV , a member of the family Rhabdoviridae which assembles infectious virus particles at the plasma membrane [27] , we used a replication competent reporter virus ( designated VSV*MQ ) that expresses a GFP transgene from an additional transcriptional unit placed between the G and L genes [28] . Interestingly , Py-2 treatment of VSV*MQ infected Huh-7 . 5 cells did neither affect intracellular level of GFP ( Figure 3A ) nor production of infectious VSV progeny particles ( Figure 3B ) indicating that unlike for HCV , production of infectious VSV particles did not rely on PLA2G4A activity . In contrast , production of infectious DENV , a relative of HCV from the genus Flavivirus that is thought to assemble at intracellular membranes [29] , was heavily impaired by Py-2 treatment ( Figure 3 ) . Strikingly , like for HCV , RNA replication was not affected ( Figure 3C ) and release of particles was only moderately reduced as is evident from ca . 10-fold lower copy numbers of viral RNA in the culture fluid of Py-2 treated cells compared to mock treated DENV infected cells ( Figure 3D ) . Importantly , when we quantified the infectivity of released particles using a limiting dilution assay we noted an approximately 1 , 000-folder lower infectivity titer for particles produced in the presence of Py-2 as compared to particles produced in the absence of the compound ( Figure 3E ) . Since Py-2 did not grossly inhibit cell entry ( Figure 3F ) we concluded that inhibition of PLA2G4A activity via Py-2 primarily impairs infectivity of released particles ( Figure 3E ) . In summary these results indicate that PLA2G4A is a key host enzyme required for efficient release and high infectivity of HCV and DENV , but not VSV particles . To corroborate our finding that PLA2G4A is involved in production of infectious HCV particles we knocked down expression of the enzyme in HCV transfected cells using RNA interference ( Figure 4 ) . Surprisingly , despite decreased abundance of PLA2G4A in siRNA-transfected cells ( Figure 4A ) , we observed at best a moderate reduction of the total cellular PLA2G4A activity as determined by a commercial enzymatic test ( Figure 4B ) . In line with the result of the enzymatic test , knock down of PLA2G4A did not impede production of infectious HCV particles ( Figure 4C ) . In contrast , reduction of virus titer correlated again with PLA2G4A inhibition upon treatment with Py-2 ( Figures 4B and C ) . To confirm that indeed PLA2G4A was directly contributing to HCV particle production rather than alternative enzymes which may share a similar enzymatic activity , we combined siRNA treatment with application of Py-2 . Under these conditions the reduction of PLA2G4A within cells should increase the susceptibility of HCV to treatment with Py-2 because due to lower abundance of the host factor a lower level of the drug should be sufficient to prevent HCV particle production . As expected , siRNA and Py-2 treatment did not decrease HCV RNA replication ( Figure 4D and E ) . In fact RNA-replication moderately increased in cells that were treated this way compared to cells receiving a scrambled siRNA and no Py-2 ( Figure 4E ) . Despite of this we found significantly lower levels of infectious virus particles secreted from cells receiving the PLA2G4A-specific siRNA and 5 or 10 µM Py-2 as compared to cells treated with the scrambled siRNA and these drug doses ( Figure 4E ) . Collectively , these data indicate that siRNA treatment does not sufficiently suppress PLA2G4A enzyme activity to limit HCV production in Huh-7 . 5 cells . However , when adding the PLA2G4A-specific inhibitor to these cells , knock down of PLA2G4A increased the susceptibility to the drug , arguing that the abundance of enzymatically active PLA2G4A is important for production of infectious particles . Mammals encode genes for more than 30 phospholipase A2s ( PLA2-s ) and related enzymes which are further divided into several classes [24] . These enzymes share the property of hydrolyzing the sn-2 position of membrane glycerophospholipids to release free fatty acids and lysophospholipids . Among PLA2-s only the PLA2G4A displays selectivity for cleaving phospholipids carrying AA at the sn-2 position [30] , [31] . While local release of AA modifies membrane properties including curvature and fluidity [32] , [33] , [34] , this lipid is also a precursor for production of bioactive prostaglandins ( PGs ) and leukotriens ( LTs ) which play essential roles in inflammatory reactions . In fact , production of PGs and LTs is reduced by ca . 90% in PLA2G4A deficient mice highlighting the dominant role of this phospholipase for generation of these lipid mediators [35] , [36] . Given these circumstances we wanted to distinguish if properties of the PLA2G4A-derived cleavage products ( AA and/or lysophosphatidic acid ) directly contribute to HCV particle production , or if these molecules may indirectly promote virus production through activating inflammatory reactions . Since AA is further metabolized by lipoxygenases and cyclooxygenases to yield prostaglandins and leukotrienes we assessed whether inhibition of AA metabolism by application of ( S ) -Flurbiprofen or Nordihydroguaiaretic Acid ( NDGA ) , inhibitors of cyclooxgygenases and lipoxygenases , respectively , prevents efficient HCV particle production . However , neither drug modulated HCV RNA replication or decreased production of HCV particles ( Figure S5 ) . In fact high doses of ( S ) -Flurbiprofen even slightly increased levels of infectious HCV ( Figure S5A ) . These results argue against the notion that AA metabolites and their biological activities are crucial for production of infectious HCV . Next , we analyzed if application of AA or related fatty acids restores production of HCV particles in the presence of the PLA2G4A inhibitor Py-2 . Remarkably , we observed a pronounced and dose-dependent restoration of infectious particle production in the presence of AA ( Figure 5A ) . Importantly , 5 , 8 , 11 , 14-Eicosatetraynoic acid ( ETYA ) a derivative of AA with four triple bonds at positions 5 , 8 , 11 , and 14 of the acyl backbone did not restore virus production ( Figure 5B ) . Likewise further natural fatty acids with 1 , 2 , 3 or 4 double bonds at various positions did not relieve the blockade of virus production caused by addition of the PLA2G4A inhibitor ( Figure S6 ) . Among all tested lipids only AA itself and to a moderate level 5 , 6-dehydro AA ( 5 , 6-DHA ) restored virus production ( Figure 5A and C ) . Notably , AA did not increase HCV cell entry since the infectivity of particles produced in the absence of the drug was not stimulated by addition of AA during cell entry ( Figure S7 ) . Next , we investigated if repression of DENV infectious particle production is also relieved by addition of AA . As is depicted in Figure S8 , we noted a trend that high doses of AA partially restore production of infectious DENV progeny in the presence of Py-2 . However , the rescue of infectious virus production was moderate and not statistically significant . To investigate if in the context of HCV , addition of AA truly rescues the blockade of PLA2G4A enzymatic activity and not simply over-stimulates production of HCV particles , we applied AA to HCV transfected cells in the absence of Py-2 . Interestingly , under these conditions we observed a moderate increase of HCV infectivity ( Figure 5D ) . This finding mirrors the moderate gain of infectivity when cells were treated with Flurbiprofen that prevents AA metabolism . Since both treatments are expected to increase availability of AA , these data suggesting that availability of AA may be sub-saturating in non-treated HCV transfected cells . Collectively , these results indicate that specific properties of AA , which is created by cleavage of glycerophospholipids carrying this lipid at the sn-2 position by PLA2G4A , are important for production of highly infectious HCV progeny particles . Pyrrolidine has been described as precursor for compounds against the NS3-protease and RNA-polymerase of HCV [37] . As the PLA2G4A-specific inhibitor Py-2 shares a heterocyclic ring with pyrrolidine we wanted to exclude that Py-2 may prevent virus production indirectly by inhibiting the viral protease or polymerase . Both enzymes contribute to active RNA replication complexes and may be required to feed in newly synthesized viral RNA into assembling virus particles . To address this , we treated Jc1-transfected cells with a polymerase inhibitor ( 2′-C-methyladenosine; 2′CMA ) , a protease inhibitor ( boceprevir ) , with Py-2 , or with quinidine , the latter being a class I antiarrhythmic drug recently found to inhibit production of infectious HCV [38] . As expected , only the RNA-replication inhibitors ( 2′CMA and boceprevir ) reduced abundance of HCV RNA in transfected cells ( Figure S9A ) dose dependently . While at the used doses ( well beyond the IC90 for all compounds ) all drugs moderately impaired release of HCV core protein ( 2–5-fold ) , only Py-2 and quinidine strongly reduced infectivity of particles ( Figure S9C ) to levels more than 20-fold lower compared to the DMSO control . Thus , it is unlikely that indirect effects of Py-2 on RNA-replication are responsible for the reduced amount of secreted particle and their impaired infectivity . Rather these findings argue that Py-2 directly interferes with HCV assembly and the infectivity of released particles . To investigate how Py-2 impairs HCV assembly , we investigated the subcellular localization of core , and PLA2G4A in the presence or absence of Py-2 . Adipose differentiation-related protein ( ADRP ) , a host protein interacting with the surface of LDs was stained in parallel . Since we were unable to detect endogenous PLA2G4A with commercial antibodies , we created a stable Huh-7 . 5 cell line ectopically expressing a GFP-tagged PLA2G4A protein . As is shown in Figure S10 we did not see gross changes of the distribution of these proteins during the short term Py-2 treatment . Moreover , localization of GFP-PLA2G4A did not differ between cells expressing HCV proteins and those cells that were not positive for HCV . These findings provide preliminary evidence that localization of epitope tagged PLA2G4A is not influenced by HCV . More work , ideally with untagged PLA2G4A will be needed to fully resolve the localization and trafficking of this protein in the presence or absence of HCV and Py-2 . Next we assessed the amount of intracellular core protein that is resistant to proteolysis by proteinase K . Reasoning that core protein that is surrounded by a membrane should be protected from digestion by the protease , this assay estimates the number of core protein that has acquired a lipid envelope . Since among members of the family flaviviridae virus particle envelopment depends on expression of functional glycoproteins and as for HCV deletion of E1-E2 abrogates production of infectious progeny [39] , we used a Jc1 mutant carrying a deletion of E1-E2 genes as a control and reference . As expected , when the protease was added to cell lysates prepared by repetitive cycles of freeze together with detergent ( Triton X-100 ) , the viral protein was completely degraded ( Figure 6A ) . However , when cell lysates were incubated with the protease in the absence of detergent , a substantial amount of core protein resisted digestion indicating protection by a membrane envelope . Notably , the amount of protected core protein was approximately 3-fold lower in Py-2 treated compared to DMSO treated Jc1 transfected cells , resembling the phenotype of cells transfected with Jc1ΔE1E2 ( Figure 6A and B ) , arguing that Py-2 had decreased the amount of enveloped core protein structures . Since trafficking of core protein to lipid droplets ( LDs ) [6] is crucial for assembly of infectious progeny particles we analyzed the influence of PLA2G4A on the accumulation of core protein on LDs . To this end , we prepared LDs from Jc1-transfected and Py-2 treated cells and analyzed the abundance of core protein on the surface of this cellular organelles . The quality of our LD preparation was monitored by detection of actin ( cytosol ) , ADRP , calreticulin ( ER ) and Golgi matrix protein ( Golgi ) in total lysates and the LD fraction ( Figure 6C ) . Importantly , calreticulin and Golgi matrix protein were below the detection limit of our assay in the LD fraction whereas ADRP was readily detected thus confirming that our procedure successfully enriched cellular LDs . Notably , Py-2 moderately increased the total cellular level of core protein but at the same time reduced the abundance of core in the LD fraction of the cell lysate . This difference evident by western blot was further confirmed using a core-specific ELISA demonstrating a ca . 2-fold higher core protein amount in the lysate of Py-2-treated cells and about 3-fold lower levels in the LD fraction . Collectively , these data indicate that Py-2-dependent inhibition of PLA2G4A impedes association of core with LDs which in turn may limit core protein envelopment and particle release . In principle inhibition of PLA2G4A by Py-2 may reduce infectivity by preventing assembly/release of particles and/or by altering particle properties including the association with lipoproteins . Therefore , we used ultracentrifugation of HCV particles through density gradients to analyze the amount and density of virus particles released from cells treated in the presence or absence of Py-2 ( Figure 7 ) . Using this approach we noted that the distribution of HCV core protein-containing structures throughout the density gradient was essentially unchanged with peak core protein levels in fractions with a density of 1 . 11–1 . 18 g/mL irrespective of Py-2 treatment ( Figure 7A ) . Notably , the overall amount of released core protein was moderately reduced ( ca . 2–3-fold; Figure 7A and C ) when particles were produced in the presence of the PLA2G4A inhibitor . Most strikingly , inhibitor treatment heavily decreased the infectivity of released HCV particles by 50- to 100-fold ( Figure 7A ) . To elucidate , why particles produced in the presence of Py-2 are less infectious , we analyzed their protein composition . To this end we transfected Jc1 carrying a FLAG-epitope tag at the N-terminus of E2 [13] into Huh-7 . 5 cells where endogenous ApoE was silenced by a specific shRNA and replaced by ectopic expression of an HA-tagged , shRNA resistant , human ApoE gene ( Figure S11 ) . This approach enabled us to monitor co-precipitation of viral and host factors with an ApoE-specific or a virus envelope-specific antibody . Notably , treatment of transfected cells with Py-2 reduced the amount of secreted ApoE and core protein by ca . 25% and 65% , respectively ( Figure 7B and 7C , respectively ) . Therefore , we normalized the culture fluids to equal quantities of ApoE or core protein before the ApoE-specific or FLAG-specific co-precipitation . Remarkably , Py-2 treatment reduced the level of core protein co-precipitating with ApoE by 5-fold ( Figure 7B ) . Likewise , inhibition of PLA2G4 lowered the association of ApoE and core with the FLAG-tagged E2 by ca . 5-fold ( Figure 7C ) . In summary , inhibition of MAPK-dependent PLA2G4A activity by Py-2 moderately decreased the titer of released HCV particles but heavily impaired infectivity of both intracellular and extracellular particles likely through gross changes of their protein composition .
In this study we manipulated key cellular signaling cascades to identify host cellular HCV dependency factors . We report that blockade of the MAPK/ERK cascade by a well-established specific inhibitor ( U0126 ) potently repressed production of infectious HCV progeny ( Figure 1 ) . Making reasonable assumptions ( activation by ERK , function at the ER membrane ) , we focused on PLA2G4A , an ERK-regulated host enzyme that is recruited to the ER-membrane by Ca2+ , as possible new HCV dependency factor for HCV assembly . Our further experiments provide three lines of evidence supporting our conclusion that the ERK-dependent activation of PLA2G4A is crucial for production of infectious progeny: First , we show that Py-2 impedes production of infectious HCV in a dose-dependent fashion ( Figure 2 ) . Notably , phospholipase A2 enzymes are subdivided into several classes including secreted PLA2s ( sPLA2s ) , Ca2+-dependent cytosolic ( cPLA2s ) , CA2+-independent cytosolic ( iPLA2s ) , platelet-activating factor acetylhydrolases ( PAF-AHs ) , lysosomal PLA2s and the most recently identified adipose-specific PLA [24] . Importantly , Py-2 potently inhibits PLA2G4A ( the α isoform of the Ca2+-dependent cytosolic PLA2s , also named cPLA2α ) in various in vitro assays [22] . In contrast it interferes with PLA2G4B ( cPLA2β ) and PLA2G4C ( cPLA2γ ) only at very high doses , probably by a non-specific mechanism , and it does not inhibit the secreted sPLA2s [22] . Congruently , BEL , a “suicide substrate” and specific inhibitor of iPLA2s with a >1 , 000-fold selectivity for iPLA2s over cPLA2s [26] , did not impede HCV particle production ( Figure 2 ) . Second , using RNA interference we show that reduction of the abundance of PLA2G4A enzyme increased susceptibility of HCV to inhibition by Py-2 ( Figure 4 ) . Surprisingly , we did not observe an influence of the knock down of PLA2G4A on HCV particle production ( Figure 4 ) . However , we note that in spite of clearly reduced abundance of the enzyme in siRNA-treated cells we measured only a small decline of PLA2G4A enzyme activity . It is possible that the active , phosphorylated PLA2G4A enzyme has a relatively long half-life which may preclude reduction of the active enzyme to a level that limits HCV infectious particle production under these experimental conditions . Third , we observed an almost complete restoration of HCV infectivity upon supplementing Py-2 treated cells with AA ( Figure 5 ) . Importantly , among all PLA2 enzymes , only the PLA2G4A displays a preference for cleaving glycerophospholipids carrying the polyunsaturated AA at the sn2-position [24] . Moreover , related fatty acids including ETYA which differs from AA only by triple-bonded C-atoms in place of the double-bonded C-atoms in AA , were unable to restore virus production ( Figure 5 and S7 ) . Even 5′6-DHA ( with a single triple-bonded C atom ) only partially compensated production of infectious HCV in the presence of Py-2 indicating that highly specific properties of AA , the cleavage product of PLA2G4A , are crucial for production of infectious HCV progeny . AA is the precursor for biologically active lipid mediators including prostaglandins , thromboxane and leukotrienes collectively termed eicosanoids . These molecules are synthesized from AA through the cyclooxygenase and lipoxygenase pathways and play important roles during inflammation . However , since inhibitors of both pathways of AA metabolism did not impede production of infectious HCV particles ( Figure S6 ) , we believe that the properties of AA itself rather than indirect effects caused by AA-metabolites are important for HCV . Our data support the conclusion that PLA2G4A activity is relevant for HCV assembly in two principal ways . First , inhibition of PLA2G4A decreased the amount of core protein associated with lipid droplets and reduced the level of core protein that is protected from proteolytic digestion ( Figure 6 ) . The latter finding may indicate a lower level of intracellular core protein that is encased in membranes – possibly within virus particles – and therefore protected from proteolysis . Moreover , we observed reduced levels of extracellular HCV particles ( 2–3-fold; Figure 7 ) . It is currently unclear why inhibition of PLA2G4A reduces the level of core protein at LDs . In principal several mechanisms account for this including an increase of core assembly and subsequent unloading from LDs or a reduced trafficking of core to LDs possibly due to aberrant processing of the protein by signal peptide peptidase cleavage . However , since we observed lower levels of secreted virus particles we consider it unlikely that increased assembly and protein unloading from LDs is responsible . Moreover we did not detect an overt processing defect of core in the presence of Py-2 ( Figure 2 and 6 ) . Notably , Gubern et al . have shown that MAPK-dependent phosphorylation of PLA2G4A at Ser505 is necessary for biogenesis of lipid droplets [40] , [41] . Therefore , it is tempting to speculate that blockade of PLA2G4A by Py-2 reduces lipid droplet biogenesis , thus limiting abundance of core protein at these organelles which are essential for HCV particle production [6] . Assuming that core protein has to be unloaded from LDs to drive budding and virus production which is consistent with the recent findings of Counihan et al . [42] , it is reasonable to suggest that reduced core protein levels at LDs may decrease membrane envelopment and particle release . While AA ( the product of the PLA2G4A-catalzed triglyceride cleavage ) is apparently not required for lipid droplet biogenesis [40] , it nevertheless seems to be essential for the second prominent influence of PLA2G4A on HCV , i . e . the production of highly infectious HCV progeny . Remarkably , both HCV and DENV produced and released in the presence of Py-2 were approximately 100-fold less infectious as compared to viruses assembled in the absence of the drug ( Figures 2 and 3 ) . Since addition of Py-2 to particles generated in the absence of the drug , did not impede cell entry ( data not shown and Figure 3 ) , we exclude that Py-2 interferes with cell entry of HCV or DENV . Rather , our results indicate that particle properties are altered when Py-2 is present . While the density spectrum of released HCV particles was unchanged , co-precipitation with ApoE- or envelope-specific antibodies provide firm evidence that blockade of PLA2G4A disturbs the composition/structure of released HCV particles . Specifically , we noted 5-fold reduced levels of core co-precipitating with anti-ApoE and likewise 5-fold lower amounts of ApoE and core co-precipitating with the envelope protein-specific pull down . These results argue that either less particles are directly associated with ApoE or that particles contain lower levels of ApoE . Since ApoE is important for infectivity of HCV particles [43] , [44] , a defect in the loading of this protein onto HCV particles may explain their reduced infectivity . In the envelope-specific pull down mediated by the FLAG-tagged viral E2 protein , we observed both 5-fold lower ApoE and also core protein . On one hand this may indicate lower abundance of both proteins in secreted enveloped HCV particles . On the other hand , this may reflect incorporation of lower numbers of glycoprotein complexes into the viral envelope and in turn reduced precipitation efficiency . Unfortunately , due to insufficient sensitivity of currently available envelope protein detection systems , we are currently unable to distinguish between these two possibilities . Nevertheless , these results argue that blockade of PLA2G4A by Py-2 prevents normal loading of viral ( core and envelope proteins E1 E2 ) and host proteins ( ApoE ) onto nascent HCV particles . As a consequence , virus attachment or the interaction with entry factors or membrane fusion might be inefficient . Notably , recent evidence suggests that AA and other poly-unsaturated fatty acids increase membrane fluidity [32] , [33] , [34] . Thus , PLA2G4A activity in the vicinity of particle production may modify membrane characteristics including curvature and fluidity . These altered properties may disturb virus budding and/or the trafficking of viral and host-derived components to the site of virus assembly and thus result in the production of particles with disturbed stoichiometry , aberrant envelope composition , and poor infectivity . A more detailed proteomic and lipidomic comparison between particles produced in the presence or absence of Py-2 should help to clarify this in the future . Our finding that HCV and DENV particle assembly depend on PLA2G4A activity while VSV apparently does not rely on this host factor indicates fundamental differences between the assembly of enveloped VSV particles compared to HCV and DENV . It remains to be shown how exactly PLA2G4A contributes to production and release of infectious DENV particles . While it has been reported that DENV may also usurp LDs for its assembly [45] , Welsch and colleagues showed that DENV assembly sites are physically linked to RNA replication sites on modified ER structures [29] . Unlike for HCV , AA did not consistently restore virus production of DENV supporting the notion that PLA2G4A may participate in HCV and DENV morphogenesis through different mechanisms . More work will be needed to fully understand the role of this host factor for these two viruses . Along these lines it will be interesting to analyze if other enveloped viruses ( e . g . Coronaviruses ) that like HCV and DENV assemble progeny particles at intracellular membranes depend on PLA2G4A as well . Such studies could in the future reveal common replication mechanisms between HCV and DENV ( and possibly other viruses ) that may provide valuable insights into conserved assembly pathways of enveloped viruses . Finally , inhibitors of the PLA2G4A which have been pursued and brought into clinical development for treating inflammatory diseases may prove useful as antiviral therapeutics for the treatment of chronic HCV infection and possibly other viral diseases .
Antibodies were obtained from the following companies: Apolipoprotein B ( Millipore ) ; PLA2G4A ( Abcam ) ; P ( S505 ) - PLA2G4A , ERK1/2 , P-ERK1/2 ( Cell Signaling ) ; Actin ( Sigma-Aldrich ) ; ADRP ( Probiogen ) ; Golgi Matrix Protein ( Epitomics ) ; Calreticulin ( Stressgen ) ; HA-Epitope ( Covance ) . Anti-NS5A antibodies ( 9E10 mouse monoclonal antibody ) were kindly provided by C . Rice ( Rockefeller University , New York ) , anti-Core ( C7-50 mouse monoclonal antibody ) was a kind gift of D . Moradpour ( Centre Hospitalier Universitaire Vaudois , Lausanne ) , and sheep polyclonal antibodies against ADRP were kindly provided by J . McLauchlan ( MRC Virology Unit , Institute of Virology , Glasgow , UK ) . Secondary antibodies were ordered from Life Technologies . 2′CMA was kindly provided by T . Tellinghuisen ( Department of Infectology , The Scripps Research Institute , Florida ) . Reagents were ordered from various manufacturers: Pyrrolidine-2 ( CAS Registry No . : 337307-06-9; Order ID: 525143 , Merck Chemicals; U0126 , PD98059 ( Cell Signaling ) ; Sorafenib ( BAY 43-9006 , Alexa Biochemicals ) ; Proteinase K ( Roche ) ; cPLA2α-siRNA ( Thermo ) ; ETYA; 5 , 6-dehydro arachidonic acid; ( S ) -Bromoenol lactone; NDGA ( Cayman ) ; Quinidine; arachidonic acid ( Sigma ) ; FLAG-agarose ( Sigma ) . Huh-7 . 5 cells were grown in Dulbecco's modified minimal essential medium ( DMEM; Life Technologies ) supplemented with 2 mM L-glutamine , nonessential amino acids , 100 U/ml of penicillin , 100 µg/ml of streptomycin , and 10% fetal calf serum ( DMEM w/FCS ) . The plasmids pFK-Luc-Jc1 and pFK-Jc1 , encoding the genotype 2a/2a chimera Jc1 with or without the firefly luciferase reporter gene have been described [14] , [18] . Chimeric HCV constructs pFK-H77/C3 encoding the genotype 1a/2a chimera [18] , pS52/JFH1 ( A4550C ) encoding the genotype 3a/2a chimera [46] , and pSA13/JFH1 ( C3405G-A3696G ) encoding the genotype 5a/2a chimera have been described [47] . For the shRNA-mediated knockdown of ApoE expression , the vector pLenti-3′-U6-EC-EP7 [48] which contains a blasticidine resistance gene was modified to harbor an shRNA specific to the 3′-untranslated region of human ApoE ( 5′-GCCGAAGCCTGCAGCCATGCG-3′ ) . As control , a construct containing a non-targeting shRNA was used . The lentiviral plasmid pWPI hApoE-Linker-HA-Gun encodes the human ApoE3 variant with an HA tag added to the 3′-end via a Glycine-Glycine-Serine-Glycine linker in the self-inactivating pWPI vector [49] which comprises a GFP-ubiquitin-neomycinphosphotransferase fusion protein as selectable marker . The gene encoding PLA2G4A ( Thermo Scientific , cDNA clone MHS4426 ) was N-terminally fused to a GFP-tag and subcloned into the pWPI vector . The creation of a Huh-7 . 5-cell line expressing GFP-PLA2G4A fusion protein is described below . Finally , pFK-Jc1-ΔE1-E2 was created by a PCR-based strategy . In this construct the entire E1 and E2 coding region is deleted and the core coding region is directly fused in frame to the coding region of p7 . Sequence information is available upon request . HCVcc particles and firefly luciferase HCV reporter virus were generated as reported previously [50] . Luciferase reporter virus infection assays were carried out as described [50] . For inhibitor assays , cells were pre-treated with cellular or viral inhibitors to completely abolish enzyme activity 41 h after transfection of HCV-RNA . One hour later , supernatants were removed and fresh medium with inhibitors was added to harvest newly synthesized virus in a period of six hours . At 48 hpt , supernatants and cell lysates were used for the infection of naïve Huh-7 . 5 cells or subjected to assays as described below . Lipid droplets were isolated according to a published protocol [51] with minor modifications . Briefly , Jc1-transfected cells of 10×100-mm plates were scraped into 50 ml PBS and pelleted by centrifugation at 1000×g . Cells were resuspended in ice-cold HLM buffer ( 20 mM Tris-HCl ( pH 7 . 4 ) ; 1 mM EDTA ) with protease inhibitors ( complete Mini; Roche ) and incubated for 10 min on ice . Cells were homogenized by eight gentle strokes with a Potter-Elvehjem tissue homogenizer and nuclei were removed from lysates by low-speed centrifugation ( 1000×g ) . Density of the post-nuclear supernatant was adjusted with sucrose ( 20% final ) and samples were loaded below a discontinuous sucrose gradient ( 0 , 5 , 20% ) . Flotation of lipid droplets through the gradient was achieved by centrifugation at 28 , 000×g for 30 min , 4°C . The white band containing lipid droplets at the top of gradient was collected and proteins were characterized by immunoblotting or core ELISA . Confluent Jc1- transfected cells were suspended in 170 µl PK buffer ( 50 mM Tris-HCl ( pH 8 . 0 ) ; 10 mM CaCl2; 1 mM DTT ) and homogenized by five freeze-and-thaw cycles . The lysate was divided into three 50 µl fractions and treated with or without 50 µg/ml proteinase K for one hour on ice . As control , the third sample additionally containted 5% ( v/v ) triton X-100 during protease digestion . The reactions were stopped by 5 mM PMSF for 10 min on ice and addition of Laemmli buffer . Samples were analysed by immunoblotting . Huh-7 . 5 cells were transfected with siRNAs specific to PLA2G4A ( D-009886-01 , -02; Thermo ) or scramble siRNA ( 4390846; Ambion ) in a forward transfection procedure according to the manufacture's protocol ( RNAiMax; Life Technologies ) . To achieve an efficient knock-down , cells were transfected twice within 96 h and re-seeded in 6-wells at a density of 2 . 5×105 cells per well . Cells were infected with a 20-fold concentrated stock of Luc-Jc1 virus and 41 h later the inhibitor assay was performed as described above . The efficiency of PLA2G4A knock-down was verified by immunoblotting . Viral RNA was prepared from infected cells using a Nucleo Spin RNAII kit ( Macherey-Nagel ) according to the manual's instructions . 5 µL of the RNA sample was used for HCV-specific quantitative reverse transcription-PCR ( qRT-PCR ) analysis using a LightCycler 480 device ( Roche ) . HCV-specific qRT-PCRs were conducted in duplicate measurements as published [52] utilizing a one-step RT-PCR LightCycler 480 RNA Master Hydrolysis Probes kit ( Roche ) and the following HCV-specific probe ( Molecular Biosystems ) and primers ( MWG-Biotech ) : HCVMGB2 [5′-6FAM ( carboxy fluoresceine ) -CACGGCTAGCTGTG-MGB-3′]; XTF5 ( 5′-GTGGCTCCATCTTAGCCCTAGT-3′ ) ; and HCMgR2 ( 5′-TGCGGCTCACGGACCTTT-3′ ) . To normalize for equal quantities of total RNA in the samples , the GAPDH-specific mRNA was detected in parallel employing GAPDH-specific oligonucleotides ( S-GAPDH , 5′-GAAGGTGAAGGTCGGAGTC-3′; A-GAPDH , 5′-GAAGATGGTGATGGG ATTTC-3′ ) and a GAPDH-specific probe ( TIB Molbiol ) , 640-GAPDH-BBQ probe ( 5′-LC640-CAAGCTTCCCGTTCTCAGCCT-BBQ-3′ ) . Reactions were performed in three stages by using the following conditions: stage 1 ( RT ) , 3 min at 63°C; stage 2 ( initial denaturation ) , 30 s at 95°C; stage 3 ( amplification ) , 45 cycles of 10 s at 94°C and 20 s at 58°C . The amount of HCV RNA was calculated by comparison to serially diluted in vitro transcripts and normalized to the amount of GAPDH , which served as a housekeeping gene . HCV Core protein within cell lysates and culture fluids was quantified with a commercially available diagnostic kit ( Architect Anti-HCV; Abbott ) . One µg of total RNA or 1/5 of RNA extracted from 100 µl cell culture supernatant was reverse transcribed into cDNA using the High Capacity cDNA Reverse Transcription Kit ( Applied Biosystems ) following the manufacturer's protocol . Quantitative RT-PCR was performed on an ABI PRISM 7000 Sequence Detection System ( Applied Biosystems ) . The reaction was carried out in a final volume of 15 µl , including 7 . 5 µl 2× Green DYE master mix ( P . J . K . , Kleinblittersdorf ) , 2 µl cDNA template , 1 . 5 µl primer mix ( 5 µM each ) , 4 µl RNase-free sterile water . Reactions were carried out using the following settings: 95°C: 10 min→40× [95°C: 30 sec→55°C: 60 sec→72°C: 60 sec] . The amounts of DENV RNA were calculated from a standard curve derived from serially diluted in vitro transcripts of known concentration . Primers used for the amplification were: sDV2-9687 , 5′-GCCCTTCTGTTCACACCATT-3′ and asDV2-9855 , 5′-CCACATTTGGGCGTAAGACT-3′ . To quantify core protein , cell culture supernatants or immunoprecipitated FLAG-Jc1 particles were diluted in PBS/1% Triton in a ratio of 1∶30 . The core ELISA was performed with a commercially available diagnostic kit ( Architect Anti-HCV , Abbott ) . ApoE was quantified according to the manufacturer's instructions ( MabTech ) . Density gradient centrifugation was performed as described recently [53] . Briefly , viruses were separated by overnight centrifugation through a 0% to 40% iodixanol step gradient at 154 , 000×g . Ten fractions of 1 ml were collected from the bottom and analyzed for virus infectivity , core protein levels , and viral RNA copies . Following Py-2 treatment , aliquots of cell culture supernatants were subjected to core or ApoE ELISA in order to equilibrate the volumes for immunoprecipitations . To capture FLAG-Jc1 particles or HA-ApoE , equilibrated supernatants were mixed with either 20 µl FLAG-agarose or 3 µg anti-HA antibody and 30 µl G-protein agarose ( Roche ) overnight at 4°C in gentle rotation . Immunoprecipitated proteins were washed three times in PBS , eluted with 50 µl PBS/1% Triton for 10 min at 50°C and analyzed by core ELISA or immunoblotting . Phospholipase A2 activity was measured according to the manufacture's protocol ( EnzChek Phospholipase A2 kit; Life Technologies ) . In brief , Huh-7 . 5 cells were harvested from 35-mm wells , resuspended in 200 µl EnzChek PLA2 reaction buffer and disrupted by sonication . To avoid any measurement of iPLA2 activity , Bromenol lactone was added to all samples at a concentration of 5 µM . Liposomes were prepared with the EnzChek Phospholipase A2 substrate and mixed with lysates at a ratio of 1∶1 to give a total volume of 100 µl . Samples were transferred in 96-wells and PLA2G4A activity was monitored by the intensity increase of a single wavelength at 515 nm in a fluorescence microplate reader ( FLx800; BioTek ) . For the generation of stable Huh-7 . 5-HA-ApoE cells , lentiviral gene transfer was used as described before [54] . Endogenous ApoE expression in Huh-7 . 5 cells was silenced using pLenti-3′-U6-EC-EP7 [48] which contains a blasticidine resistance gene and an shRNA specific to the 3′-untranslated region of human ApoE ( 5′-GCCGAAGCCTGCAGCCATGCG-3′ ) . Subsequently , ApoE expression was restored by transduction with pWPI hApoE-Linker-HA-Gun described above . Lentiviral particles were generated by transfection of pCMV ΔR . 74 , pcz VSV-G and a derivative of either pLenti-3′-U6-EC-EP7 or pWPI at a ratio of 3∶1∶3 into 293T cells . Lentiviral particles were collected 48 h post transfection and used to transduce target cells . Selection was carried out in the presence of 5 µg/ml Blasticidine or 0 . 75 mg/ml G418 . For generation of Huh-7 . 5-GFP-PLA2GA4 cells , Huh-7 . 5 cells were transduced with lentiviruses carrying the pWPI-GFP_PLA2G4A vector . Transduced cells were selected in the presence of 5 µg/mL Blasticidine . The protocol for immunostaining was carried out as described previously ( Frenzen , Hueging , Steinmann , PLoS Pathogens April 2011 Volume 7 Issue 4 e1002029 ) . Immunostainings of Core and ADRP proteins were performed at dilutions of 1∶7000 respectively 1∶500 . Texas Red and Cy-5 secondary antibodies were used at dilutions of 1∶1000 . Statistical data analysis was performed using the free statistical environment R . Data were initially visualized using histograms , boxplots and QQ-plots , and normality of the distributions was assessed . Statistical significance of differences was then calculated using Welch's t-test if data were sufficiently well approximated by a normal distribution , or using the Wilcoxon rank sum test as a non-parametric alternative for non-normal data . P-values were calculated and statistical significance reported as highly significant ( *** ) if p≤0 . 01 , significant ( ** ) if p≤0 . 05 , and marginally significant ( * ) if p≤0 . 1 . Differences were considered not significant ( n . s . ) for p>0 . 1 . | The human genome encodes more than 30 phospholipase A2s . These enzymes cleave fatty acids at the C2 atom of phosphoglycerides and thus modulate membrane properties . Among all PLA2s only PLA2G4A , which is recruited to perinuclear membranes by Ca2+ and activated by extracellular stimuli via the mitogen activated protein kinase pathway , specifically cleaves lipids with arachidonic acid . Metabolism of arachidonic acid yields prostaglandins and leukotriens , important lipid mediators of inflammation . We show that inhibition of PLA2G4A produces aberrant HCV particles and that infectivity is rescued by addition of arachidonic acid . Our results suggest that a specific lipid ( arachidonic acid ) is essential for production of highly infectious HCV progeny , likely by creating a membrane environment conducive for efficient incorporation of crucial host and viral factors into the lipid envelope of nascent particles . Strikingly , PLA2G4A is also essential for production of highly infectious Dengue Virus ( DENV ) particles but not for vesicular stomatitis virus ( VSV ) . These observations argue that HCV and DENV which unlike VSV produce particles at intracellular membranes usurp a common host factor ( PLA2G4A ) for assembly of highly infectious progeny . These findings open new perspectives for antiviral intervention and highlight thus far unrecognized parallels in the assembly pathway of HCV and DENV . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"dengue",
"hepatitis",
"gastroenterology",
"and",
"hepatology",
"lipid",
"signaling",
"infectious",
"disease",
"control",
"mapk",
"signaling",
"cascades",
"erk",
"signaling",
"cascade",
"hepatitis",
"c",
"infectious",
"diseases",
"signaling",
"in",
"cellular",
"processes",
"membranes",
"and",
"sorting",
"biology",
"infectious",
"disease",
"modeling",
"molecular",
"biology",
"signal",
"transduction",
"cell",
"biology",
"viral",
"diseases",
"molecular",
"cell",
"biology",
"signaling",
"cascades"
]
| 2012 | MAP-Kinase Regulated Cytosolic Phospholipase A2 Activity Is Essential for Production of Infectious Hepatitis C Virus Particles |
Danforth's short tail ( Sd ) is a semidominant mutation on mouse chromosome 2 , characterized by spinal defects , urogenital defects , and anorectal malformations . However , the gene responsible for the Sd phenotype was unknown . In this study , we identified the molecular basis of the Sd mutation . By positional cloning , we identified the insertion of an early transposon in the Sd candidate locus approximately 12-kb upstream of Ptf1a . We found that insertion of the transposon caused overexpression of three neighboring genes , Gm13344 , Gm13336 , and Ptf1a , in Sd mutant embryos and that the Sd phenotype was not caused by disruption of an as-yet-unknown gene in the candidate locus . Using multiple knockout and knock-in mouse models , we demonstrated that misexpression of Ptf1a , but not of Gm13344 or Gm13336 , in the notochord , hindgut , cloaca , and mesonephros was sufficient to replicate the Sd phenotype . The ectopic expression of Ptf1a in the caudal embryo resulted in attenuated expression of Cdx2 and its downstream target genes T , Wnt3a , and Cyp26a1; we conclude that this is the molecular basis of the Sd phenotype . Analysis of Sd mutant mice will provide insight into the development of the spinal column , anus , and kidney .
Danforth's short tail ( Sd ) is a semidominant spontaneous mutant mouse characterized by severe spinal defects , urogenital defects , and anorectal malformations [1] , [2] , [3] . Heterozygous and homozygous Sd animals display a broad range of abnormalities in the vertebral column , including reduction or absence of the dens axis , reduction of all vertebral bodies in the dorsoventral axis , split vertebrae , and truncation of the caudal vertebral column [4] , [5] , [6] . The vertebral columns of Sd/Sd and Sd/+ mice are usually truncated at the seventh thoracic and the sixth caudal vertebral body , respectively [7] . The urogenital system in Sd heterozygotes may display malformations ranging from displaced to missing kidneys . Homozygotes invariably have missing or severely malformed and dislocated kidneys . The rectum and anal opening are missing , and the embryonic cloaca persists . Homozygous animals die within 24 h after birth [4] . Although Sd is known to map to mouse chromosome 2 , little is known about the molecular nature of the mutation . Double mutants between the Sd and undulated ( un ) alleles showed reduced expression of Pax1 and enhancement of the vertebral malformations [8] . Pax1 expression is regulated by signals from the notochord [9] , [10] , thus providing a potential molecular link for the interaction between un and Sd . Zachgo et al . obtained a lacZ enhancer trap insertion called Etl4lacZ , which is tightly linked to Sd . If Etl4lacZ is present in trans ( i . e . , on the chromosome that is wild type ( WT ) for Sd ) , the Sd phenotype is enhanced [11] . In contrast , if Etl4lacZ is present in cis ( i . e . , on the same chromosome as Sd ) , the phenotype is attenuated , suggesting a direct interaction of the transgene insertion with the Sd gene at the DNA level . However , neither the Sd mutation nor the Sd gene is known [12] , [13] . We previously obtained a mutant mouse line , SktGt , through gene-trap mutagenesis , and identified the Skt gene . We found that the SktGt locus was located 0 . 95 cM distal to the Sd locus , and that the Etl4lacZ site was located in the third intron of the Skt gene [14] . Because the Sd region had been shown to be located 0 . 15–0 . 3 cM distal to the marker D2Mit362 [13] , [15] and 0 . 75 cM proximal to Etl4lacZ [11] , it was clear that the Sd locus is located between D2Mit362 and the Skt gene . In this study , we identified the cause of the Sd mutation by generating an Sd-derived cosmid contig of approximately 0 . 5-Mb around the Sd locus and using it to guide the production of genetically engineered mice with particular genetic alterations .
Although the Sd mouse was identified in 1940 , detailed histological findings have not yet been fully described . In this study , we mainly analyzed three tissues—the vertebrae , urogenital tract , and kidney—because characteristic features are found in these tissues of Sd mice . Sd/Sd homozygotes had similar , but much more severe , abnormalities than Sd/+ heterozygotes in terms of truncation of the vertebrae at day 0 postpartum ( Figure 1A , Text S1 ) , defects of the nucleus pulposus in the intervertebral discs , anorectal malformations , and renal hypoplasia/agenesis ( Figure 1B ) . We also revealed hypoplasia of the dens ( Figure 1C ) and sacral hypoplasia ( Figure 1D , 1E ) by high-resolution computed tomography ( Text S1 ) . Thus , Sd is considered a mouse model for caudal regression syndrome ( CRS ) , characterized by vertebral , anorectal , and urogenital abnormalities . We also carried out a lung-floating test to analyze the cause of death . As shown in Figure 1F , the lungs of Sd/Sd neonates sank in water . Histological sections of the lung revealed atelectasis of the lung ( Figure 1G ) , indicating that no breathing occurred after birth . The Sd locus had been shown to be located 0 . 15–0 . 3 cM distal to D2Mit362 and 0 . 75 cM proximal to Etl4lacZ [11] , [13] , [15] . We demonstrated that the insertion site of lacZ in the Etl4lacZ mutation was within the third intron of Skt [14]; thus , it was clear that Sd is located between D2Mit362 and Skt . Furthermore , we showed that the genetic distance between SktGt and Sd was 0 . 95 cM , and hence that Skt was genetically separated from Sd [14] . We created a cosmid library using embryonic day ( E ) 11 . 5 homozygous Sd/Sd embryos . We screened this cosmid library with 32 different DNA probes , obtaining 28 cosmid clones within the Sd region between D2Mit362 and SktGt . Based on physical mapping of 19 cosmid clones and 25 PCR products , the assembled contig spans a 542-kb region containing Sd ( Figure S1 ) . The region spanned by the contig contained three known genes , Ptf1a , Msrb2 , and Skt , and five expressed sequence tags of unknown function: Gm13344 , Gm13336 , 4921504 , E06Rik , and Otud1 ( Figure S1 , Text S1 ) . Interestingly , we found one cosmid clone for which one Not I digestion product was longer than that expected based on its end-sequence tags and wild-type genome informatics ( C57BL/6 and Sv-129 ) . We performed shotgun sequencing of the insert of this cosmid and determined it to be 36 , 440-bp long , which was considerably longer than the expected 27 , 936-bp ( Figure S2A–S2C , Text S1 ) . We submitted the 36 , 440-bp sequence to GenBank with accession number AB70168 . Sequence analysis revealed that this cosmid contained a retrotransposon near the Gm13344 , Gm13336 , and Ptf1a genes . This retrotransposon was highly homologous to murine early transposon ( ETn ) endogenous retrovirus ( ERV ) 3 ( ETnERV3 ) —Family: ERVK , Class: long-terminal repeat ( 98 . 6% identity ) —whose size was 8 , 497 nucleotides ( AB701682 ) ( Figure 2A ) . This ETn was found to be tightly linked to the Sd phenotype , showing no recombination in 1 , 157 offspring . We performed Southern blot analysis using flanking DNA-specific genomic probes from the 5′-region to genotype Sd/Sd , Sd/+ , and WT ( +/+ ) mice . One 17 , 970-bp and one 9 , 461-bp band were detected in Sd/Sd E18 . 5 and +/+ embryos , respectively . Both were detected in Sd/+ E18 . 5 embryos ( Figure 2B ) . In addition , using PCR primer pairs shown in Figure 2A , the genotype of offspring from the heterozygous intercross could be easily determined ( Figure 2C ) . These data clearly suggest that the ETn insertion is associated with Sd phenotypes . The ETn insertion may disrupt an as-yet-unknown gene present in the insertion site or the presence of the ETn itself may cause the Sd phenotype . To distinguish these possibilities , we used a method that we developed for exchangeable gene targeting using Cre/lox [16] , [17] , [18] . Using this system , we were able to disrupt the target locus in the first step , and then produce an ETn knock-in allele ( kiETn ) by Cre-mediated site-specific recombination . To examine whether the Sd phenotype was caused by disruption of an as-yet-unknown gene present at the insertion site of the ETn , we constructed a targeting vector that contained a 5′ homology region , the neomycin resistance gene ( neo ) flanked by loxP and lox2272 , and a 3′ homology region . Using this targeting vector , we inserted neo flanked by loxP and lox2272 into embryonic stem ( ES ) cells at the site where the ETn is found in the Sd mouse ( Figure 3A ) . We obtained neo knock-in mice through mating of germline chimeras ( Figure S3A , S3B ) . The neo mice did not show any phenotype ( Figure S3C ) . Thus , we could rule out the possibility that Sd is caused by the disruption of an as-yet-unidentified gene at the insertion site of the ETn . We then examined whether the presence of ETn is required for the Sd phenotype . We first isolated an Xba I fragment containing the ETn with 287-bp upstream and 630-bp downstream regions from cosmid clone C3 . Then , we prepared a replacement vector that contained the Xba I fragment flanked by loxKR3 and lox2272 . By electroporating this replacement vector and a Cre expression vector , we created a knock-in allele in ES cells in which neo was replaced with ETn . The puromycin resistance ( puro ) gene was removed by expressing Flp , thus producing the kiETn allele ( Figure 3A ) . The established kiETn heterozygous animals were fertile and showed a kinked tail phenotype . In this experiment , none of wild type mice ( 0/29 ) showed kinky tail , while 27% ( 6/22 ) kiETn/+ mice showed kinky tail ( Figure 3B ) . There was no sex difference . Homozygotes showed characteristic Sd phenotypes , such as short tails with a defect of the nucleus pulposus of the intervertebral discs , anorectal malformations , and hypoplasia of the kidney , which were more severe than those in Sd heterozygotes ( Figure 3B , 3C ) . All homozygous kiETn animals showed perinatal lethality . These data strongly supported the hypothesis that the ETn insertion is the cause of the Sd phenotype . However , the kiETn mutant phenotype was less severe than that observed in Sd mice . The kiETn allele was produced using an Xba I fragment containing the ETn with 287-bp 5′ and 630-bp 3′ flanking genomic sequences; these genomic flanking regions were duplicated in the kiETn allele . In addition , the loxKR3/loxP and Frt sequences were retained in the kiETn allele . Therefore , it is possible that the retention of these sequences attenuated the effect of the ETn on the phenotype . To examine the effects of the ETn on transcription of the Gm13344 , Gm13336 , and Ptf1a genes , we first obtained full-length sequences for the Gm13344 , Gm13336 , and Ptf1a transcripts by rapid amplification of cDNA ends ( RACE ) and reverse transcription PCR ( RT-PCR ) ( Text S1 ) . The full-length sequences of the Gm13344 , Gm13336 , and Ptf1a transcripts were determined by compiling the sequences of the 5′ RACE and 3′ RACE products with those of expressed sequence tags that showed 100% homology to the RACE products . The final Gm13344 and Gm13336 cDNA sequences were 1 , 557-bp ( AB701678 ) and 917-bp ( AB701680 ) long , respectively ( Figure 4A ) . In the case of Gm13344 , we identified two alternative splicing products , 1 , 557-bp and 1 , 403-bp ( AB701679 ) long; splicing in the first exon produced the latter transcript . Unexpectedly , we found a 1 , 105-bp fusion transcript ( AB701681 ) containing the first and second exons of Gm13336 and 456-bp of the ETn sequence in Sd mutant embryos , which we termed the mutant Gm13336 transcript ( mGm13336 ) ( Figure 4A ) . Our sequencing data suggested that the Gm13344 and Gm13336 genes do not contain a significant open reading frame ( ORF ) . Moreover , Gm13344 partially overlaps with Gm13336 , and Gm13336 partially overlaps with Ptf1a . Interestingly , quantitative RT-PCR analyses revealed increased expression of all four transcripts in the Sd mutant at E9 . 0 and E9 . 5 ( Figure 4B , 4C ) , but not at E10 . 0 . To determine whether any of these four transcripts is responsible for Sd , we first generated two lines of transgenic mice carrying either a proximal genomic fragment including Gm13344 and the ETn ( Gm13344-ETn ) or a distal genomic fragment including the ETn , Gm13336 , and Ptf1a ( ETn-Gm13336-Ptf1a ) ( Figure 5A ) . Although all four transcripts were expressed in transgenic mice at E10 . 5 or neonatally ( Figure 5B ) , Gm13344-ETn transgenic embryos showed no phenotype ( Figure 5C ) , while ETn-Gm13336-Ptf1a transgenic mice showed a short tail similar to that of the Sd mutant ( Figure 5C ) . These results strongly suggested that the increased expression of Gm13336 , mGm13336 , and/or Ptf1a is the cause of Sd . Because Gm13336 and Ptf1a overlap each other , the strategy we employed to identify the Sd gene was as follows . First , we created a null allele for all three transcripts—Gm13336 , mGm13336 , and Ptf1a . Then , we expressed each gene by returning it into this locus by the exchangeable gene targeting method , to determine whether the Sd phenotype could be reproduced . For this purpose , we successfully obtained germline-competent ES cell lines using blastocysts obtained by mating Sd/+ mice and +/+ mice . Four of 15 ES cell lines carried the Sd allele and three were positive for the Sry gene ( Figure S4A , Text S1 ) . All the chimeric mice showed a short tail ( Figure S4B ) . We used this germline-competent Sd/+ ES cell line for disruption of the Gm13336-Ptf1a allele . The vector used for homologous recombination in Sd/+ ES cells is shown in Figure 6A . The neo cassette was inserted between the first and second exons of Gm13336 , resulting in deletion of the first exon of Ptf1a . Nine targeted ES clones lacking Gm13336-Ptf1a were identified by Southern blot analysis with both a 5′ probe and a 3′ probe ( Figure 6A ) and were used to generate chimeric mice . We obtained eight chimeric mice and four of them were germline chimeras . The Gm13336-Ptf1a+/neo mice were healthy and fertile , and indistinguishable from their negative littermates . Then , we examined whether the Gm13336-Ptf1aneo allele and the ETn segregated in the next generation . We obtained 20 offspring from three germline chimeras . None showed segregation of the Gm13336-Ptf1aneo and ETn alleles ( Figure S5A ) , suggesting that these offspring carry the Gm13336-Ptf1aneo and ETn alleles on the same chromosome [ETn-Gm13336-Ptf1aneo/+-+; cis configuration] . These mice were intercrossed to produce homozygous ETn-Gm13336-Ptf1aneo/ETn-Gm13336-Ptf1aneo mice . The ratio of +/+ , ETn-Gm13336-Ptf1aneo/+-+ , and ETn-Gm13336-Ptf1aneo/ETn-Gm13336-Ptf1aneo mice at E18 . 5 was 13∶25∶15 ( n = 53 ) , which is accordance with the expected Mendelian ratio . In ETn-Gm13336-Ptf1aneo/ETn-Gm13336-Ptf1aneo embryos , no Gm13336 or Ptf1a transcripts were detected ( Figure 6B ) , indicating the creation of a null allele . Interestingly , ETn-Gm13336-Ptf1aneo/ETn-Gm13336-Ptf1aneo embryos exhibited no abnormalities in vertebral , urogenital , or anorectal development ( Figure 6C , 6D ) , despite the presence of the ETn allele . However , the pancreas was missing in ETn-Gm13336-Ptf1aneo/ETn-Gm13336-Ptf1aneo E18 . 5 embryos , as expected ( Figure S5B ) . Using the targeted ES clone , we obtained one germline chimera , which transmitted the ETn and Gm13336-Ptf1aneo alleles independently into its offspring . In this case , F1 mice carried the ETn and the Gm13336-Ptf1aneo alleles on different chromosomes [ETn-+/+-Gm13336-Ptf1aneo; trans configuration] and thus showed a similar phenotype to that of the heterozygous Sd mutant . These findings clearly indicated that the cis configuration of ETn-Gm13336-Ptf1a causes the Sd phenotype . To identify the gene causing Sd , we prepared two replacement vectors . One contained the Ptf1a ORF flanked by loxKR3 and lox2272 [16] , [18] . Using Cre-mediated recombination , we replaced the phosphoglycerate kinase 1 promoter ( PGK ) -neo gene with the Ptf1a ORF in ETn-Gm13336-Ptf1aneo ES cells ( Figure 7A ) . The other replacement vector contained exons 1 and 2 of the Gm13336 gene with a CAG promoter flanked by loxKR3 and lox2272 . Using Cre-mediated recombination , we replaced the PGK-neo gene with the CAG-exons 1 and 2 of Gm13336 ( CAG-Gm13336 ( 1–2 ) ) in ETn-Gm13336-Ptf1aneo ES cells ( Figure 7A ) . Although we confirmed the expression of Gm13336 and mGm13336 in ETn-Gm13336-Ptf1aGm13336 ( 1–2 ) /+-+ and ETn-Gm13336-Ptf1aGm13336 ( 1–2 ) /ETn-Gm13336-Ptf1aGm13336 ( 1–2 ) mice , we observed no abnormal phenotypes ( Figure 7B ) . In contrast , the ETn-Gm13336-Ptf1aPtf1a/+-+ mice showed a short tail similar to that in Sd/+ mice , while ETn-Gm13336-Ptf1aPtf1a/ETn-Gm13336-Ptf1aPtf1a neonates showed no tail and a short trunk ( Figure 7B and Figure S6A ) . Histological examination revealed defects of the nucleus pulposus of the intervertebral discs and anorectal malformations similar to the defects in Sd/Sd mice , although the kidney defect was less severe ( Figure 7C ) . As expected , the pancreas was restored to normal in ETn-Gm13336-Ptf1aPtf1a/ETn-Gm13336-Ptf1aPtf1a mice ( Figure 7C ) . The vertebral columns of ETn-Gm13336-Ptf1aPtf1a/+-+ and ETn-Gm13336-Ptf1aPtf1a/ETn-Gm13336-Ptf1aPtf1a mice were usually truncated at the eighth caudal and the tenth thoracic vertebrae , respectively ( Figure S6B ) . To exclude the possibility that Gm13336 or mGm13336 is responsible for Sd , we introduced CAG-Gm13336 or CAG-mGm13336 into the gene-trap ES cell line 21-B137 ( Figure S7A , Text S1 ) . We confirmed the expression of these transcripts and observed no abnormalities ( Figure S7B ) . If the Ptf1a gene is responsible for Sd , Ptf1a should be expressed in tissues where the various phenotypes are observed . To examine the expression pattern of Ptf1a during embryonic development , we first tried to detect Ptf1a mRNA by in situ hybridization or PTF1a protein by immunohistochemistry . However , we failed to detect the expression of Ptf1a by either method , probably because of low-level expression and the low sensitivity of these methods . Thus , we used Cre-mediated recombination to replace the PGK-neo cassette in ETn-Gm13336-Ptf1aneo ES cells with the lacZ gene ( Figure 8A ) . These mice were examined for lacZ expression by whole mount X-gal staining during embryonic development at E8 . 5 , E9 . 5 , E10 . 5 , and E11 . 5 ( Figure 8B ) . Interestingly , lacZ expression was detected in the notochord and hindgut at E8 . 5 and E9 . 5 . LacZ expression extended to the cloaca and mesonephros at E10 . 5 and to the pancreatic bud at E10 . 5 and E11 . 5 . LacZ expression was strongly detected in the notochord , mesonephros , and cloaca at E9 . 5 ( Figure 8C ) . Kawaguchi et al . previously reported that expression of lacZ in Ptf1aCre;ROSA26R mice was restricted to the pancreas during development in Ptf1a-Cre transgenic mice in which the Cre gene was inserted into the Ptf1a locus [19] . Therefore , this ectopic expression in ETn-Gm13336-Ptf1alacZ embryos was considered to be induced by ETn insertion . Taken together , the ectopic expression pattern of Ptf1a is consistent with the phenotypes observed in Sd . To examine the effects of ectopic Ptf1a expression , we carried out transcriptional profiling using E10 . 0 Sd/Sd and WT whole embryos ( Tables S1 and S2 ) ( Text S1 ) . As expected , Ptf1a was upregulated 3 . 7-fold in Sd/Sd embryos . Among the significantly downregulated genes , we focused on the Cdx2 and T genes , because Cdx2 is known to regulate the expression of genes such as T , Wnt3a , and Cyp26a1 , which are essential for development of the posterior embryo [20] . In fact , mice mutant for these genes show tail phenotypes similar to those in Sd mice [21] , [22] , [23] , [24] , [25] , [26] , [27] . Thus , the expression of these genes may be suppressed by the ectopic expression of Ptf1a in tissues such as the notochord , gut , and mesonephros . We analyzed the mRNA expression of Ptf1a , Cdx2 , T , Wnt3a , and Cyp26a1 in ETn-Gm13336-Ptf1aPtf1a/+-+ and ETn-Gm13336-Ptf1aPtf1a/ETn-Gm13336-Ptf1aPtf1a embryos as well as in Sd E9 . 5 embryos . The level of Ptf1a expression in ETn-Gm13336-Ptf1aPtf1a/ETn-Gm13336-Ptf1aPtf1a embryos was about 80% of that in Sd embryos , but was 20-fold higher than that in WT embryos ( Figure 8D ) . As mentioned above , RNA profiling showed a 3 . 7-fold increase , but this discrepancy could be caused by differences in sensitivity and specificity . Both Cdx2 and T expression in ETn-Gm13336-Ptf1aPtf1a/ETn-Gm13336-Ptf1aPtf1a and Sd embryos decreased to about 40% of that in WT embryos at E9 . 5 . Wnt3a and Cyp26a1 expression was decreased to 42% and 62% of WT levels , respectively , in ETn-Gm13336-Ptf1aPtf1a/ETn-Gm13336-Ptf1aPtf1a embryos , while their expression in Sd/Sd embryos was similar to that in WT embryos . However , only two genes , Cdx2 and T , were downregulated at E10 . 0 and E11 . 5 in ETn-Gm13336-Ptf1aPtf1a/ETn-Gm13336-Ptf1aPtf1a mice ( Figure S8 ) . We further analyzed whether expression of Ptf1a can suppress the expression of Cdx2 , T , Wnt3a , and Cyp26a1 in ES cells . We inserted the CAG-Ptf1a gene into the 21-B137 allele using the same method ( see Figure S9A ) . As expected , the mRNA expression of Cdx2 , T , Wnt3a , and Cyp26a1 was significantly decreased ( Figure S9A ) . Furthermore , we transiently expressed Ptf1a in WT ES cells by electroporating in the expression vector CAG-Ptf1a . As shown in Figure S9B , the expression of Cdx2 , T , Wnt3a , and Cyp26a1 was decreased compared with that in ES cells transfected with the control CAG-enhanced green fluorescent protein ( EGFP ) ( Text S1 ) . As suggested by our ES cell data shown in Figure S9 , Ptf1a expression can downregulate the expression of four genes: Cdx2 , T , Wnt3a , and Cyp26a1 . Accordingly , all four genes were downregulated in E9 . 5 ETn-Gm13336-Ptf1aPtf1a/ETn-Gm13336-Ptf1aPtf1a embryos ( Figure 8D ) . However , only two genes , Cdx2 and T , were downregulated at E10 . 0 and E11 . 5 ( Figure S8 ) , similar to the result observed at E9 . 5 in Sd embryos . The precise spatial or temporal regulation of Ptf1a gene expression in Sd mice could differ from that in ETn-Gm13336-Ptf1aPtf1a mice because of the difference in allele structure . Taken together , our results strongly suggest that ectopic expression of Ptf1a induced by ETn insertion suppresses the Cdx2 gene and its downstream targets such as T , Wnt3a , and Cyp26a1 , resulting in the characteristic phenotypes observed in Sd mice ( Figure 8E ) .
In this study , we have revealed the nature of the Sd mutation to be the insertion of an ETn causing ectopic expression of Ptf1a in the caudal region of the embryo , resulting in suppression of Cdx2 and its downstream target genes . Endogenous retroviruses are present in the genomes of all vertebrates [28] , [29] . Retrotransposons are genetic elements that can amplify themselves in a genome and are ubiquitous components of the DNA of many eukaryotic organisms . They are responsible for the majority of ERV-induced de novo germline mutations [30] . Most commonly , germline mutations caused by retrotransposon insertions occur in an intron , disrupting gene expression by causing premature polyadenylation , aberrant splicing , or ectopic transcription driven by the long terminal repeat . For ETn insertions , the most commonly reported defect is premature polyadenylation within the ETn , coupled with aberrant splicing because of a few commonly used cryptic splice signals [31] . However , ETn-promoted ectopic gene expression has not previously been observed [31] . Kano et al . reported that dactylaplasia ( Dac ) is a LTR retroptransposon insertion caused by the type D mouse endogenous provirus element ( MusD ) , and that the ectopic MusD expression at the apical ectodermal ridge of limb buds correlates with the dactylaplasia phenotype [32] . However , in this Dac mutation , ETn-promoted ectopic expression of any endogenous genes has not been observed . In this study , we revealed the insertion of an ETn approximately 12 kb upstream of the transcription initiation site of the Ptf1a gene , resulting in the misexpression of Ptf1a . As the expression of Gm13344 and Gm13336 was also increased by this insertion , ETn acts as an enhancer , instead of as a transcription initiator . To the best of our knowledge , this is the first report of an ETn insertion causing ectopic gene expression . Ptf1a , which encodes a basic helix-loop-helix transcription factor , was originally reported to be a pancreatic determiner that drives undifferentiated cells in the foregut endoderm to differentiate into a pancreatic lineage [19] , [33] . In humans , Ptf1a was identified as responsible for the human permanent neonatal diabetes mellitus associated with cerebellar ataxia , and was reported to be involved in cerebellar development [34] . In fact , ectopic expression of Ptf1a conferred inhibitory gamma aminobutyric acid characteristics to neural progenitors , which were normally fated to become glutamatergic excitatory neurons [35] . These loss- and gain-of-function experiments suggested that Ptf1a itself acts as a cell fate determinant . However , in our case , Ptf1a overexpression and/or ectopic expression resulted in downregulation of Cdx2 and its downstream targets T , Wnt3a , and Cyp26a1 . The shortened tail phenotype of the Sd mutant is similar to that of the Cdx2 mutant , although heterozygous Cdx2 mutants show an anterior homeotic shift of the cervical and thoracic spine and rib abnormalities that are not observed in Sd mice [23] . The shortened tail phenotype of the T mutant [23] , [36] is also similar to that of the Sd mutant . A mutation in the Wnt3a gene results in a lack of caudal somites , a disrupted notochord , and failure to form a tail bud [37] , [38] . Cyp26a1-null mutants die during mid–late gestation and show a number of major morphogenetic defects , such as truncation of the tail , deficiencies of the external genitalia , anal atresia , and horseshoe kidneys [25] , [26] . Interestingly , expression of T and Wnt3a in the tail bud was downregulated in Cyp26a1-deficient mice . Taken together , these results suggest that the phenotypes observed in Sd mice are caused by the combination of partial deficiency of Cdx2 and its downstream target genes . We demonstrated that the mRNA expression of Cdx2 , T , Wnt3a , and Cyp26a1 was significantly decreased in ES cells overexpressing Ptf1a . Ptf1 is a trimeric transcription factor comprising Ptf1a , an E protein ( such as Tcf4 or Tcf12 [39] ) , and a third protein such as the mammalian Suppressor of Hairless ( RBP-J ) or its paralog RBP-L [40] . However , the binding site for Ptf1 has not been reported in the Cdx2 gene , nor in the T , Wnt3a , or Cyp26a1 genes . Instead , there are two Tcf-binding elements in the promoters of both the Cdx2 and T genes in the mouse [41] , [42] . It is possible that overexpressed Ptf1a can bind to Tcf4 , preventing the binding of Tcf4 to the promoter , leading to attenuation of the expression of Cdx2 and T . Decreased expression of Cdx2 may cause activation of β-catenin-mediated transcriptional activity , because Cdx2 directly binds β-catenin and disrupts the β-catenin–Tcf protein complex [42] , [43] , [44] . However , nuclear β-catenin interacts with the Tcf/lymphoid enhancer factor ( Lef ) family of DNA-binding proteins to regulate the expression of numerous Wnt target genes [45] , [46] , [47] . Thus , the removal of Tcf by Ptf1a may result in the degradation of β-catenin and downregulation of the Wnt target genes . CRS is a congenital heterogeneous constellation of caudal anomalies that includes varying degrees of agenesis of the spinal column , anorectal malformations , and genitourinary anomalies . Its pathogenesis is unclear , but it could be the result of excessive physiologic regression of the embryonic tail . As described above , the various mouse mutants have shown that caudal agenesis occurs a result of hypodevelopment of the anterior–posterior axis . The Sd mouse is considered a model for human CRS based on phenotypic similarity in the spine , hindgut , and urogenital system . The exact etiology in humans is unknown , except in some cases of Currarino syndrome . Currarino syndrome is a form of CRS with hemisacrum , anorectal malformations , and presacral mass , such as teratoma [48] . Previous reports have shown that HLXB9 is a major causative gene for Currarino syndrome [49] , [50] . HLXB9 is a homeobox protein that contains a glycine-and-alanine-rich region and a strongly acidic region next to the amino-terminal and carboxy-terminal side of the homeodomain , respectively [51] . However , a null mutation of the Hlxb9 gene in mice showed agenesis of the dorsal pancreas but no skeletal truncation [52] , [53] . It is of interest that both Ptf1a and Hlxb9 are expressed in the pancreas , and that Ptf1a is required for Hlxb9 expression [54] . Although it is not clear whether PTF1A is involved in human CRS , the evidence from this study implicates PTF1A and possibly HLXB9 in the caudal abnormalities . Further studies using Sd mice will provide further insight into the development of human CRS .
Sd mice purchased from the Jackson Laboratory ( Bar Harbor , ME ) were backcrossed to C57BL/6 mice for at least ten generations . All experiments were performed in accordance with the Declaration of Helsinki and were approved by the Kumamoto University Ethics Committee for Animal Experiments ( authorization number in Kumamoto University: C23-262 , C24-278 ) . PCR was used to genotype Sd alleles . For the WT allele , the 5′ primer 5′Sd-S1 ( 5′-GAAAGCAAAGGGCTGCTTAC-3′ ) and the 3′ primer 3′Sd-A1 ( 5′-TATTCTTGCAGGGAGAGTTG-3′ ) were used to amplify a 283-bp fragment . To detect the Sd allele , the 5′ primer 5′Sd-S1 ( 5′-GAAAGCAAAGGGCTGCTTAC-3′ ) and the 3′ primer 5′Tn-A1 ( 5′-TCTCGTGTGATCTGTCTGTC-3′ ) , located in the ETn , were used to amplify a 228-bp fragment . The PCR conditions were as follows: denaturation at 94°C; followed by 30 cycles of denaturation at 94°C for 30 s , annealing at 56°C for 30 s , and extension at 72°C for 30 s . PCR products were visualized on a 1% agarose gel using ethidium bromide . For Southern blotting , genomic DNA was digested overnight with Sph I and subjected to electrophoresis on a 1% agarose gel . DNA was transferred onto a positively charged nylon membrane ( Roche , Indianapolis , IN ) . After baking at 80°C for 1 h , the membrane was hybridized with a flanking genomic DNA-specific probe ( Figure 2A ) . ES cells were cultured in KSR-GMEM medium consisting of Glasgow Minimum Essential Medium ( Sigma , St Louis , MO ) with 1×nonessential amino acids ( Gibco Invitrogen , Grand Island , NY ) , 0 . 1 mM β-mercaptoethanol , 1 mM sodium pyruvate , 1% fetal bovine serum ( HyClone; Thermo Fisher Scientific Inc . , Waltham , MA ) , 14% Knockout Serum Replacement ( Gibco Invitrogen ) , and 1100 U/ml leukemia inhibitory factor ( ESGRO; Chemicon , Temecula , CA ) . The knock-in method used in this study was developed by us and has been described [16] , [17] , [18] . In the first step , the targeting vector—containing a 5′ homology region , loxP , Frt , PGK-neo cassette , lox2272 , polyadenylation signal ( pA ) , Frt , 3′ homology region , and an MC1 promoter-diphtheria toxin A fragment with a pA ( MC1-DT-A ) —was constructed using pBluescript II containing the PGK-neo cassette ( p03 ) [55] . Targeting vectors were electroporated into feeder-free KTPU8 ES cell lines derived from the TT2 ES cell line , according to previously described methods [56] , [57] . Three targeted ES clones were obtained from 288 G418-resistant clones . ES cells were aggregated with ICR morulas to produce chimeric mice . Germline transmission was confirmed in all three lines . In the second step , the replacement vector—containing loxKR3 ( KR3 ) , ETn , Frt , puromycin N-acetyl-transferase ( puro ) , lox2272 , Frt , MC1-DT-A , and loxP—was electroporated into the targeted ES cell clones to establish the kiETn allele [18] . The PGK-puro cassette was removed by transient Flp expression . One replaced ES clone was obtained from 28 puromycin-resistant clones . These kiETn ES cells were used to produce germline chimeras . Heterozygous kiETn mice were backcrossed to C57BL/6 for at least five generations . Then , heterozygous kiETn mice were intercrossed to produce homozygous kiETn mice . Founder kiETn mice were identified by PCR and Southern blotting . Genomic DNA was extracted from an ear clip . The ETn fragment digested by Xba I contained both 5′ and 3′ flanking genomic sequences; partial tandem duplication ( 5′: 287-bp; 3′: 630-bp ) of flanking genomic sequences was confirmed by PCR analyses . To detect the kiETn allele , the 5′ primer 5′Sd-S1 ( 5′-GAAAGCAAAGGGCTGCTTAC-3′ ) , located in the 5′ flanking genomic region , and the 3′ primer 5′Tn-A1 ( 5′-TCTCGTGTGATCTGTCTGTC-3′ ) , located in the ETn , generated 228-bp and 650-bp fragments , respectively . To detect the WT allele , the 5′ primer 5′Sd-S1 ( 5′-GAAAGCAAAGGGCTGCTTAC-3′ ) , located in the 5′ flanking genomic region , and the 3′ primer 3′Sd-A1 ( 5′-TATTCTTGCAGGGAGAGTTG-3′ ) , located in the 3′ flanking genomic region , generated a 283-bp fragment . To detect the targeted allele , the 5′ and 3′ primers neo-F ( 5′-AGAGGCTATTCGGCTATGAC-3′ ) and neo-R ( 5′-CACCATGATATTCGGCAAGC-3′ ) , respectively , both located in the targeting vector , generated a 545-bp fragment . For Southern blotting , genomic DNA was digested overnight with Sph I and subjected to electrophoresis on a 1 . 0% agarose gel . DNA was transferred onto a positively charged nylon membrane ( Roche ) . After baking at 80°C for 1 h , the membrane was prehybridized and then hybridized using flanking genomic DNA-specific probes ( Figure 2A ) prepared using a Digoxigenin DNA Labeling and Detection Kit ( Roche ) . PCR amplification of each DNA fragment and gene was performed using TaKaRa EX or LA Taq ( Takara , Kyoto , Japan ) , according to the manufacturer's protocol . The primer sequences were as follows: Gm13344 ( accession number AB701678 and AB701679 ) : 5′-ACGAATGGGGTGTTCAGACG-3′ ( sense ) and 5′-CGACTGCCAGACCCAGGAAG-3′ ( antisense ) , generating two alternative splicing products , 446-bp and 297-bp fragments; Gm13336 ( accession number AB701680 ) : 5′-TGACGCTTTGTGAGTGATCC-3′ ( sense ) and 5′-AACACTCCTGTGATGTGTAG-3′ ( antisense ) , generating a 226-bp fragment; mGm13336: 5′-TGACGCTTTGTGAGTGATCC-3′ ( sense ) and 5′-GAACAATACGATTTCTTTTTACCTG-3′ ( antisense ) , generating a 488-bp fragment; Ptf1a: 5′-TGAGGGACCTACCCGAATTG-3′ ( sense ) and 5′-ACAATATGCACAAAGACGCG-3′ ( antisense ) , generating a 1 , 105-bp fragment; Actb ( β-actin; control ) : 5′-ATGTACGTAGCCATCCAGGC-3′ ( sense ) and 5′-AAGAAGGAAGGCTGGAAAAG-3′ ( antisense ) , generating a 407-bp fragment; Gapdh ( glyceraldehyde 3-phosphate dehydrogenase; control ) : 5′-GGAAAGCTGTGGCGTGATG-3′ ( sense ) and 5′-CTGTTGCTGTAGCCGTATTC-3′ ( antisense ) , generating a 392-bp fragment . Thermal cycling was carried out with denaturation at 94°C; followed by 30 , 35 , or 40 cycles of denaturation at 94°C for 30 s , annealing at 56°C for 30 s , and extension at 72°C for 30 s . PCR products were visualized on 1% agarose gels using ethidium bromide . Reverse transcribed products were used for quantitative real-time PCR using an Applied Biosystems 7500 Real-Time PCR system ( Applied Biosystems , Foster City , CA ) . TaqMan Gene Expression Master Mix and TaqMan Gene Expression Assays for Gm13336 ( a custom-made Gm13336 TaqMan probe ) , mGm13336 ( a custom-made mGm13336 TaqMan probe ) , Ptf1a ( Mm 00479622 ) , Cdx2 ( Mm 01212280 ) , T ( Mm 01318252 ) , Wnt3a ( Mm 00437337 ) , Cyp26a1 ( Mm 00514486 ) , and Actb ( Mm 00607939 ) were purchased from Applied Biosystems . Reactions were carried out under the following conditions: 2 min at 50°C and 10 min at 95°C; followed by 40 cycles of 15 s at 95°C and 1 min at 60°C . Calibration was conducted using the relative standard curve method . To construct a standard curve , a standard sample cDNA was prepared from E10 . 5 embryos for Cdx2 , T , Wnt3a , Cyp26a1 , and Actb , or Ptf1a cDNA for Ptf1a , or Gm13336 cDNA for Gm13336 , or mGm13336 cDNA for mGm13336 . For each PCR assay , the standard curve was generated using the same standard sample . The relative concentration of the target gene in each sample was calculated from the constructed standard curve , and the ratio of the relative concentration of the target gene to Actb in each sample was calculated . This ratio represented the relative expression of the target gene normalized to Actb compared with the standard sample . Actb is recommended by Applied Biosystems as a suitable endogenous internal control for TaqMan RT-PCR analyses . Real-time PCR for Gm13344 transcripts was performed using the THUNDERBIRD SYBR qPCR mix ( Toyobo , Osaka , Japan ) . The following primer pair was used for the shorter splice variant Gm13344 transcripts: 5′-TGTGCTGGACCCAAACATAGCCAAAG-3′ ( sense ) and 5′-CGACTGCCAGACCCAGGAAG-3′ ( antisense ) , generating a 293-bp fragment . The relative concentration of Gm13344 in each sample was calculated from the constructed standard curve , and the ratio of the relative concentration of Gm13344 to Actb was calculated . This ratio represented the relative expression of the target gene normalized to Actb compared with the standard sample . To isolate the 21 , 815-bp DNA fragment containing Gm13344-ETn and the 24 , 714-bp DNA fragment containing ETn-Gm13336 from a cosmid clone , cosmid DNA was digested with Xho I and Bsm BI , respectively . Isolated DNA fragments were microinjected into fertilized eggs obtained from C57BL/6 mice , at a final concentration of 1 µg/ml in Tris-EDTA buffer . Genomic DNA containing the coding region of the Ptf1a gene was isolated from the cosmid clone containing the ETn . The pBluescript II construct containing the PGK-neo cassette ( p03 ) and the MC1-DT-A fragment was used as a backbone to construct the targeting vector . The targeting vector comprised a 5′ homology region , loxP , Frt , PGK-neo , lox2272 , pA , Frt , 3′ homology region , and MC1-DT-A ( Figure 6A ) . The 5′ homology region contained the first exon , first intron , second exon , and part of the second intron of Gm13336; part of the first exon of Ptf1a was deleted by insertion of the neo cassette . The targeting vector was electroporated into Sd/+ ES cells as described above . Nine targeted ES clones were obtained from 192 G418-resistant clones . ES cells were aggregated with ICR morulas as described above . Germline chimeras were obtained from four ES lines . This strain of mouse was designated ETn-Gm13336-Ptf1aneo . ETn-Gm13336-Ptf1aneo/+-+ mice were backcrossed to C57BL/6 mice for at least three generations . Then , ETn-Gm13336-Ptf1aneo/+-+ mice were intercrossed to produce ETn-Gm13336-Ptf1aneo/ETn-Gm13336-Ptf1aneo mice . Founder ETn-Gm13336-Ptf1aneo mice were identified by PCR and Southern blotting . Genomic DNA was extracted from an ear clip . To detect the targeted allele , the 5′ primer neo-F ( 5′-AGAGGCTATTCGGCTATGAC-3′ ) and the 3′ primer neo-R ( 5′-CACCATGATATTCGGCAAGC-3′ ) , located in the neo cassette , generated a 545-bp fragment . To detect the WT allele , the 5′ primer 5′AKKO-S1 ( 5′-ATTGCTCAGAACCCCTAGGG-3′ ) , located in the 5′ flanking genomic region of the second exon of Ptf1a , and the 3′ primer 3′AKKO-A1 ( 5′-GATTCCCTGAGCTGTGAAGC-3′ ) , located in the 3′ flanking genomic region of the second exon of Gm13336 , generated a 1 , 777-bp fragment . For Southern blotting , genomic DNA was digested overnight with Eco RI and Spe I and electrophoresed on 1 . 0% agarose gels . DNA was transferred onto a positively charged nylon membrane ( Roche ) . After baking at 80°C for 1 h , the membrane was prehybridized and then hybridized using 5′ and 3′ flanking genomic DNA-specific probes ( Figure 6A ) prepared using a Digoxigenin DNA Labeling and Detection Kit ( Roche ) . To insert the Ptf1a ORF into the ETn-Gm13336-Ptf1a allele , ETn-Gm13336-Ptf1aneo ES cells were used . The ES cell clones were electroporated with the Cre expression vector and a replacement vector assembled from pKR3-Frt-del . pA-puro-2272 with the cloned Ptf1a ORF to establish ETn-Gm13336-Ptf1aPtf1a ES cell and mouse lines ( Figure 7A ) . ETn-Gm13336-Ptf1aPtf1a mice were backcrossed to C57BL/6 mice for at least three generations . Gm13336-Ptf1aPtf1a/+-+ mice were intercrossed to produce Gm13336-Ptf1aPtf1a/Gm13336-Ptf1aPtf1a mice . To insert the CAG-Gm13336 ( 1–2 ) into the ETn-Gm13336-Ptf1aneo allele , ETn-Gm13336-Ptf1aneo ES cells were used . The ES cell clones were electroporated with the Cre expression vector and a replacement vector assembled from pKR3-Frt-del . pA-puro-2272 with the cloned Gm13336 ( 1–2 ) to establish ETn-Gm13336-Ptf1aCAG-Gm13336 ( 1–2 ) ES cell and mouse lines ( Figure 7A ) . ETn-Gm13336-Ptf1aCAG-Gm13336 ( 1–2 ) mice were backcrossed to C57BL/6 mice for at least three generations . Then , Gm13336-Ptf1aCAG-Gm13336 ( 1–2 ) /+-+ mice were intercrossed to produce Gm13336-Ptf1aCAG-Gm13336 ( 1–2 ) /Gm13336-Ptf1aCAG-Gm13336 ( 1–2 ) mice . ETn-Gm13336-Ptf1aneo ES cells were used to insert lacZ into the ETn-Gm13336-Ptf1a allele . The ES cell clones were electroporated with a replacement vector assembled from pKR3-Frt-del . pA-puro-2272 and cloned lacZ to establish ETn-Gm13336-Ptf1alacZ mouse lines ( Figure 8A ) . ETn-Gm13336-Ptf1alacZ mice were backcrossed to C57BL/6 mice for at least two generations . Embryos and neonates were fixed in phosphate-buffered 15% formaldehyde overnight , rinsed twice for 1 h in phosphate-buffered saline ( PBS ) , dehydrated through increasing concentrations of ethanol , equilibrated with xylene , embedded in paraffin wax , and sectioned at 4 µm . Sagittal sections were stained with hematoxylin and eosin and examined by light microscopy . Samples were fixed for 30 min at room temperature in fix solution [1% formaldehyde , 0 . 2% glutaraldehyde , and 0 . 02% NP-40 in PBS] . Fixed samples were washed twice with PBS and incubated overnight at 30°C in staining solution ( 5 mm potassium ferricyanide , 5 mm potassium ferrocyanide , 2 mm MgCl2 , 0 . 5% X-gal in PBS ) . Samples were rinsed twice in PBS and then post-fixed in 10% formaldehyde . For observation of whole-mount X-gal staining , samples were made transparent using benzylalcohol/benzylbenzoate ( 1∶2 ) , after dehydration with a series of ethanol steps ( 25% , 50% , 70% , 100% , and 100% , 1 h each ) . For histological analysis , samples were sectioned at 8 µm and counterstained with Nuclear Fast red ( Funakoshi , Tokyo , Japan ) after X-gal staining . The results are presented as the mean ± standard deviation ( SD ) of independent experiments as detailed separately in each corresponding figure legend . Data were compared using the Student's t-test and were considered significantly different at p<0 . 05 . | Caudal regression syndrome ( CRS ) is a congenital heterogeneous constellation of caudal anomalies that includes varying degrees of agenesis of the spinal column , anorectal malformations , and genitourinary anomalies . Its pathogenesis is unclear . However , it could be the result of excessive physiologic regression of the embryonic caudal region based on analyses of the various mouse mutants carrying caudal agenesis . Among the mouse mutants , the Danforth's short tail ( Sd ) mouse is considered a best model for human CRS . Sd is a semidominant mutation , characterized by spinal defects , urogenital defects , and anorectal malformations , thus showing phenotypic similarity to human CRS . Although Sd is known to map to mouse chromosome 2 , little is known about the molecular nature of the mutation . Here , we demonstrate an insertion of one type of retrotransposon near the Ptf1a gene . This resulted in ectopic expression of Ptf1a gene in the caudal region of the embryo and downregulation of Cdx2 and its downstream targets , leading to characteristic phenotypes in Sd mouse . Thus , Sd mutant mice will provide insight into the development of the spinal column , anus , and kidney . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"biology",
"genetics",
"and",
"genomics"
]
| 2013 | Ectopic Expression of Ptf1a Induces Spinal Defects, Urogenital Defects, and Anorectal Malformations in Danforth's Short Tail Mice |
Genome-wide expression Quantitative Trait Loci ( eQTL ) studies in humans have provided numerous insights into the genetics of both gene expression and complex diseases . While the majority of eQTL identified in genome-wide analyses impact a single gene , eQTL that impact many genes are particularly valuable for network modeling and disease analysis . To enable the identification of such broad impact eQTL , we introduce CONFETI: Confounding Factor Estimation Through Independent component analysis . CONFETI is designed to address two conflicting issues when searching for broad impact eQTL: the need to account for non-genetic confounding factors that can lower the power of the analysis or produce broad impact eQTL false positives , and the tendency of methods that account for confounding factors to model broad impact eQTL as non-genetic variation . The key advance of the CONFETI framework is the use of Independent Component Analysis ( ICA ) to identify variation likely caused by broad impact eQTL when constructing the sample covariance matrix used for the random effect in a mixed model . We show that CONFETI has better performance than other mixed model confounding factor methods when considering broad impact eQTL recovery from synthetic data . We also used the CONFETI framework and these same confounding factor methods to identify eQTL that replicate between matched twin pair datasets in the Multiple Tissue Human Expression Resource ( MuTHER ) , the Depression Genes Networks study ( DGN ) , the Netherlands Study of Depression and Anxiety ( NESDA ) , and multiple tissue types in the Genotype-Tissue Expression ( GTEx ) consortium . These analyses identified both cis-eQTL and trans-eQTL impacting individual genes , and CONFETI had better or comparable performance to other mixed model confounding factor analysis methods when identifying such eQTL . In these analyses , we were able to identify and replicate a few broad impact eQTL although the overall number was small even when applying CONFETI . In light of these results , we discuss the broad impact eQTL that have been previously reported from the analysis of human data and suggest that considerable caution should be exercised when making biological inferences based on these reported eQTL .
The CONFETI framework is constructed to systematically avoid the tendency of other confounding factor analysis methods to model broad impact eQTL as confounding variation . This is accomplished by leveraging Independent Component Analysis ( ICA ) to identify generative sources of multivariate gene expression variation and then screening candidates based on component correlations with genotypes , which are then omitted from the confounding factor correction ( Fig 1 ) . ICA is widely used in machine learning for blind source separation problems to detect non-Gaussian signals from multivariate data and has been applied to a diverse set of problems including voice and image separation [86 , 87] . The reason ICA is particularly well suited for identifying candidate broad impact eQTL is that the method is designed to separate independent sources of multivariate variation . ICA assumes that the observed data for each sample is a linear combination of non-Gaussian statistically independent components . When applying ICA , the vector of expression values for an individual are modeled as weighted sum of independent components: y → i = a i 1 s 1 → + a i 2 s 2 → + ⋯ + a i k s k → = ∑ j = 1 k a i j s j ( 1 ) where y → i is a g-dimensional vector of gene expression values for a single sample , and independent components s j → are g-dimensional vectors of gene weights that are shared among all samples and the scalar component coefficients aij represent the contribution of each independent component s j → for sample i ( Fig 1 ) . When considering all samples together , the above can be simply expressed as a matrix decomposition: Y = A S ( 2 ) where Y is an n × g matrix with ith row y → i . A is the n × k mixing matrix with the jth column holding component coefficients a j → for component j , and S is the k × g independent component matrix in which the jth row is s j → . A and S are estimated by finding a projection of Y that maximizes the non-Gaussianity of the gene weight distribution of each row in S . In CONFETI these are identified by using the FastICA algorithm for reliable and fast computation [88] . Since ICA recovers factors by assessing non-Gaussianity and not the amount of variation explained as in methods such as Principal Component Analysis ( PCA ) or any other factor analysis method [86] , ICA is able to more clearly resolve separate factors responsible for variation , while a PCA or factor analysis will tend to identify composite effects , which are likely to be mixtures of multiple factors ( S1 Fig ) . The critical assumption for application of ICA in the CONFETI framework is that broad impact eQTL will have non-Gaussian impacts on the multivariate expression profile and that the effects of these eQTL will be relatively independent of other genetic and non-genetic factors . Complete independence is not necessary , since the framework only has to identify and retain enough of the expression variation due to a broad impact eQTL to make it detectable with an association test . The assumption that broad impact eQTL will tend to have non-Gaussian impacts is not particularly restrictive given that we expect eQTL with large enough effects to impact only a subset of the total number of genes and therefore be detectably non-Gaussian . The assumption that broad impact eQTL are relatively independent of each other is also not overly restrictive in humans given the low linkage disequilibrium observed among non-local genotypes throughout the genome . While the assumption that broad impact eQTL are largely independent of non-genetic factors is not always expected to hold , it seems likely in many cases unless there is a reason to expect broad impact eQTL to strongly interact with non-genetic factors such as sample-specific environmental effects or technical effects arising from differences between laboratories and procedures . Furthermore , in cases where broad impact eQTL are completely conflated with non-genetic factors , these broad impact eQTL will be indistinguishable from non-genetic contributions to the observed multivariate gene expression variation and will be modeled away by any confounding factor method . In summary , the only accurately detectable broad impact eQTL are those that have properties that are expected to make them identifiable by ICA . The complete CONFETI framework involves running ICA on multivariate gene expression data , an automated detection step to identify candidate broad impact eQTL by assessing associations with genotypes , and omission of these factors for the construction of the random effect sample covariance matrix used in a mixed model confounding factor analysis ( Fig 1 ) . While this approach could be used in combination with confounding factor methods that use a fixed covariate approach [69 , 74 , 76 , 89–92] , the framework more naturally integrates with a mixed model approaches to confounding factor analysis , since the random effect modeling in these methods provides a high dimensional modeling of confounding variation . A covariance matrix constructed from the non-genetic independent components is used to model confounding factors as random effects in a linear mixed model eQTL approach . We note that our framework differs from ICA methods for eQTL detection that treat the identified ICs as meta-genes , where these methods cannot reliably distinguish the specific gene effects of individual eQTL [83 , 93] . The only method that we are aware of close to this framework is ISVA , which uses ICA within the Surrogate Variable Analysis ( SVA ) method for iteratively modeling pre-specified fixed effects and confounding variation [91] . ISVA is not appropriate for eQTL analysis since it begins the iterative approach by pre-specifying the fixed effects and therefore pre-supposing the existence of a relationship , which would introduce a bias towards finding eQTL false positives . CONFETI on the other hand uses ICA to separate candidate broad impact eQTL without the need of pre-specifying the existence of the eQTL . We also note that in the mixed model based method PANAMA [72] , the authors discuss a strategy for avoiding the over-correction of trans-eQTL by jointly estimating the covariance matrix with genotype effects to avoid including those effects in the correction [72] . However , this approach is not a feature of PANAMA included in the LIMIX package [94] , which the authors have directed us to use . Moreover , the gene loadings in PANAMA are integrated out in the estimation step making it difficult to analyze the factors that are being corrected . In summary , the CONFETI framework utilizes the optimal properties of ICA to detect broad impact eQTL by excluding genetic effects from confounding variation accounted for in a mixed model , thereby taking advantage of the performance increases provided by mixed model confounding factor analysis without reducing the ability to identify broad impact eQTL .
To apply ICA to gene expression data and generate a sample covariance matrix , we developed a custom R package ( https://github . com/jinhyunju/confeti ) . The independent component estimation features are using functions adopted from the fastICA R package [95] which implemented the computationally efficient and robust FastICA algorithm [88] based on a fixed-point algorithm to find directions maximizing the Negentropy to identify statistically independent components ( ICs ) . The number of ICs that can be estimated is the smaller of the sample size or the number of features ( genes ) , and the sign of any particular estimated component is arbitrary . As the estimated ICs do not have any particular order and have the potential to change based on the input of number of components to estimate [91 , 96 , 97] , the package supports diagnostics for assessing optimal IC number such as functionality to estimate replicating ICs between multiple runs for ensemble ICA estimation . To provide a fair comparison between ICA and PANAMA [72] , which both require as input the number of components to be considered prior to estimation , we set the number of ICs to be estimated in the fastICA algorithm to explain the same variance as for the set of principal components accounting for 95% of the variance in the data . After decomposing the observed data Y into A and S we test for any significant associations between the component coefficients ( columns of A ) and all genotypes . As in fixed effect eQTL models , we fit a linear regression model with the IC coefficient as the dependent variable and the genotype values as independent variables . After calculating p-values for each IC coefficient and genotype pair , we identified candidate broad impact eQTL using a global Bonferroni corrected p-value threshold of 0 . 05 . Components with at least one significant association are marked as candidate genetic components . After filtering out r ( 0 ≤ r < k ) components with significant genotype association , we reconstruct expression matrix Y* originating from non-genetic factors using the remaining k − r components: Y * = A * S * ( 3 ) where Y* is an n × g matrix , A* is a n × ( k − r ) matrix and S* is a ( k − r ) × g matrix . Given that the overall CONFETI method makes use of the phenotype and genotype data both in the filtering out of candidate genetic effects and in the identification of significant genotype-gene expression associations , using the full dataset could lead to model over-fitting impacts in the selection and removal of ICs . To assess this issue , we compared the approach of using CONFETI on the full dataset to a strategy where we split the genotype data into two random subsets . For the splitting strategy , we used one of the genotype subsets for filtering candidate genetic effects and the remaining genotypes for the eQTL analysis , we then repeated the analysis flipping the subsets that are used for filtering and eQTL analysis , and the combined the results . With this splitting strategy , genotypes used for the removal of candidate genetic effects do not overlap with the genotypes that are being tested for eQTL , such that each genotype is only accessed once in each subset . From the analysis of multiple datasets , we found that the results obtained by using the full dataset and the splitting strategy largely overlapped with only minor differences ( S2 Fig ) . A possible reason for this observation is that over-fitting issue in the CONFETI framework differs from more standard cases in machine learning applications in that the estimated independent components are not being directly used as features , but are rather included in the model to account for sample similarity structures that violate the independence assumption of the model , i . e . , selected features are not being tested for associations . While we present the splitting strategy as an option for selecting and removing ICs for the users of CONFETI , given agreement with results when using the full dataset , and the additional complexity and computational costs in data splitting , separate analysis , and combining steps , we suggest applying CONFETI when considering the full dataset and adopt this approach in these analyses . We used two approaches to construct the sample covariance matrix K for the random effect part of the mixed model . Our first approach was to use a simple location-scale normalization of each gene of Y*: Z i p * = ( Y i p * - μ p ) / σ p ( 4 ) and then calculate sample covariance matrix: K = cov ( Z * ) ( 5 ) We label this approach CONFETI-I since it can be thought of as a specific , lower dimensional approach to Intersample Correlation Emended ( ICE ) , one of the first methods to estimate a sample structure for confounding factor analysis [62] by estimating the sample covariance matrix using the full dimensional observed expression data . For our second approach , we couple CONFETI with PANAMA ( Probabilistic ANAlysis of genoMic dAta ) [72] that estimates the covariance structure using a maximum likelihood framework . Using this approach , the likelihood objective can be stated as: p ( Y*|Kpanama ) =∏p=1gN ( y·p*→|Kpanama+σp2I ) ( 6 ) ( θ ^ , C ^ ) = argmax θ , C p ( Y * | C , θ ) ( 7 ) where C is an n × Q matrix initialized by projecting the observed data onto the first Q principal components explaining 95% of the variance and is further optimized in the process , and θ is the set of hyperparameters consisting of { { α q 2 } , σ p 2 } . Each α q 2 then represents the optimized weight of the qth column of C , C⋅q in constructing the sample covariance matrix: K = ∑ q = 1 Q α q 2 ^ C ^ · q C ^ · q T ( 8 ) We label this approach CONFETI-P , where we use of the implementation of PANAMA included in the LIMIX package [94] for the estimation of K . We model the genetic effects from SNPs and covariates as fixed effects and confounding factor effects as random effects , such that the expression levels for gene p in n individuals are: y → · p = X β p → + c p → + ϵ p → ( 9 ) c p → ∼ N ( 0 → , τ p 2 K ) ( 10 ) ϵ p → ∼ N ( 0 → , σ p 2 I ) ( 11 ) where n is the number of samples , g the number of genes , s the number of SNPs , and v the number of covariates . Each gene expression vector y → · p has dimension n × 1 and is mean centered . The n × ( 1 + v ) genotype and covariate matrix X contains a single genotype as the number of minor alleles coded as 0 , 1 , 2 and any additional v number of covariates . β p → is the ( 1 + v ) × 1 dimensional coefficient vector representing the fixed effect of the SNPs and covariates on gene p . The confounding effect is included in the model as a n × 1 random effect c p → sampled from a multivariate normal distribution with covariance τ p 2 K , where K is the n × n sample covariance matrix constructed the corresponding confounding correction method , τ p 2 is a scalar weight for K in the random effect , and ϵ p → is a n × 1 vector representing the independent error for gene p with scalar weight σ p 2 . We compared CONFETI-I and CONFETI-P to a simple linear regression with no confounding factor correction ( LINEAR ) , including PCA projections as fixed effects ( PCA ) , probabilistic estimation of expression residuals ( PEER ) [92] , and mixed model confounding factor methods ICE [62] and PANAMA [72] . For mixed model based confounding factor correction methods , we limited our comparison to methods that pre-calculate a sample covariance matrix ( K ) , which is kept constant when testing individual genotypes against phenotypes , to avoid the computational burden of recalculating K for every phenotype . For each comparison of methods on simulated or real data , we ran each method to be as equivalent as possible , including the same covariates and using the same linear mixed model fitting function . For CONFETI-I , CONFETI-P , PANAMA , and ICE we used lrgprApply ( ) function from the R package lrgpr [98] to fit the linear mixed model and calculate p-values for the genotype effects using a Wald test . Following the methodology of the GTEx analysis [14] , the number of factors for PEER were decided based on the sample size . We used 30 factors for datasets with sample size between 150 and 250 , and 35 factors for datasets with more than 250 samples . We used the same number of components for PCA correction . To fit the eQTL model using PEER , PCA , and LINEAR we used the glmApply ( ) function from lrgpr and used a Wald test for significance testing . To mirror real cases where a reasonable number of broad impact eQTL have been repeatedly identified , we used yeast as a model [63–65] . To create simulated datasets , we used 2956 yeast genotypes from the study of Smith et al . [99] and randomly sampled 3000 yeast gene annotations to simulate cis- and trans-eQTL relationships . To simulate eQTL , a matrix with a dimension of number of genotypes × number of expression phenotypes was first created that marks genotype and phenotype pairs cis- if the starting position of the gene and the genotype were within 100 , 000 base pairs distance and trans- if the distance was greater . From this matrix we sampled 2500 genotype and phenotype pairs which consisted of 80% cis- and 20% trans-genotypes . In total , for each simulated dataset , we included 2000 cis-eQTL , 500 trans-eQTL , and 10 broad impact eQTL . We simulated each broad impact eQTL to affect 10% of the expression phenotypes . Effect sizes for cis-eQTL were sampled from N ( 0 . 8 , 1 ) and effect sizes for trans-eQTL and broad impact eQTL were sampled from N ( 0 . 48 , 1 ) ( 70% attenuation of trans-effects ) to reflect observed effect sizes in real data . After the eQTL effects were simulated , we added normally distributed random noise sampled from N ( 0 , 1 ) . For confounding factor effects , we simulated two types of confounding factors: sparse and dense . For sparse confounding factors 30% of phenotypes were affected with effect sizes drawn from N ( 1 , 0 . 5 ) , and for the dense confounding factors , the effect over all genes followed a standard normal distribution N ( 0 , 1 ) . We tested 2 scenarios , each with 30 confounding factors: sparse only , and mixed ( 15 sparse and 15 dense ) . We simulated and analyzed 50 datasets for each of these two scenarios , a total of 100 datasets . We ran each of the methods CONFETI-I , CONFETI-P , PANAMA , ICE , PEER , PCA , and LINEAR on each of the 100 datasets using the method settings and parameters as described above . To evaluate performance for each method , we ranked the eQTL for each method according to their p-values and then calculated the True Positive Rate ( TPR ) and False Positive Rate ( FPR ) and generated Receiver Operating Characteristic ( ROC ) curves for each method , where we also calculated the area under the curve for each method across the simulation scenarios . True eQTL were further labeled as cis- , trans- or broad impact and the recovery rate for each category at different FDR thresholds was calculated by dividing the number of true genotype phenotype pairs that were called significant by the total number of true genotype phenotype pairs in each category . To provide an upper bound metric on how well methods could recover each of these eQTL types , we also simulated the same scenarios without any confounding factors and reported the ROC curves after running LINEAR . We labeled these results ‘TMR’ for ‘Theoretical Maximum Recovery’ since these represent the maximum recovery expected in theory if confounding factors were perfectly modeled by the confounding factor methods . We analyzed data from the Multiple Tissue Human Expression Resource ( MuTHER ) [10] project , the Depression Genes Networks study ( DGN ) [13] , the Netherlands Study of Depression and Anxiety ( NESDA ) [85] , and from the Genotype-Tissue Expression ( GTEx ) consortium [14] to compare the performance of the methods and to potentially identify broad impact eQTL in humans . Given that true eQTL are not known for human data we used replication as a metric for performance . While this is an imperfect metric and will tend to undercount true positives , replication does provide relative control over non-systematic false positives , such that a method that is overly liberal in calling of eQTL false positives will be appropriately assessed . We ran eQTL analysis on the adipose , lymphoblastoid cell line ( LCL ) , and skin datasets obtained through the MuTHER project [10] . Based on the matched twins information , there were 161 monozygotic and 220 dizygotic twin pairs in the dataset . We only selected samples that had both genotype and gene expression measurements for both individuals in each twin pair for all three tissue types . To assess replication within a tissue type , we split each tissue specific dataset into two subsets separating each twin pair into different subsets . This created two subsets for each tissue type resulting in 327 samples for adipose , 329 for LCL , and 253 samples for skin ( Table 1 ) . For each subset there were 28 , 964 genes in Adipose , 28 , 894 genes in each LCL , and 28 , 893 genes in Skin . Genotype information was provided by the TwinsUK consortium , and we used only non-imputed genotypes from the downloaded data with minor allele frequencies higher than 5% ( a total of 246 , 298 genotypes ) . We also analyzed data from the DGN [13] and NESDA [85] studies . These independent studies analyzed blood samples and have large sample sizes . Normalized gene expression measurements and genotype files were obtained for DGN that were analyzed previously [13] . Genes which could not be unambiguously mapped to an Entrez Gene ID were excluded as well as SNPs which were not present in dbSNP . In the final DGN dataset there were 922 samples with 15 , 169 genes and 719 , 149 genotypes . Genotype data , gene expression data , and information regarding twin pairs for NESDA were downloaded via dbGaP ( phs000486 . v1 . p1 ) . SNPs with minor allele frequency less than 0 . 05 or which were not present in dbSNP were excluded . In the final NESDA datasets there were 641 , 753 genotypes and 45 , 137 genes with expression level measurements . To match the sample sizes in the two datasets to be within a similar size range for assessing replication , we split the NESDA dataset by available twin status information similar to the strategy used in the MuTHER analysis . This resulted in two subsets from the NESDA dataset with 636 samples in each subset ( Table 2 ) . For the analysis of the GTEx datasets , we selected 4 pairs of tissues ( Adipose , Artery , Heart , Skin ) from GTEx release v6 ( dbGaP Accession phs000424 . v6 . p1 ) with over 150 samples that have both RNA-seq gene expression and SNP array genotypes ( Table 3 ) . For gene expression , we included all genes which could be unambiguously mapped to Entrez Gene IDs ( 24 , 686 genes ) . Within each tissue , we excluded any genes which had zero measurements in more than 80% of samples as well as genes with highly skewed distributions , with more than 85% of measurements in the top or bottom 20% . After these filters were applied , the number of genes for each tissue was between 19 , 207 and 20 , 108 . For genotypes , we excluded SNPs with missing genotypes and those with minor allele frequency <0 . 05 . We also pruned SNPS within 10kb with pairwise r2 > 0 . 99 and removed SNPs which were deprecated in dbSNP ( 1 , 270 , 565 SNPs remaining ) . We fit CONFETI-I , CONFETI-P , PANAMA , ICE , PEER , PCA , and LINEAR for every phenotype and genotype pair in each of the datasets using the method settings and parameters as described above . To control for population structure , we included principal components derived from the genotypes , using the first five ( DGN analysis ) , three ( MuTHER analysis ) , and two ( GTEx and NESDA analysis ) principal components as covariates in each analysis . While a permutation approach is often applied to avoid any systematic inflation or deflation of the p-values , this was computationally infeasible for this study given the number of datasets analyzed and number of methods applied to each dataset . We therefore calculated the genomic inflation factor λ for each expression phenotype , a statistic which has been shown to provide a good metric for assessing model fit and appropriate p-value distributions [70 , 72] . The λ statistic was calculated per gene using the median p-value mp as λ p = qchisq ( 1 - m p ) / qchisq ( 0 . 5 ) ( 12 ) where qchisq is a quantile function for the chi-square distribution with 1 degree of freedom . For each method we assessed inflation using λp values for every gene to calculate λdiff , p = 1 − λp . After calculating p-values for all phenotype and genotype pairs , we adjusted the p-values using Benjamini-Hochberg multiple hypothesis correction . The corrected p-values represent upper bounds on False Discovery Rate ( FDR ) [100] . We used a threshold of 0 . 01 on the adjusted p-values to mark significant eQTL . An eQTL ( significant SNP gene pair ) was labeled as cis- if the SNP and gene were located on the same chromosome within 1 Mb , and trans- otherwise . To avoid potential artifacts caused by ambiguous RNA-seq alignment we screened trans-eQTL using two methods . First , we used annotated gene relationships available from NCBI ( ftp://ftp . ncbi . nih . gov/gene/DATA/gene_group . gz ) to identify trans-eQTL where the SNP was within 1Mb of a gene related to the eQTL gene ( such as a pseudogene or functional gene ‘parent’ of a pseudogene ) . Because not all gene relationships were captured in the NCBI annotation , we searched for additional , potentially unannotated pseudogenes using the BLAT tool [101] to align all gene transcripts to the genome and identified all genomic regions matching at least 50% of each transcript . We omitted any trans-eQTL where the SNP was within 1Mb of a region matching the eQTL gene transcript . This “pseudo-trans” screening revealed that a number of the replicating trans-eQTL were artifacts arising due to incorrect/ambiguous mapping of RNA-seq reads that are in fact caused by cis-regulation of a gene , which shares sequence similarity with the eQTL gene . We also visually inspected eQTL for artifact or false positive indicators ( e . g . , individual genotype associations inconsistent with local linkage disequilibrium ) . In order to avoid double-counting eQTL associated with multiple linked SNPs , we selected at most one significant cis- and trans- SNP per cytoband per gene . Using this criteria , we measured the replication of eQTL between and across different tissues counting the overlapping cytoband and gene pairs that were called significant in each dataset . We marked broad impact eQTL by searching for genotypes that showed more than a single trans-eQTL associations on different chromosomes that replicated between at least one twin or tissue pair .
In our analysis of simulated data , we assessed the performance of the eQTL analysis methods CONFETI-I , CONFETI-P , PANAMA , ICE , PEER , PCA , and LINEAR on their ability to identify three types of eQTL , cis- , trans- and broad impact , in the presence of confounding factors . We also included the theoretical maximum recovery ( TMR ) as an upper limit of eQTL detection for each eQTL category , where the phenotype data has only normally distributed random noise added without any confounding factor effects . For both sparse and dense confounding factor effects , all methods showed significant improvements over LINEAR ( linear regression without confounding factor correction ) , and CONFETI-I correctly identified the most eQTL at every FDR threshold ( Fig 2 ) . We found that linear mixed model based methods recovered individual cis- and trans-eQTL more accurately in comparison to linear fixed effect based correction methods PEER and PCA , where one explanation for this observation could be the lower power of fixed effect correction models by the increased number of parameters [102] . For broad impact eQTL in particular , CONFETI-I and CONFETI-P outperformed all other methods by a large margin illustrating the value of distinguishing genetic and non-genetic factors in the correction . The difference between the confounding factor methods decreased with a combination of sparse and dense confounding factors compared to cases with just sparse confounding factors ( S3 Fig ) , although the general trends remained consistent . This is likely due to the relative amount of total variance explained by each confounding factor and broad impact eQTL . In the dense confounding factor scenario , the confounding factors contribute a significantly higher proportion of the total variance compared to broad impact eQTL . In such a case , distinguishing genetic variance from non-genetic variance has less influence on the covariance matrix correction , since the majority of the variation in the data is originating from the confounding factors , and the resulting difference between methods in identifying true eQTL is expected to be smaller . Overall , approaches such as PANAMA , ICE , PEER , and PCA which do not explicitly remove genetic effects from their correction , increased the accuracy in identifying individual cis- and trans-eQTL but incorrectly modeled broad impact eQTL as confounding factors . While the extent to which any simulated data will capture the true confounding factor conditions and genetic architectures of real eQTL datasets is unknown , these simulations demonstrate that the CONFETI framework can provide a considerable performance improvement compared to mixed model confounding factor methods in some situations , and performed at least as well as other methods overall . We ran each of the eQTL analysis methods on the six datasets from MuTHER [10] ( twin pairs in Adipose , LCL and Skin Tissues ) , the DGN [13] and NESDA [85] datasets ( blood ) , and eight datasets from GTEx [14] ( Adipose , Visceral vs . Subcutaneous; Artery , Aorta vs Tibial Artery; Heart , Atrial Appendage vs . Left Ventricle; Skin , Leg vs . Suprapubic ) . For each method applied to each dataset , we inspected the median λ genomic inflation factor [103] as a measure of appropriate model fit and control of false positives and false negatives rates . Linear mixed model based correction methods showed a slight inflation in comparison to linear fixed effect based methods with ICE showing the highest degree of inflation of p-values in every dataset . Overall , all methods were within acceptable fit levels of inflation or deflation when including genotype PCs as covariates ( S4 Fig ) . When considering different significance thresholds for individual datasets , we found that cis-eQTL discovery starts to asymptote while trans-eQTL discovery does not ( Fig 3 , S5 and S6 Figs ) . This is consistent with the overall smaller effect size of trans-eQTL , which makes them more difficult to detect . Confounding factor correction methods greatly increased the number of cis-eQTL identified in every dataset in comparison to LINEAR , demonstrating the increase of power by accounting for systematic variation . Linear mixed model correction methods CONFETI-I , CONFETI-P , PANAMA , and ICE identified comparable numbers of cis-eQTL in each dataset , followed by fixed effect correction methods PCA and PEER . Similarly , CONFETI-I , CONFETI-P , PANAMA , and ICE increased the number of identified trans-eQTL . However , the number of trans-eQTL identified by PEER and PCA were comparable or even lower than the results of LINEAR in some datasets . While the DGN analysis yielded almost 3 to 4 fold increase for cis and trans-eQTL identification compared MuTHER and GTEx datasets , both subsets of NESDA found fewer cis-eQTL and similar numbers of trans-eQTL . While the decrease in NESDA sample size produced by splitting the datasets into subsets of twins could have affected the results , we would still expect the number of cis and trans-eQTL discoveries to increase compared to the datasets analyzed in MuTHER and GTEx , which had roughly half the sample size . One potential factor influencing the results might be the higher multiple hypothesis testing correction burden in the subsets of NESDA mainly driven by the additional number of gene expression measurements . However , this alone could not explain the significantly lower number of eQTL found in the NESDA dataset , since we would expect to see a steeper increase of cis-eQTL discoveries at lower FDR thresholds based on the increased sample size compared to MuTHER and GTEx datasets . We investigated the replication of eQTL found in each twin pair in the MuTHER dataset , across the DGN and NESDA datasets , and for each tissue pair in the GTEx datasets . Based on the results of individual tissues , we used a significance threshold of FDR < 0 . 01 to further investigate the replication of eQTL focusing on high confidence results . We found that similar to eQTL discovery in each dataset , confounding factor correction increased the number of replicating cis and trans-eQTL with linear mixed model based methods showing the most significant increase ( S7 and S8 Figs ) . For MuTHER and GTEx , we observed a large number of replicating cis-eQTL in all twin pairs and tissue pairs , respectively , and a significantly lower number of replicating trans-eQTL , a result that was also observed in other studies [18 , 104] . In each twin pair and tissue pair , CONFETI-I , CONFETI-P , PANAMA , and ICE identified similar numbers of replicating cis- and trans-eQTL , which were significantly higher than PCA and PEER . Between linear mixed model correction methods , the majority of eQTL were being found by multiple methods and only a few eQTL were unique to each method . This indicated that linear mixed model based correction increases the power of the model over linear fixed effect corrections , however that the differences between methods in constructing the sample covariance matrix lead to few novel discoveries per dataset ( S9 and S10 Figs ) . Twin pairs showed a higher degree of replication compared to similar tissue pairs , which could be explained by the heterogeneity between tissue subtypes in the GTEx dataset ( Fig 4 ) . The replication ratio for cis-eQTL showed little difference between methods and was considerably higher than the replication ratio of trans-eQTL , which also showed higher variation between methods . For the DGN and NESDA datasets the linear mixed model correction methods showed a higher increase in the number of replicating eQTL over linear fixed effect correction methods . The replication rate between the NESDA subsets were comparable to the results for MuTHER and GTEx , with most of the cis-eQTL identified in both datasets with few unique discoveries . However , due to the imbalance of identified eQTL between the DGN and NESDA datasets , the number of replicating eQTL were limited by the eQTL discovered in the NESDA subsets and resulted in lower replication rates with approximately 10% for cis-eQTL and below 1% for trans-eQTL . We further investigated the results for replicating broad impact eQTL . Before the artifact correction protocol , we found replicating broad impact eQTL in GTEx datasets , but excluding pseudogenes from the replicating eQTLs effectively removed all replicating broad impact eQTL from the GTEx dataset . This is consistent with the findings in a study by Jo et al . [105] , in which the authors state that they were unable to identify any individually significant genes with trans-eQTL after testing the associations between a single locus and all expressed genes in both subcutaneous and visceral subsets . This paper did report rs7037324 and rs1867277 on the 9q22 locus of being associated with TMEM253 and ARFGEF3 in the thyroid tissue , and rs2706381 and rs1012793 on the 5q31 locus to be associated with PSME1 and ARTD10 in skeletal muscle . However , these tissues had no replicates where we could assess eQTL replication across the same broad tissue type and were not included in our analysis . Both the DGN and NESDA studies reported broad impact eQTL separately [13 , 85] , but in our analysis we were unable to find any replicating broad impact eQTL among the datasets . We were able to identify a few broad impact eQTL that replicated in the MuTHER LCL dataset ( Fig 5 ) . Most of these impacted only a few genes in trans , where rs3817963 impacted the highest number of genes ( S1 Table ) , including a cis gene HLA-DRA , and six trans ( CCDC28B , CSNK2A1 , ERG , LIMS1 , RPL34 , XRCC6 ) . The enrichment of regulatory signals in the LCL dataset on the region of chromosome 6 which rs3817963 is located is proximal to the major histocompatibility complex ( MHC ) region , which is critical in immune cell function . A cluster of replicating eQTL on the same region of chromosome 6 was also found in the Adipose and Skin twin pair , however , we only found replicating eQTL associated with genes on different chromosomes in the LCL dataset ( S11 Fig ) . We also found an individual case of a broad impact eQTL in the MuTHER Skin twin pair , which was found by ICE . Using PEER we did identify a genotype impacting two genes ( C8orf82 , MYL5 ) that were a subset of reported broad impact eQTL genes in the study by Small et al . [61] ( S1 Table ) . We did not find all of the broad impact eQTL reported by previous studies in the MuTHER Adipose dataset [10 , 61] , which might be a function of our conservative testing threshold . We therefore used the approach of considering the replicating broad impact eQTL we could identify by focusing only on genotypes with significant cis-eQTL as a strategy for adjusting the significance threshold . While using only a subset of genotypes effectively lowered the significance threshold for identifying eQTL overall and led to the identification of few additional replicating broad impact eQTL , it created little difference overall ( Section 1 in S1 Text ) . We also investigated whether independent components significantly associated with genotypes could be used to identify broad impact eQTL . We found that a number of the components that were marked as candidate genetic effects resembled the significance level of individual eQTL with a small number of highly contributing genes . However , ICA does not have a stringent sparsity restriction in estimating the components , so distinguishing between genes , which are highly contributing to the component and noise is challenging ( Section 2 in S1 Text ) . We note that methods enforcing sparsity in the estimation process of components [106 , 107] could be an alternative to ICA in directly identifying broad impact eQTL from the data .
We have introduced the confounding factor correction framework CONFETI , which uses Independent Component Analysis ( ICA ) to avoid over-correcting genetic effects in eQTL mixed model confounding factor analysis . CONFETI provides an easy to implement solution for a known problem with eQTL confounding factor methods: the tendency of these methods to model the effects of eQTL with broad impacts on many genes as confounding variation . In sum , the CONFETI approach provides a method for finding broad impact eQTL while leveraging the advantages of confounding factor analysis for eQTL discovery , a capability that has not been systematically implemented in currently available confounding factor analysis software . In our real data evaluation of CONFETI and other methods , we found that confounding factor correction methods , especially linear mixed model based methods , increased the findings of replicating eQTL . This was also the case for identifying broad impact eQTL that replicated at a genome-wide significance level between datasets . While we did not find any replicating broad impact eQTL for the GTEx tissue pairs , we did find a number of broad impact eQTL when analyzing the MuTHER LCL dataset . Given that broad impact eQTL appear to have relatively small per gene impacts and the larger sample size of MuTHER compared to the GTEx datasets we analyzed , this supports power , and therefore sample size , as being a critical issue when detecting broad impact eQTL . However , this is clearly not the only critical factor , since only one broad impact eQTL was identified by PEER in the MuTHER Adipose dataset , only one was identified by ICE in the Skin dataset , and no broad impact eQTL were identified when comparing results for the considerably larger DGN and NESDA studies . Given LCL are likely to allow a more controlled and homogeneous measurement of gene expression variation compared to the mixed cell populations sampled in vivo for MuTHER adipose and skin datasets , and the even great heterogeneity across distinct studies of DGN and NESDA , it seems likely that different levels of sampling heterogeneity are also influencing broad impact eQTL discovery . We were not able to replicate a small number of broad impact eQTLs reported by previous studies in the MuTHER Adipose dataset [10 , 61] . One possible explanation could be the lower sample size of our analysis resulting from the splitting of twins in each dataset for replication . Another issue to consider is that both studies had less stringent thresholds for identifying significant trans-eQTL compared to the FDR of less than 1% threshold used in our study . Small et al . narrowed down the targets to investigate by testing a single genotype rs4731702 , which significantly lowered the multiple testing burden [61] , and both studies had a threshold of P <5 × 10−8 which corresponded to an FDR threshold of less than 10% in Grundberg et al . [10] . Given that trans associations are the most prone to statistical false positives , it seems reasonable to view these previous reports of broad impact eQTL with caution . In contrast to humans , broad impact eQTL have been easier to detect in model organisms and trans-eQTL seem less dispersed [58] . Given the landscape of broad impact eQTL in humans , the question is therefore what sample sizes and study conditions will be required to detect broad impact eQTL that are robust ? Answering this question will require more genome-wide eQTL studies with larger sample sizes , more control over heterogeneity , and careful analysis with strategies designed to remove broad impact eQTL false positives . | The discovery of expression Quantitative Trait Loci ( eQTL ) from the analysis of genome-wide genotype and gene expression data has played an important role in the study of cellular processes and complex disease . Here , we introduce CONFETI: Confounding Factor Estimation Through Independent component analysis , an analysis framework that has been designed to identify eQTL with broad impacts on the expression levels of many genes . The CONFETI framework takes advantage of Independent Component Analysis ( ICA ) to separate putative genetic and non-genetic factors in a confounding factor mixed model analysis , such that broad impact eQTL are not corrected out of the analysis as confounding variation . We show that CONFETI has better performance for identifying broad impact eQTL compared to the most widely applied confounding factor correction methods when applied to simulated data . We also applied CONFETI and these same methods to identify eQTL that replicate between twin pairs from the MuTHER consortium , the Depression Genes Networks study ( DGN ) , the Netherlands Study of Depression and Anxiety ( NESDA ) , and common tissue type pairs in the Genotype-Tissue Expression ( GTEx ) consortium . Surprisingly , while CONFETI had comparable replication performance compared to other methods , we were able to identify and replicate a very small number of broad impact eQTL overall . We discuss reports of broad impact eQTL in humans and suggest that they should be interpreted with caution . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"variant",
"genotypes",
"random",
"variables",
"covariance",
"factor",
"analysis",
"twins",
"genetic",
"mapping",
"simulation",
"and",
"modeling",
"multivariate",
"analysis",
"developmental",
"biology",
"mathematics",
"statistics",
"(mathematics)",
"research",
"and",
"analysis",
"methods",
"mathematical",
"and",
"statistical",
"techniques",
"gene",
"expression",
"principal",
"component",
"analysis",
"probability",
"theory",
"heredity",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"statistical",
"methods"
]
| 2017 | An independent component analysis confounding factor correction framework for identifying broad impact expression quantitative trait loci |
Holoprosencephaly ( HPE ) is a remarkably common congenital anomaly characterized by failure to define the midline of the forebrain and midface . HPE is associated with heterozygous mutations in Sonic hedgehog ( SHH ) pathway components , but clinical presentation is extremely variable and many mutation carriers are unaffected . It has been proposed that these observations are best explained by a multiple-hit model , in which the penetrance and expressivity of an HPE mutation is enhanced by a second mutation or the presence of cooperating , but otherwise silent , modifier genes . Non-genetic risk factors are also implicated in HPE , and gene–environment interactions may provide an alternative multiple-hit model to purely genetic multiple-hit models; however , there is little evidence for this contention . We report here a mouse model in which there is dramatic synergy between mutation of a bona fide HPE gene ( Cdon , which encodes a SHH co-receptor ) and a suspected HPE teratogen , ethanol . Loss of Cdon and in utero ethanol exposure in 129S6 mice give little or no phenotype individually , but together produce defects in early midline patterning , inhibition of SHH signaling in the developing forebrain , and a broad spectrum of HPE phenotypes . Our findings argue that ethanol is indeed a risk factor for HPE , but genetically predisposed individuals , such as those with SHH pathway mutations , may be particularly susceptible . Furthermore , gene–environment interactions are likely to be important in the multifactorial etiology of HPE .
Holoprosencephaly ( HPE ) is a congenital anomaly characterized by failure to define the midline of the forebrain and midface [1] . HPE occurs with the remarkable frequency of ∼1∶250 conceptions but , due to intrauterine lethality , live-born prevalence is ∼1∶10 , 000 [2] , [3] . A phenotypic continuum of HPE defects is broadly classified into three categories based on the degree of midline cleavage of the forebrain [4] , [5] . Alobar HPE , the most severe form , is characterized by complete failure to partition the forebrain into left and right hemispheres; semilobar and lobar HPE are progressively less severe subtypes . The spectrum of craniofacial midline anomalies ranges from cyclopia in the most severe cases to single nostril , midface hypoplasia with cleft lip and/or palate , hypotelorism ( abnormally close-set eyes ) and solitary median maxillary central incisor in progressively less affected individuals . Mild facial midline abnormalities , called HPE microforms , can occur in the absence of brain malformations . The etiology of HPE is heterogeneous , with genetic and environmental factors implicated [5]–[7] . Heterozygous mutations in Sonic Hedgehog ( SHH ) pathway components are found in both inherited and sporadic HPE , including SHH itself , the primary receptor PTCH1 , the co-receptors CDON ( also called CDO ) and GAS1 , and the transcription factor , GLI2 [6] , [8]–[11] . HPE is characterized by extensive phenotypic variability; this variability is seen in both sporadic cases and within pedigrees [12] . As many as one-third of mutation carriers in pedigrees do not exhibit a clinical phenotype , and mutations found in many sporadic HPE patients are inherited from unaffected parents [12] , [13] . These observations strongly suggest that heterozygous mutation of HPE genes is insufficient to produce severe anomalies and have led to the proposal that HPE is a multiple-hit disorder which arises from a complex interplay of developmental , genetic ( both mutations and more common modifier alleles ) and environmental factors [5]–[7] , [14] . Consistent with this notion , HPE cases with double genetic variations were recently reported , including individuals in which one mutation was inherited and the other arose de novo [13] . In addition to the evidence for multiple-genetic-hit models of HPE , it has been hypothesized that gene-environment interactions may be involved , with a synergistic combination of genetic and non-genetic insults , but there is little or no direct evidence for this . Epidemiological studies suggest that preexisting maternal diabetes and fetal alcohol exposure are potential non-genetic risk factors for HPE , but such studies have been difficult because they rely on birth prevalence not overall prevalence , resulting in small sample sizes and , consequently , wide confidence intervals and interstudy variability [15] , [16] . A question of public health relevance is whether genetic predisposition to HPE ( e . g . , heterozygous mutation of a SHH pathway component ) sensitizes individuals to environmental agents . The challenges faced by HPE epidemiology suggest that the ability to assess gene-environment interactions in patient cohorts might be impracticable and that animal models are required to address this point [15] , [16] . Although fetal alcohol exposure has been implicated in HPE in animal models , results with the mouse , the model organism that best mirrors human genetic susceptibility to HPE , have been inconsistent , with most strains resistant to ethanol and sensitive strains showing low penetrance of HPE phenotypes [17]–[22] . CDON and BOC are related , cell surface SHH-binding proteins that promote SHH pathway activity as co-receptors with PTCH1 [23]–[28] . We have recently identified heterozygous , loss-of-function CDON mutations in HPE patients , including at least one that arose de novo [9] . However , deletions of one copy of CDON have also been identified in individuals without overt HPE phenotypes [29] . These findings are consistent with the notion that additional events , genetic or environmental , may be required for production of HPE in CDON mutation carriers . Studies with mice support this concept . Cdon−/− mice display HPE with strain-dependent severity [28] , [30] . Cdon−/− mice on a C57BL/6NTac background have semi-lobar HPE with a single nostril with high penetrance , whereas these mice on a 129S6/SvEvTac background ( 129S6 . Cdon−/− mice ) show only HPE microforms with low penetrance . Although Shh+/− mice and Boc−/− mice do not have HPE , removal of one copy of Shh or gene dosage-dependent removal of Boc from 129S6 . Cdon−/− mice results in much more severe HPE phenotypes [25] , [27] . These results suggest that 129S6 . Cdon−/− mice have a largely subthreshold defect in SHH signaling that renders them sensitive to second hits , and they are useful as a model for the multifactorial nature of HPE [25] , [27] . 129S6 mice are resistant to ethanol teratogenesis [18] . In this study , we therefore tested whether Cdon−/− mice of this strain are sensitized to ethanol-induced HPE . We report that loss of CDON and in utero ethanol exposure in 129S6 mice results in synergistic and specific defects in early midline patterning , inhibition of SHH signaling in the rostroventral midline , and a broad spectrum of HPE phenotypes . Therefore , loss of Cdon is sufficient to confer sensitivity to ethanol-induced HPE . Our findings argue that ethanol is indeed a risk factor for HPE , but genetically predisposed individuals , such as those with SHH pathway mutations , may be particularly susceptible . Furthermore , gene-environment interactions are likely to be involved in the multifactorial etiology of HPE .
We adapted a protocol for in utero ethanol exposure used for studies on fetal alcohol spectrum disorders [31] , [32] to assess whether 129S6 . Cdon−/− mice were sensitized to ethanol-induced HPE ( see Materials and Methods , Figure S1 and Table S1 ) . All studies were performed with mice on the 129S6 background , and animals are referred to only by genotype unless otherwise noted . Cdon+/− mice were intercrossed and pregnant females received IP injections of ethanol or saline control at E7 . 0 and 4 hours later ( when embryos are at the gastrulation stage ) . At E8 . 0 , ethanol-treated embryos had , on average , between one and two fewer somite pairs than saline-treated controls , regardless of genotype; ethanol-treated embryos at E9 . 0 and E10 . 0 had a similar deficit in somite numbers despite having many more somites at these stages , indicating that ethanol induced an early , transient developmental delay of approximately two to four hr that was independent of Cdon status ( Table S2 ) . Embryos were then assessed for HPE phenotypes between E10 . 0 and E19 . 0 . Cdon+/+ and Cdon+/− embryos with or without ethanol did not have HPE , and saline-treated Cdon−/− embryos showed only microform HPE at low penetrance , similar to untreated Cdon−/− embryos . In contrast , ∼75% of ethanol-exposed E10 . 0–E19 . 0 Cdon−/− embryos displayed HPE-related phenotypes of varying severity . Therefore , mutation of Cdon and fetal ethanol exposure synergized to produce the HPE spectrum . We examined E10 . 0 embryos ( 30–34 somites ) for defects in rostroventral midline formation by measuring the distance between the left and right nasal pits ( Figure 1A–1E ) . Ethanol-treated Cdon−/− embryos had significantly reduced distance between the nasal pits , as compared to saline-treated Cdon−/− embryos and saline- or ethanol-treated Cdon+/+ embryos , none of which were different from each other ( Figure 1J , 1K ) . Furthermore , whole-mount and section analyses revealed that 13 . 5% ( n = 111 ) of E10 . 0 ethanol-exposed Cdon−/− embryos displayed a severe HPE phenotype , including loss of telencephalic structure , failure to divide the eye field , and absence of Rathke's pouch ( Figure 1A–1I , Table 1 ) . This is a more pronounced phenotype than that seen even in Cdon−/− embryos on a sensitized genetic background [28] . These most severely affected embryos died in utero and at E11 . 0 were in the process of resorption; >70% ( n = 37 ) of the remaining E14 . 0–E19 . 0 Cdon−/− embryos had phenotypes that ranged from lobar HPE ( characterized by a partitioned forebrain with abnormal ventral midline continuity ) with single nostril , deficient philtrum , diminished nasal septal cartilage , and rudimentary vomeronasal organs , to microform HPE ( Figure 2 , Table 1 ) . We note that lobar HPE is a relatively subtle forebrain phenotype also seen in Cdon;Boc double mutants on this background [27] , whereas microform HPE is restricted to the facial midline and is often found in Cdon−/− mutants of a mixed genetic background [30] . Additionally , some ethanol-treated E14 . 0 Cdon−/− mice displayed coloboma ( data not shown ) , an eye phenotype sometimes associated with human HPE [33] . E19 . 0 cranial bone/cartilage preparations revealed that , whereas saline controls of any genotype and ethanol-treated Cdon+/+ mice had normal cranial and palatal bone patterning , up to 71% ( n = 14 ) of ethanol-treated Cdon−/− mice displayed underdeveloped maxillary shelves and a foreshortened and/or fused premaxillary bone , both HPE-associated midline defects [27] , [30] ( Figure 3A–3E , Table 1 ) . Furthermore , four of five ethanol-treated Cdon−/− mice , but none of the other mice , had misshapen primary and secondary palates ( Figure 3K–3N , Table 1 ) . Consistent with these E19 . 0 palate defects , coronal sections of E14 . 0 embryos revealed defective outgrowth of palatal shelves in ethanol-treated Cdo−/− embryos ( Figure 3O–3R ) . The mandible is generally spared in HPE but SHH signaling is required for mandibular development [34] , and agnathia spectrum phenotypes ( i . e . , hypoplasia through complete loss of the mandible ) have been reported in up to 10% of HPE patients [35] . Three of 14 ethanol-treated E19 . 0 Cdon−/− mice displayed agnathia spectrum defects , two with fused , hypoplastic mandibles and one with complete agnathia ( Figure 3F–3J , Table 1 ) . Therefore , the synergistic interaction of loss of Cdon and fetal ethanol exposure resulted in a wide spectrum of HPE defects at high penetrance and also produced anomalies more rarely associated with human HPE , at a similar low penetrance . Cdon−/− mice display HPE with strain-dependent severity . Mice lacking the Cdon paralog Boc do not have HPE , regardless of genetic background , but removal of Boc from Cdon−/− mice enhances their HPE phenotype [27] . We therefore tested whether Boc−/− embryos are also sensitized to ethanol-induced HPE . In contrast to Cdon−/− embryos , Boc−/− embryos exposed in utero to ethanol under the same protocol did not display detectable HPE or facial midline phenotypes at E14 . 0 ( Figure S2 ) . SHH produced by the prechordal mesendoderm ( PCM ) is required for initiating development of the midline of the forebrain and midface [36] . PCM-derived SHH induces expression of Shh itself and SHH pathway target genes in the ventral midline of the developing diencephalon and , subsequently , telencephalon [37] , [38] . Expression of Shh in the ventral diencephalon also requires Six3 , an HPE gene that encodes a homeodomain transcription factor [39] , [40] . We used whole-mount in situ methods to assess Shh and Six3 expression , as well as apoptosis and cell proliferation , at 24–36 hr after the initial in utero ethanol treatment of embryos . Ethanol treatment had no effect on Shh expression in the axial mesoderm at the headfold ( presomite ) stage , or in the PCM and notochord at the 8–9 somite stage ( Figure 4A–4H; Table 2 ) . Furthermore , Six3 was expressed normally in the ventral forebrain of ethanol-treated Cdon−/− embryos of 8–9 somites , just prior to Shh induction in that structure ( Figure S3; Table 3 ) . Ethanol did not induce apoptosis in the midline of 4–6 somite embryos , although the number of TUNEL+ cells in the lateral regions of the anterior neural plate was increased by ethanol in a manner independent of Cdon genotype ( Figure S4 ) . Cell proliferation , as assessed by immunostaining for phospho-histone H3 , was unaffected at the 4–6 somite stage by either ethanol treatment or Cdon genotype ( Figure S4 ) . Taken together , Cdon−/− embryos exposed to ethanol did not display obvious alterations in several aspects of rostroventral midline development 24–36 hr after ethanol treatment . We next examined expression of Shh and SHH target genes in the developing forebrain at stages just subsequent to this period . Shh expression in the diencephalon was initiated normally at the 14–15 somite stage in ethanol-treated Cdon+/+ embryos and in saline-treated Cdon−/− embryos ( Figure 4I–4K ) . In contrast , Shh expression in the rostral diencephalon was strongly reduced in ethanol-exposed Cdon−/− embryos at this stage ( Figure 4L; Table 3 ) . This pattern of diminished Shh expression in ethanol-treated Cdon mutants was also observed at later stages when , in control embryos , the expression zone expanded anteriorly in the diencephalon ( 20–21 somites ) ( Figure 4M–4P; Table 3 ) , and when Shh had been induced in the ventral telencephalon ( 35–36 somites ) ( Figure 4Q–4T; Table 3 ) . Consistent with the reduction in Shh expression in the ventral forebrain of ethanol-treated Cdon−/− embryos , expression of the direct SHH target genes Ptch1 , Gli1 and Nkx2 . 1 was reduced specifically in the ventral forebrain of these embryos at the 35–36 somite stage ( Figure 5A–5H , 5M–5Q ) . Reduction in SHH target gene expression was seen with partial penetrance ( Table 2 ) , similar to the partial penetrance and range of HPE phenotypes produced by this protocol ( Table 1 ) . SHH is required for maintenance of Fgf8 expression in the commissural plate of the rostral telencephalon but not expression at the midbrain-hindbrain boundary [41] , and Fgf8 expression in the former , but not the latter , structure was diminished in ethanol-treated Cdon mutants at this stage ( Figure 5I–5L; Table 2 ) . As a control for specificity of inhibition of SHH-dependent gene expression , we assessed expression of Msx1 and Msx2 , which are markers of the migrating neural crest cells that contribute to craniofacial structures but are not targets of SHH signaling [34] , [42] . The expression patterns of Msx1 and Msx2 were not affected in ethanol-treated Cdon−/− embryos ( Figure S5; Table 2 ) . We conclude that synergistic interaction between mutation of Cdon−/− and ethanol exposure during early development results in delayed and diminished induction of Shh expression specifically in the ventral forebrain , leading to a failure to pattern the rostroventral midline and , consequently , to HPE . To apply a more quantitative approach to these results , RNA was extracted from dissected forebrains of E10 . 0 embryos , and quantitative RT-PCR ( qRT-PCR ) was performed . Although there was some variability , likely due to the partial penetrance observed at the level of in situ hybridization ( Table 1 and Table 2 ) , mRNA levels of Shh , Nkx2 . 1 , Ptch1 , Fgf8 and Gli1 were each reduced in the ethanol-treated Cdon−/− embryonic heads relative to controls ( Figure S6 ) . Moreover , it is likely that the qRT-PCR results underestimate the reduction in expression of these genes in the most affected region of ethanol-treated Cdon−/− embryos ( the rostroventral midline ) because more caudal and lateral forebrain structures , where changes in expression were not obvious , were by necessity included in the dissected tissue . SHH signaling is required for dorsoventral patterning of the neural tube [43] , and Cdon−/− embryos on a mixed 129S6×C57BL/6NTac background show a reduction in FOXA2+ floor plate cells [25] . Additionally , Shh levels were reduced in the floor plate of ethanol-treated Cdon−/− embryos at the 20–21 somite stage ( Figure 4M–4P ) . We therefore tested whether ethanol-exposed Cdon−/− embryos had defective neural tube patterning . Sections of saline- and ethanol-treated control and Cdon−/− embryos were analyzed by immunofluorescence at E10 . 0 for FOXA2 , NKX2 . 2 , NKX6 . 1 , PAX6 and PAX7 , markers whose respective expression zones span the entire dorsoventral axis of the developing neural tube . No differences in the expression patterns of any of these factors were seen between any of the four conditions ( Figure S7 ) . Furthermore , the neural tube of ethanol-exposed Cdon−/− embryos had a normal morphology . It is likely that the reduction in floor plate Shh expression in these embryos does not have an overt effect on further neural tube patterning because notochord-derived SHH was not perturbed and is largely sufficient for this process [43] . Therefore , the synergistic effect of loss of Cdon and in utero ethanol exposure was restricted to the most rostral portion of the ventral midline , resulting in HPE spectrum defects , but not obvious neural tube defects . The timing and location of defects in Shh expression and signaling shown in Figure 3 and Figure 4 are consistent with the HPE phenotypes seen in the majority of ethanol-treated Cdon−/− embryos ( Figure 2 ) . However , the alobar HPE found in 13 . 5% of E10 . 0 embryos is less easily explained by such alterations in SHH pathway activity . Furthermore , it must be presumed that ethanol exerts its effects during the relatively brief window in which Cdon−/− embryos are exposed ( peak levels occur at approximately E7 . 25; Figure S1 ) , even if defects in midline development occur subsequently . Patterning events that result in the formation of the PCM occur during the period of ethanol exposure; this process is regulated by the NODAL signaling pathway , and mutations in NODAL pathway components are observed in human HPE [6] , [44] . Foxa2 and Gsc ( Goosecoid ) are two markers of the anterior primitive streak , from which the PCM is derived , and they function cooperatively to specify anterior mesendoderm [45]–[47] . We found that some ethanol-treated Cdon−/− embryos at the late streak stage ( isolated at E7 . 25–E7 . 5 ) displayed substantially reduced expression of Foxa2 ( 3 of 5 embryos ) and Gsc ( 2 of 4 embryos ) as compared to saline-treated Cdon+/+ controls ( Figure 6A–6H; Table 3 ) . This decrease in Foxa2 and Gsc expression was not due simply to the brief delay in development seen in all ethanol-treated embryos ( see above and Table S2 ) because ethanol-exposed Cdon+/+ embryos did not show obvious defects in Foxa2 or Gsc expression ( Figure 6A–6H; Table 3 ) . Saline-treated Cdon−/− embryos also resembled controls ( Figure 6A–6H; Table 3 ) . Therefore , these defects arose as a consequence of an interaction between ethanol exposure and loss of Cdon . Similar effects on Foxa2 and Gsc expression are seen in embryos with defects in NODAL signaling [48]–[50] . Another derivative of the anterior primitive streak is the anterior definitive endoderm ( ADE ) , and embryos with defective NODAL signaling display decreased expression of the ADE marker Cer1 ( Cerberus 1 ) at the late streak and early allantoic bud stages [49] , [50] . In contrast to Foxa2 and Gsc expression , Cer1 expression at the early bud stage was not obviously altered by ethanol treatment in 5 of 6 Cdon−/− embryos ( Figure 6I–6L; Table 3 ) . The synergy between ethanol exposure and loss of CDON is therefore detectable at a developmental stage prior to SHH function [51] . Consistent with these early effects of ethanol treatment in Cdon−/− embryos , we find that Cdon is expressed at the late-streak and early bud stages in the ectoderm , mesoderm and allantoic bud as assessed by in situ hybridization and ß-galactosidase activity derived from a LacZ reporter knocked into the Cdon locus ( Figure 6M–6O ) . To assess whether treatment with ethanol at this early stage was critical to production of HPE spectrum phenotypes , pregnant females were administered ethanol at E8 . 0 , rather than E7 . 0 , and embryos collected at E14 . 0 for analysis . In contrast to the external features of HPE seen in 13 out 18 Cdon−/− embryos treated at E7 . 0 and analyzed at E14 . 0 ( Table 1 ) , none of 13 Cdon−/− embryos treated at E8 . 0 displayed such phenotypes when analyzed at this stage ( Figure S8 ) .
CDON functions as a co-receptor in the SHH pathway , binding both to Hedgehog ligand and to the primary receptor , PTCH1 [9] , [52] . Loss-of-function missense mutations in CDON identified in human HPE cases result in variant proteins that do not support ligand-dependent signaling and display defective interactions with PTCH1 [9] . Previous studies with Cdon−/− embryos on sensitive and resistant genetic backgrounds ( C57BL/6NTac and 129S6 , respectively ) argued that loss of CDON results in disruption of SHH signaling in and/or from the PCM , leading to defective induction of Shh and SHH target gene expression in the ventral forebrain , with consequent effects on forebrain and facial midline patterning [27] , [28] , [30] . The precise timing and location of the first observable defect in Shh expression in ethanol-treated 129S6 . Cdon−/− embryos ( between the 8–9 and 14–15 somite stages in the rostroventral diencephalon ) , as well as the severity of the forebrain and facial midline phenotypes in the majority of such embryos ( lobar HPE with strong midfacial anomalies ) , is similarly consistent with a defect in SHH signaling from the PCM to the presumptive ventral forebrain and/or responsiveness of the latter to PCM-derived inductive signals . However , 13 . 5% of ethanol-treated Cdon−/− embryos had severe HPE at E10 . 0 ( and were being resorbed by E11 . 0 ) , a stronger phenotype than that anticipated by the above mechanism . Furthermore , ethanol treatment at E8 . 0 , rather than the standard E7 . 0 , did not produce HPE in Cdon−/− embryos . The PCM arises from the anterior primitive streak . Development of the anterior primitive streak occurs during the time of embryonic ethanol exposure with E7 . 0 administration , and defects in this developmental process are associated with HPE [6] , [44] . We therefore analyzed expression of Foxa2 and Gsc , two genes that mark the anterior streak and function to pattern the PCM [45]–[47] . Fifty-to-sixty percent of ethanol-treated Cdon−/− embryos displayed substantially reduced expression of these genes . Although ethanol caused a transient , two- to four-hour delay in development that was independent of embryo genotype , only Cdon−/− embryos showed decreased expression of Foxa2 and Gsc in response to ethanol . Nodal+/−;Gdf1−/− mice and Nodal+/−;Chrd−/− mice , which have defective NODAL pathway signaling ( and , at least in the case of Nodal+/−;Chrd−/− mice , overactive BMP signaling ) show a similar diminution of Foxa2 and Gsc expression and develop HPE [48]–[50] . This stage of development is prior to expression of Shh , but we report here that Cdon is expressed at the late streak and early bud stages . These results suggest that CDON plays an earlier role in development than its known role as a SHH co-receptor . CDON is a multifunctional co-receptor , and promotes signaling in a Hedgehog-independent manner when associated with various other cell adhesion molecules and signaling receptors [53] , [54] . Perhaps CDON is also able to function with NODAL , BMP or other ligands or their antagonists; it should be noted , however , that embryos with defective NODAL pathway signaling also display reduction in expression of the ADE marker , Cer1 [49] , [50] , and this was not observed in ethanol-treated Cdon−/− embryos . Although the mechanism whereby CDON exerts effects in primitive streak stage embryos is not clear , the need for ethanol exposure to reveal this role suggests that , in the absence of additional insults , it is subtle or redundant with other factors . A model consistent with all these data is that CDON can function at multiple points in rostroventral midline patterning , one of which is via promotion of SHH signaling . In 129S6 mice , ethanol initiates defects in midline patterning specifically in genetically sensitized ( i . e . , Cdon−/− ) embryos , with the variable severity of the HPE phenotype – ranging from severe HPE to no overt effect beyond that associated with loss of Cdon alone on this background – arising stochastically . It will be interesting to test in the future whether mice carrying mutations specific for the SHH pathway ( e . g . , Shh+/− mice ) are sensitized to ethanol treatment; similarly , animals potentially sensitized by heterozygosity for NODAL pathway components could also be investigated . A previous study on chick embryos treated at various developmental stages with the Hedgehog pathway inhibitor cyclopamine concluded that a phenotypic HPE spectrum could be produced by varying the timing of SHH pathway blockade; i . e . , early inhibition led specifically to severe phenotypes , while later time points of inhibition led to progressively less severe defects [37] . Our findings indicate that brief exposure to a teratogen early in development can produce a similar broad range of phenotypes in a genetically susceptible host: transient ethanol exposure during gastrulation of 129S6 . Cdon−/− mice , produced not only alobar HPE , which is associated with early patterning defects , and but also milder forms of HPE with HPE-related phenotypes that are associated with much later patterning defects ( e . g . , in palatogenesis ) . Fetal alcohol exposure has been linked to HPE and alterations in SHH pathway activity in other animal models , including mice and zebrafish [17]–[22] , [55] , but a specific genetic interaction between ethanol and the SHH pathway has not been reported . Most mouse strains , including 129S6 , are resistant to ethanol , and ethanol induces HPE only with low penetrance even in the most widely studied strain , C57BL/6J [17]–[19] . In zebrafish , ethanol produces cyclopia but also severe defects along the entire anterior-posterior axis [21] . In contrast , the 129S6 . Cdon−/− plus in utero ethanol-treatment model is notable for its specificity , including: 1 ) timing of exposure ( administration of ethanol at E7 . 0 but not E8 . 0 was effective ) ; 2 ) structures affected ( the ventral forebrain and craniofacial midline displayed defects but the neural tube did not ) ; and 3 ) mutation of a bona fide HPE gene , Cdon , but not an HPE modifier gene , Boc , synergized with ethanol . Therefore , this model incorporates major known and predicted features of human HPE , including: 1 ) a multifactorial etiology that reveals gene-environment interactions in the specific inhibition of SHH pathway activity in the rostroventral midline; and 2 ) a broad spectrum of HPE phenotypes , including low penetrance phenotypes such as agnathia . Genetic removal of Cdon plus or minus removal of the paralogous gene Boc on two different genetic backgrounds results in animals that display distinct windows within the range of HPE phenotypes observed in human cases ( e . g . , in individuals heterozygous for loss-of-function SHH mutations ) ( Figure 7 ) . However , no combination of gene loss and strain background resulted in as wide a spectrum of HPE phenotypes as that seen in the human population , even within pedigrees . In contrast , ethanol-treated Cdon−/− mice display a nearly complete HPE spectrum , including low penetrance phenotypes ( Figure 7 ) . These findings suggest that interaction between genetic and teratogenic factors may well underlie many HPE cases and have considerable public health and clinical implications . Approximately 12% of pregnant women use alcohol and 2% do so heavily ( i . e . , “binge” drinking ) [56] . Furthermore , the time of exposure in this model [31] is equivalent to the third week of human gestation , a time when many women are unaware they are pregnant . Epidemiological evidence for ethanol as an HPE risk factor is ambiguous [15] , [16] . The results presented here argue that ethanol is indeed a risk factor for HPE , but genetically predisposed individuals , such as those with SHH pathway mutations , may be particularly susceptible . Because 129S6 . Cdon−/− mice are capable of identifying both genetic ( e . g . , Boc ) and non-genetic ( e . g . ethanol ) risk factors for HPE that are insufficient to produce HPE alone ( i . e . , silent cofactors ) ( [25] , [27]; this study ) , we propose that they serve as a potential animal model for the assessment and identification of plausible risk factors for HPE .
All animal work was approved by the Institutional Animal Care and Use Committee ( IACUC ) . Our animal facility is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International ( AAALAC ) . Two to three month-old Cdon+/tm1Rsk ( Cdon+/− ) mice on a 129S6/SvEvTac ( 129S6 ) background [28] , [30] were mated for one hour in the dark and plugged females were collected . The time of the plug was designated as embryonic day ( E ) 0 . 0 . Ethanol administration was performed as described by Sulik et al . and Webster et al . [31] , [32] with slight modification . In initial experiments , pregnant female mice were injected intraperitoneally twice with 15 µl per gm body weight of a solution of 25% ethanol in saline ( 2 . 9 gm/kg ) , at E7 . 0 and 4 hr later . Saline injections were used as a control . Offspring were examined at E14 . 0 , and 22 . 2% of ethanol-treated Cdon−/− embryos ( but not Cdon+/+ or Cdon+/− embryos ) displayed external signs of HPE ( n = 56 ) . Injected females were also assessed for blood alcohol concentration over 10 hr after the first injection with the Pointe Scientific Alcohol Reagent Set ( A7504 ) ( Figure S1 ) . As previously reported [32] , blood alcohol concentrations peaked 1 hr after the second injection , but the concentration achieved in 129S6 mice was only ∼60% that described for C57BL/6J mice , the most widely-used strain . We therefore increased the dose by 20% ( 15 µl per gm body weight of a solution of 30% ethanol in saline; 3 . 48 gm/kg ) . The blood alcohol levels achieved were very similar to those reported previously with the 2 . 9 gm/kg dose in C57BL/6J mice [32] , and >70% of E14 . 0 Cdon−/− embryos ( but not Cdon+/+ or Cdon+/− embryos ) under this protocol displayed external signs of HPE ( Table S1 ) . This protocol was used in the studies reported here unless noted otherwise . In some experiments 129S6 . Boc+/− mice [27] were used in place of Cdon+/− mice and in others , plugged Cdon+/− females received ethanol at E8 . 0 rather than E7 . 0 . 129S6 mice are resistant to ethanol teratogenicity ( [18]; this study ) , and we note that these concentrations are above those generally achieved clinically in humans . Among more than 150 pregnant females treated with ethanol , ∼40% lost their litters in less than 24 hr but none died of acute toxicity . Results reported are from litters that survived this initial period , after which resorptions were unusual ( Table S1 ) . For all analyses , embryos were collected at the embryonic day indicated in the text . At presomitic stages , embryos were further staged by morphology and in situ hybridization . For E8 . 0 and later , embryos of equivalent somite number were compared . Although ethanol-treated embryos had , on average , between one and two fewer somite pairs than saline-treated controls regardless of genotype ( see Results ) , analysis of multiple embryos that differed by one to two somites at the various stages shown in Figure 4 demonstrated that this transient delay was not the cause of the altered gene expression patterns seen in ethanol-treated Cdon−/− embryos . Embryos were dissected out and fixed overnight in 4% paraformaldehyde in PBS . Embryos were then dehydrated through a graded ethanol series , embedded in paraffin and sectioned at 8 µm . Hematoxylin and eosin ( H&E ) staining was performed as described [27] . Slides were then dehydrated through graded ethanol and xylene and mounted with Permount ( Fisher Scientific ) . Immunofluorescence analysis of developing neural tubes was performed on frozen sections as described [57] , and images were taken on a Zeiss Axioplan 2 microscope . The number of cells expressing a particular marker , or the relative size of marker expression domains , was measured with ImageJ software . The antibodies used were: mouse anti-FOXA2 ( 1∶20; DSHB ) , mouse anti-NKX2 . 2 ( undiluted; DSHB ) , mouse anti NKX6 . 1 ( 1∶20; DSHB ) , mouse anti-PAX6 ( 1∶20; DSHB ) , mouse anti-PAX7 ( 1∶20; DSHB ) , and Alexa 488 anti-mouse IgG ( 1∶500; Invitrogen ) . For whole mount in situ hybridization , embryos were prepared as described previously [58] , except that they were treated with 10 µg/ml proteinase K ( QIAGEN ) in phosphate-buffered saline , 0 . 1% Tween-20 ( PBT ) according to stage . Embryos were rinsed , postfixed and hybridized with digoxygenin-labeled probes in hybridization mix [50% formamide , 1 . 3× SSC , 5 mM EDTA , 50 µg/ml yeast RNA , 0 . 2% Tween 20 , 0 . 5% 3-[ ( 3-cholamidopropyl ) dimethylammonio] propanesulfonate , and 100 µg/ml heparin] overnight at 65°C . After washing and blocking , embryos were incubated overnight with alkaline phosphatase-conjugated anti-digoxigenin antibody ( 1∶2000; Roche ) in blocking buffer ( 2% blocking reagent [Roche] ) , 20% heat-inactivated lamb serum in 100 mM maleic acid , pH 7 . 5 , 150 mM NaCl , and 0 . 1% Tween 20 [MABT] ) . After washing in TBST ( Tris-buffered saline with 0 . 1% Tween-20 ) and ( NTMT ) 100 mm NaCl , 100 mm Tris-HCl , pH 9 . 5 , 50 mm MgCl2 , and 0 . 1% Tween −20 , signals were developed with BM Purple AP Substrate ( Roche ) . In situ terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling ( TUNEL ) assay was performed according to the manufacturer's instructions ( Roche ) . Whole mount immunohistochemistry for phospho-histone H3 ( Upstate , 06-570 ) was performed with embryos fixed overnight in 4% paraformaldehyde in PBS [59] , [60] . Embryos were dehydrated through a graded methanol series to 100% . Endogeneous peroxidase was inactivated by a one-hour treatment in 5% hydrogen peroxide in methanol . Embryos were rehydrated into 80% , 50% and 20% methanol in PBT ( PBS+0 . 75% tween-20 ) and were washed for 10 minutes in PBT . Embryos were then incubated with blocking solution ( PBT+10% goat serum ) for one hour and incubated overnight at 4°C with anti phospho-Histone H3 antibody at a dilution of 1∶100 in blocking solution . Embryos were subsequently washed in PBT four times over one hour and then incubated overnight at 4°C with biotinylated anti-rabbit IgG ( Vector Laboratories ) at 1∶200 in blocking solution . After washing 4 times with PBT , embryos were then incubated with HRP-conjugated biotin-avidin as instructed by the manufacturer ( Vector Laboratories , Vectastain ABC kit , PK-4000 ) . Staining was developed with Sigma Fast DAB substrate ( Sigma , D4418 ) . Stained embryos were cleared in 80% glycerol and photographed with a Jenoptik ProgRes C3 camera attached on a Nikon SMZ 1500 stereomicroscope . Captured images were assembled by Helicon Focus software ( Helicon Soft ) . Dissected embryos were stained for ß-galactosidase activity as described previously [27] , [28] , [30] with the following modification . Embryos were fixed with 2% paraformaldehyde and 0 . 2% glutaraldehyde in phosphate-buffered saline ( PBS ) for 10 minutes on ice and stained in PBS , pH 7 . 0 , 2 mM MgCl2 , 0 . 01% NP-40 , and 0 . 02% sodium deoxycholate , 17 . 5 mM each K3Fe ( CN ) 6 and K4Fe ( CN ) 6 and 1 mg/ml 5-bromo-4-chloro-3-indolyl-β-d-galactoside ( Roche ) . Stained embryos were cleared in 80% glycerol and PBS for photography . Bones and cartilage of E19 embryos were stained with alizarin red and alcian blue as described [61] . Briefly , embryos were collected , fixed in 95% ethanol for 4–5 days and transferred to acetone for 3 days . The embryos were then rinsed with water and stained for 24 hours in 0 . 05% Alcian Blue in 20% glacial acetic acid in 95% ethanol . After washing in 95% ethanol for 2–3 days , soft tissues were dissolved in 1% KOH for 1 hour and stained in 0 . 75% Alizarin Red in 1% KOH for 4 hours . Stained embryos were kept in 20% glycerol/1% KOH until skeletons became clearly visible . Embryos were transferred through 50% , 80% and 100% glycerol for photography and storage . qRT–PCR analysis of Shh , Gli1 , Ptch1 , Nkx2 . 1 and Fgf8 expression was performed on E10 . 0 control ( Cdo+/− or Cdo+/+ ) and Cdo−/− embryos . Embryonic forebrains were dissected out and transferred into 100 µl of RNAlater ( Qiagen ) . Total RNA was purified using RNeasy Mini Kit ( Qiagen ) . cDNA was synthesized with the Superscript III first strand synthesis system ( Invitrogen ) . qPCR was performed using PerfeCta SYBR Green FastMix for iQ ( Quanta bioscience ) with Bio-Rad iCycler iQ5 . Data were normalized to Gapdh levels and presented as fold change over control . qRT-PCR primers were from the Harvard PrimerBank ( Primerbank IDs: Gapdh , 6679937a1; Shh , 21617861a1; Gli1 , 6754002a1; Ptch1 , 6679519a1; Fgf8 , 22094093a1 ) and reference [62] ( Nkx2 . 1 ) . | Holoprosencephaly ( HPE ) , a congenital anomaly characterized by failure to form the midline of the forebrain and midface , occurs as frequently as 1 in 250 conceptions . Mutations in genes that direct formation of the forebrain and facial midline are associated with HPE , but the clinical outcome is extremely variable and many mutation carriers are unaffected . This has led to the hypothesis that more than one genetic insult is required to cause HPE . Although multiple mutations have been identified in some individuals with HPE , they represent a small percentage of the total . Non-genetic risk factors are also implicated in HPE , and it is possible that this defect may arise as a consequence of a mutation interacting with an environmental exposure . However , evidence for this possibility is lacking . We have developed a mouse HPE model in which there is dramatic synergy between mutation in the Cdon gene , which is mutated in some HPE patients , and in utero exposure to ethanol , a suspected but unproven HPE risk factor . Our findings argue that ethanol is a risk factor for HPE , but genetically predisposed individuals may be particularly susceptible . Furthermore , gene–environment interactions are likely to be important in the multifactorial etiology of HPE . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"developmental",
"biology",
"genetics",
"biology",
"genetics",
"and",
"genomics",
"toxicology"
]
| 2012 | Cdon Mutation and Fetal Ethanol Exposure Synergize to Produce Midline Signaling Defects and Holoprosencephaly Spectrum Disorders in Mice |
T cell immunoglobulin and mucin protein 3 ( TIM-3 ) is a type I cell surface protein that was originally identified as a marker for murine T helper type 1 cells . TIM-3 was found to negatively regulate murine T cell responses and galectin-9 was described as a binding partner that mediates T cell inhibitory effects of TIM-3 . Moreover , it was reported that like PD-1 the classical exhaustion marker , TIM-3 is up-regulated in exhausted murine and human T cells and TIM-3 blockade was described to restore the function of these T cells . Here we show that the activation of human T cells is not affected by the presence of galectin-9 or antibodies to TIM-3 . Furthermore , extensive studies on the interaction of galectin-9 with human and murine TIM-3 did not yield evidence for specific binding between these molecules . Moreover , profound differences were observed when analysing the expression of TIM-3 and PD-1 on T cells of HIV-1-infected individuals: TIM-3 was expressed on fewer cells and also at much lower levels . Furthermore , whereas PD-1 was preferentially expressed on CD45RA−CD8 T cells , the majority of TIM-3-expressing CD8 T cells were CD45RA+ . Importantly , we found that TIM-3 antibodies were ineffective in increasing anti-HIV-1 T cell responses in vitro , whereas PD-L antibodies potently reverted the dysfunctional state of exhausted CD8 T cells . Taken together , our results are not in support of an interaction between TIM-3 and galectin-9 and yield no evidence for a functional role of TIM-3 in human T cell activation . Moreover , our data indicate that PD-1 , but not TIM-3 , is a promising target to ameliorate T cell exhaustion .
Inhibitory costimulatory signals play a decisive role in the outcome of T cell responses and there is an ever-growing number of pathways that have been implicated in such processes . The successful clinical use of CTLA-4 antibodies to enhance anti-melanoma T cell responses underlines the therapeutic potential of antibodies targeting negative costimulatory T cell pathways [1] . During the last years , it has been acknowledged that inhibitory pathways also significantly contribute to the exhausted state of T cells , which results from persistent antigen stimulation in chronic virus infections or cancer . The inhibitory PD-1 was identified as a marker for such dysfunctional T cells and blockade of PD-1 signals - in most cases realized with antibodies to PD-L1 - was shown to revert the dysfunctional state of exhausted T cells [2]–[4] . Peretz et al . have found that co-expression of PD-1 and CD160 , another inhibitory receptor , defined a subset with advanced dysfunction [5] . We have recently shown that PD-1 is up-regulated in CD4 and CD8 T cells from HIV-infected individuals , who have impaired immune reconstitution despite successful antiretroviral therapy , indicating that PD-1 blockade might be beneficial also in these patients [6] . PD-1 is also a promising target to improve T cell immunity in cancer patients and clinical trials to assess the safety and efficacy of blocking PD-1 and PD-L1 antibodies have been initiated [7] , [8] . Additional inhibitory receptors have been demonstrated to be associated with a dysfunctional phenotype and Blackburn et al . have shown that exhausted T cells express up to seven different inhibitory receptor molecules [9] . Several studies indicate that among these molecules , the T cell immunoglobulin and mucin domain 3 ( TIM-3 ) has an important role in maintaining the dysfunctional state of exhausted T cells , as TIM-3 blockade restored proliferation and cytokine production upon antigenic challenge in these cells [10]–[13] . Consequently , it has been suggested that blocking TIM-3/TIM-3 ligand interactions might be of clinical utility by restoring the function of virus or tumour-specific T cells . However , TIM-3 appears to be a pleiotropic immune receptor and many aspects of human TIM-3 functions have not been clarified to date . Murine TIM-3 was reported to negatively regulate T cell responses via interacting with galectin-9 [14] but currently it is not known whether human TIM-3 acts as galectin-9 receptor and the functional role of galectin-9 during the activation of human T cells has not been investigated . Furthermore , apart from a study on NY-ESO-1 specific CD8+ T cells derived from melanoma patients , the effects of TIM-3 and PD-L1 blockade have not been compared [12] . In this study , we have addressed the functional role of galectin-9 during the activation of human T cells . In addition , we have performed extensive experiments to study the interaction between TIM-3 and galectin-9 molecules of human and mouse origin . Finally , we used Gag/Nef - specific T cells derived from HIV-1-infected individuals as model system to compare the ability of antibodies to TIM-3 and PD-ligands to counter-act T cell dysfunction and exhaustion .
Previously , it was reported that galectin-9 binds murine TIM-3 , thereby negatively regulating murine T cells [14] . Since little is known regarding the functional role of galectin-9 on human T cells , we performed experiments in which T cells were activated in the presence of human galectin-9 . In these experiments , we used our previously described system of T cell stimulator cells that can stimulate human T cells via membrane-bound anti-CD3 antibody fragments [15] . Control stimulator cells and stimulator cells expressing high levels of human galectin-9 were established and analyzed for expression of anti-CD3 antibodies and galectin-9 ( Figure 1A ) . Human T cells were co-cultured with these stimulator cells and T cell proliferation and cytokine production were assessed . As shown in Figure 1B , the presence of galectin-9 during stimulation did not result in a reduced proliferative response in human T cells . By contrast , stimulator cells expressing the well-established inhibitory ligand PD-L1 , used under the same conditions , significantly reduced human T cell proliferation , whereas activating costimulatory ligands ( CD80; CD86 ) strongly enhanced T cell proliferation when expressed on the stimulator cells ( Figure S1A ) . Furthermore , we found that the production of cytokines was also not affected by human galectin-9 ( Figure 1C ) . Whereas galectin-9 was reported to induce cell death in murine T cells [14] , we could not observe such effects in human T cells: In T cells activated with T cell stimulator cells expressing galectin-9 there was no increase in apoptotic T cells . By contrast , the number of T cells undergoing apoptosis was higher when T cell stimulator cells expressing human FasL were used to activate T cells ( Figure S1B ) . Previous studies reported that the inhibitory effect of galectin-9 on murine T cells is mediated at least in part by TIM-3 [14] , [16]–[18] . Although galectin-9/TIM-3 interaction was reported to protect human CD4 T cells against HIV-1 infection [19] , to date there are no reports demonstrating a specific binding of human galectin-9 to human TIM-3 . Therefore , we performed a series of experiments to study a potential interaction between these two molecules . In the first set of experiments , we probed Bw cells expressing human galectin-9 on their surface with an immunoglobulin fusion protein representing the extra-cellular domain of TIM-3 fused to the Fc-part of human IgG1 ( TIM-3-Ig ) or with B7-H3-Ig used as control . In these experiments , a specific interaction between cell-expressed human galectin-9 and TIM-3-Ig was not observed ( Figure 2A ) . Binding of CD80-Ig to CD28 expressing Bw cells was readily detected under the same experimental conditions even to cells that expressed low levels of CD28 ( Figure S2A ) . In addition , we probed Bw cells expressing human TIM-3 or Bw-control cells with biotin-labelled recombinant human galectin-9 at different concentrations . When used at higher concentrations galectin-9 was found to interact with parental Bw cells . Galectins are known to bind to β-galactose containing glyco-conjugates and thus the interaction with β-galactose containing glycoproteins expressed on the surface of Bw cells is likely to account for this binding . Importantly , this interaction was independent of TIM-3 , since the binding signals were not enhanced on cells expressing high levels of human TIM-3 on their surface ( Figure 2B ) . In addition , we performed binding studies where galectin-9 antibodies were used for detection . Also in these experiments we did not observe increased binding to the TIM-3 expressing cells ( Figure S2B ) . Finally , we performed an ELISA , where immobilized recombinant human galectin-9 was probed with an immunoglobulin fusion protein representing the extra cellular domain of human TIM-3 ( TIM-3-Ig ) . For control purposes , the binding of ILT5-Ig , mouse-CTLA4-Ig , TROP2-Ig , TREML2-Ig from the same manufacturer and human IgG was analyzed in parallel . Also in these experiments we did not observe a specific interaction between human TIM-3 and galectin-9 , since compared to the control fusion proteins , the binding signals observed with TIM-3-Ig were not higher ( Figure 2C ) . Furthermore , when we performed ELISA binding experiments that cover a broad concentration range ( 0 . 3–9 µg/ml ) we did not obtain higher binding signals with TIM-3-Ig compared to murine CTLA4-Ig that was used as a control ( Figure S2C ) . In these experiments , we also included TIM-3-Ig and a control fusion protein ( EpCam-Ig ) produced in our laboratory . Compared to the commercial fusion proteins ( all expressed in NSO-cells ) , the ELISA signal obtained with our fusion proteins was generally much weaker , which could indicate that the NSO-expressed fusion proteins harbour modifications that account for the galectin-9 binding that was observed with all commercial preparations tested . Importantly , with both types of TIM-3-Ig preparations we did not observe higher binding signals compared to the respective control proteins ( Figure S2C ) . Since receptor-ligand interactions are usually conserved between mice and men , we also performed a series of experiments to analyze the interaction between murine TIM-3 and galectin-9 . However , as expected from the experiments with their human orthologues and in contrast to a previous report [14] , a specific interaction between cell expressed murine galectin-9 and murine TIM-3-Ig fusion proteins was not detected ( Figure 2D ) . Likewise , membrane-resident murine TIM-3 did not bind soluble recombinant murine galectin-9 , and ELISA experiments did also not yield evidence for a specific interaction between murine TIM-3 and galectin-9 ( Figure 2E , F ) . Binding studies in the presence of calcium ions were also performed with the same outcome ( data not shown ) . In a next set of experiments , we assessed the function of TIM-3 in human T cell activation using an antibody ( clone 2E2 ) that was previously described and commonly referred to as blocking TIM-3 antibody [10]–[12] , [19]–[23] . CD4 T cells were stimulated with plate-bound antibodies to CD3 and CD28 , and proliferation and cytokine secretion were measured following 48 hours of stimulation . Whereas Hastings et al . reported that TIM-3 antibody 2E2 enhanced cytokine production , but not T cell proliferation under these experimental conditions [20] , we did not observe any functional effects of clone 2E2 as neither proliferation nor cytokine production was affected by the addition of this TIM-3 antibody ( Figure 3 ) . Since CD4 T cells express little TIM-3 during primary stimulation we also performed experiments where the effects of TIM-3 antibody 2E2 were assessed on T cells harbouring TIM-3 at a high density . PBMC were stimulated with antibodies to CD3 and CD28 under Th1 polarizing conditions . In line with previous reports this treatment resulted in T cells with high surface TIM-3 expression ( Figure 4A; inset ) . However , presence of TIM-3 antibody 2E2 during the stimulation of such T cells with allogeneic DC again did not result in an increase in proliferation or cytokine production . By contrast , we found that a blocking PD-L1 antibody strongly enhanced proliferation and cytokine production in these experiments ( Figure 4A , B ) . TIM-3 has recently been described as a marker of human T cell exhaustion and it was reported to be expressed at high levels on CD8 T cells of HIV-1-infected patients . Moreover , TIM-3 antibodies were reported to revert the dysfunctional state of T cells , which suggests that TIM-3 has a function in the maintenance of T cell exhaustion . We thus assessed TIM-3 expression on CD8 T cells of viremic HIV-1-infected patients and of patients under antiretroviral therapy with no detectable viral load in peripheral blood ( virologically suppressed ) . For comparison , the expression of PD-1 , a well-established marker for human T cell exhaustion , was also analyzed . We found TIM-3 to be expressed on only a very small proportion of CD8 T cells in these individuals , whereas PD-1 was strongly up-regulated ( Figure 5A ) . Importantly , the weak staining signal obtained with the TIM-3 antibodies was not due to a weaker reactivity of the antibody used , since TIM-3 and PD-1 antibodies reacted both very strongly with transductants expressing their cognate antigens ( Figure S3 ) . Although the percentage of TIM-3 positive CD8 T cells was slightly higher in HIV-1-infected donors , the difference was not statistically significant , whereas PD-1 was significantly increased in HIV infected patients ( Figure 5B ) . TIM-3 expression did not correlate with disease progression parameters ( CD4 cell count and viral load; data not shown ) . Furthermore , we observed that TIM-3 positive CD8 T cells predominantly belong to the CD45RA+ T cell subset , which is in contrast to PD-1 expressing T cells that were mainly CD45RA− ( CD45RO+; Figure 5C ) . In respect of TIM-3/CD45RA co-expressing T cells , significantly higher numbers could be found in viremic HIV-1-infected patients than in healthy controls , indicating that there is an enrichment of these cells in patients with uncontrolled chronic HIV-1 infection ( Figure S4A ) . When co-staining with TIM-3 and PD-1 antibodies we also observed little co-expression of these molecules on CD8 cells of viremic HIV patients ( Figure S4B ) . CD45RA+T cells are found on both ends of the life cycle of T cells and characterize either naïve T cells or terminally differentiated T cells . Since TIM-3 expression has been linked to chronic viral infection and persistent antigen stimulation shifts the T cell repertoire from naïve to terminally differentiated T cells it is likely that TIM-3/CD45RA+ T cells belong to the latter . Thus , we analyzed expression of CD57 , a marker for senescent T cells , on TIM-3 positive CD8 T cells . We found that the majority of TIM-3 expressing CD8 T cells indeed co-express CD57 and these cells thus belong to late-differentiated T cells . In line with the differences in CD45RA expression , the percentage of PD-1 expressing cells was significantly lower in CD57-expressing CD8 T cells than in CD57-negative CD8 T cells ( Figure 5D ) . Taken together , these data indicate that compared to PD-1 expressing T cells , the number of TIM-3 expressing T cells is much lower in chronically HIV-1-infected patients . Moreover , we show that TIM-3 and PD-1 characterize different CD8 T cells in these individuals . Several studies have reported that similar to PD-1 , TIM-3 also has a functional role in T cell exhaustion associated with HIV-1-infection , but also with other chronic viral infections or cancer [10]–[13] , [22] . Most of these studies have used the TIM-3 antibody 2E2 to investigate the functional role of TIM-3 during stimulation of CD8 cells with virus or cancer antigens . We used Gag- and Nef-specific T cells as a model system to compare TIM-3 antibody 2E2 and blocking PD-L antibodies regarding their ability to enhance HIV-1-specific T cell responses in vitro . In these experiments we found that PD-L blockade but not TIM-3 antibodies consistently enhanced proliferation measured by methyl-3[H]-thymidine uptake as well as IFN-γ production in HIV-1-specific T cells ( Figure 6A , B; Table 1 ) . Similar results were obtained in CFSE-dilution experiments: PD-1 blockade strongly increased the numbers of CD8 T cells that had proliferated in response to HIV-1 peptides , whereas TIM-3 antibody 2E2 had no effect ( Figure 6C; Table 1 ) .
TIM-3 was identified during an attempt to find new Th1-specific cell surface proteins by screening a large panel of monoclonal antibodies raised against murine Th1 T cell clones and lines . Antibodies that stained Th1 cells , but did not react with Th2 cells were found to interact with TIM-3 , which was also highly expressed on cytotoxic CD8 cells [24] . The authors reported that TIM-3 antibodies enhanced experimental autoimmune encephalomyelitis ( EAE ) and suggested that TIM-3 has a role in autoimmune diseases by regulating macrophage number and activation . However , subsequent studies implicated TIM-3 in tolerance induction by interfering with Th1 proliferation and maintaining Treg function [25] , [26] . Interestingly , TIM-3 expression was also reported for innate immune cells , where it was found to promote rather than inhibit immune responses [21] . Taken together , these data attribute a very complex functional role to this molecule , which depending on the site of expression may act to promote or inhibit murine immune responses upon interaction with its ligands . Zhu et al . described galectin-9 as a binding partner for murine TIM-3 and reported that galectin-9/TIM-3 interaction inhibits Th1-responses in mice [14] . Much less is known regarding the functional role of TIM-3 on human immune cells . Importantly , there are no reports how galectin-9 impacts on the activation of human T cells . Therefore , we have used our T cell stimulators , an experimental system that we have devised for the functional evaluation of human costimulatory or coinhibitory ligands , to address a potential functional role of human galectin-9 on human T cells . We found that presence of galectin-9 during the activation of human T cells does not inhibit T cell responses or induce cell death . Since there are no reports that corroborate a specific interaction for human galectin-9 and human TIM-3 , we have performed a series of binding studies to analyze a potential interaction between human galectin-9 and TIM-3 . Our data did not yield any evidence for such an interaction . In most cases receptor-ligand interactions are conserved between mice and men , and we thus went on to study a potential interaction of murine TIM-3 with murine galectin-9 . Using several different experimental settings , we did not detect a specific interaction between these molecules . This is in apparent contrast to a study by Zhu and coworkers [14] . It should however be stressed that in their study the interaction of surface galectin-9 with TIM-3 has not been analyzed . Rather , TIM-3 fusion proteins were used for binding studies with fixed and permeabilized CHO cells transfected with galectin-9 expression constructs . This unusual type of binding experiments was justified by claiming that according to a previous publication by Rabinovich et al . “transient transfection does not produce expression of galectins on the cell surface” [27] . However , this study focuses on galectin-1 and does not contain any statements or data regarding the cell surface expression of galectins in general . It is evident that a specific interaction of surface or soluble galectin-9 with TIM-3 is a pre-requisite for an involvement of galectin-9 in TIM-3 mediated effects . The results of our study largely rule out such an interaction , suggesting that TIM-3 functions are independent of galectin-9 . Su et al . have shown that functional effects of galectin-9 do not differ in wild-type and TIM-3 knockout mice , which would also be consistent with our findings and suggests that galectin-9 functions are independent of TIM-3 [28] . Furthermore although TIM-3 has been implicated in Treg function , Treg suppression was not impaired in galectin-9 deficient mice indicating that TIM-3 would exert such function independent of galectin-9 [29] . There are several reports suggesting that TIM-3 has a functional role in T cell exhaustion . We have found that compared to CD8 T cells that harbour the classical exhaustion marker PD-1 , the number of TIM-3 expressing T cells is much lower in viremic HIV-1-infected patients as well as in individuals where the virus is suppressed by antiretroviral therapy . Furthermore , exhausted T cells have been described as being CD45RO+ [30] and in line with this , the classical exhaustion marker PD-1 was preferentially expressed on CD45RA−CD8 T cells , which are CD45RO positive . By contrast , most TIM-3+ CD8 cells were found to belong to the CD45RA+ subset . The analysis of CD57 expression also suggested differences between PD-1 and TIM-3 expressing T cells: TIM-3 was co-expressed with CD57 , whereas the percentage of PD-1 expressing cells was higher in the CD57− subset . Thus , the majority of TIM-3 expressing cells express CD45RA and CD57 , indicating that TIM-3 defines a terminally differentiated cell population that is distinct from classical PD-1 expressing exhausted T cells . Jones et al . have also reported that in HIV-1-infected individuals the TIM-3 expressing T cell population is distinct from the PD-1 expressing T cells [10] . They found that the TIM-3 expressing cells are impaired in their ability to produce cytokines , which is in line with a terminally differentiated phenotype of these cells observed in our study . Differences in the amount of TIM-3 expressing T cells in these two studies might result from different staining techniques but also due to the fact that the cohort studied by Jones et al . included a high number of acute/early HIV infections where they observe the highest levels of TIM-3 expressing T cells . In contrast , the cohort studied here comprises chronically infected patients only . Previous studies have in most cases used the TIM-3 antibody 2E2 - which has been suggested to function as a TIM-3 antagonist – to assess the role of human TIM-3 during T cell activation and exhaustion . We have used Gag and Nef-specific T cells derived from HIV-1-infected individuals as a model system to compare TIM-3 antibody 2E2 and blocking PD-L antibodies regarding their capacity to enhance HIV-1-specific T-cell responses in vitro . In these experiments we found that whereas PD-L blockade significantly enhanced proliferation and IFN-γ production in HIV-1-specific T cells , TIM-3 antibody 2E2 was completely ineffective in this respect ( Figure 6; Table 1 ) . In contrast to a previous study we found that TIM-3 antibody 2E2 was also ineffective in enhancing cytokine production of anti-CD3/CD28 stimulated human CD4 cells [20] . Importantly , we also investigated the effect of TIM-3 antibodies during the stimulation of human Th1 polarized human CD4 T cells with allogeneic monocyte derived DC . Despite the high expression of TIM-3 on these T cells TIM-3 antibodies have also been ineffective in these experiments . Taken together our study does not yield any evidence for a role of TIM-3 during human T cell activation processes . Previous reports suggest that TIM-3 exerts its functional role upon interacting with ligands , since TIM-3 antibodies and TIM-3 fusion proteins had similar effects . We have repeatedly screened high quality expression libraries generated from resting and in vitro activated human PBMC or human DC [31] , [32] with TIM-3 fusion proteins to identify TIM3-ligands . These attempts did not yield TIM-3 binding clones ( Figure S5 ) . Although it cannot be ruled out completely that TIM-3 interacts with molecules that were not represented in the cell pools used for screening , it might also indicate that human PBMC and DC do not express TIM-3 ligands . TIM-1 and TIM-4 bind phosphatidylserine ( PtdSer ) , which is exposed on the surface of apoptotic cells via a conserved binding pocket termed metal ion-dependent ligand binding site ( MILIBS ) localized on the N-terminal end of their IgV domain [33] . Importantly , human as well as murine TIM-3 molecules also harbour such a motif and DeKruyff et al . have demonstrated that immunoglobulin fusion proteins representing TIM-3 bind to PtdSer in liposomes in a calcium-dependent manner . Furthermore , they found that cells transfected with human TIM-3 bind and phagocytose apoptotic cells [34] . In line with these results it was described that TIM-3-mediated phagocytosis of apoptotic cells is crucially involved in cross-presentation and was linked to peripheral tolerance [35] . Thus it is quite possible that a major functional role of TIM-3 lies in the interaction with apoptotic cells . Engulfment of apoptotic cells by phagocytes can result in potent anti-inflammatory effects and prevent autoimmunity [36]–[38] . TIM-3 is expressed on APC including microglial cells of the CNS [21] . Thus , as already pointed out by DeKruyff et al . , enhanced experimental autoimmune encephalomyelitis or reduced transplant tolerance that was observed in mice upon blockade or absence of TIM-3 could be explained by impaired phagocytic activity [34] . Identification of TIM-3 ligands expressed on intact cells would greatly facilitate investigations on potential additional functions for this molecule in immunity .
The ethics review board of the General Hospital and the Medical University of Vienna approved collection of blood samples from HIV-1-infected patients and from healthy donors . Written informed consent was obtained from the HIV-1-infected study participants and from the healthy donors prior to blood sampling . All samples were anonymized and research confirmed to the guidelines of the ethics review board of the General Hospital and Medical University of Vienna . RPMI-1640 supplemented with 10% FBS , antibiotics and amphotericin was used to culture the mouse thymoma cell line Bw5147 ( short designation within this work: Bw cells ) , T cell stimulator cells and for functional assays with cells derived from healthy donors . Functional assays with cells from HIV-1-infected individuals were done in AIM-V medium ( Invitrogen , Carlsbad , CA ) in presence of costimulatory antibodies to CD28 ( #28 . 2 , Biolegend , San Diego , CA ) and CD49d ( # L25; BD-Pharmingen , San Diego , CA; final concentration 0 . 1 µg/ml each ) . The coding sequences of human and murine galectin-9 and human and mouse TIM-3 were PCR-amplified and cloned into the retroviral expression vector pCJK2 [15] . Sequence analysis was performed and constructs that contained sequences that were identical to the coding sequences of Genbank entries: NM_032782 ( human TIM-3 ) , BC_106851 ( murine TIM-3 ) , NM_001159301 ( human galectin-9 ) and NM_002308 ( murine galectin-9 ) were selected for further use . Bw-transductants expressing these molecules and T cell stimulator cells expressing human galectin-9 were generated using previously described retroviral transduction protocols [15] . Expression of these molecules was confirmed using goat-anti-human galectin-9 , goat-anti-mouse galectin-9 ( both from R&D Systems , Minneapolis , MN ) , mAbs to human ( #F38-2E2 , short designation 2E2 ) or mouse TIM-3 ( #B8 . 2C12; both from Biolegend ) . Bound antibodies were detected using appropriate PE-conjugated secondary reagents from JacksonImmunoResearch ( West Grove , PA ) . Mouse and human TIM-3 immunoglobulin ( Ig ) fusion proteins , human ILT5 Ig , human TROP2 Ig , human TREML2 Ig , mouse CTLA4 Ig fusion proteins - all expressed in NSO-cells – and recombinant human ( rh ) and recombinant mouse ( rm ) galectin-9 were purchased from R&D systems . Human IgG preparations were bought from JacksonImmunoResearch . In additional experiments immunoglobulin fusion proteins representing the extra-cellular domains of human TIM-3 or human EpCam or B7-H3 produced in our laboratory were used . These fusion proteins were generated and expressed in 293T cells using previously described methods [39] . In flow cytometry experiments the binding of Ig-fusion proteins was detected using goat-anti-human-IgG-Fc-PE antibodies ( Jackson ImmunoResearch ) . Rh-galectin-9 and rm-galectin-9 were labelled using Biotin-X-NHS ( Sigma-Aldrich Chemie GmbH , Taufkirchen , Germany ) . Binding of biotinylated galectin-9 was detected using Strepdavidin-PE ( BD Pharmingen ) . Flow cytometric analysis was done using a FACSCalibur flow cytometer supported by CELLQUEST software ( Becton Dickinson ) . Fluorescence intensity is shown on a standard logarithmic scale . For ELISA-based binding studies , recombinant galectin-9 was reconstituted in PBS and immobilized overnight at room temperature on high-protein binding ELISA plates ( Maxisorp; NUNC/Thermofisher Fremont , CA; coating concentration 0 . 5 µg/ml and 1 µg/ml for human and murine galectin-9 , respectively ) . The plate was washed and blocked with PBS-0 . 5% BSA ( 1 h at 37°C ) . Immunoglobulin fusion proteins were added at the indicated concentrations in PBS-0 . 5% BSA and incubated for 2 hours at 37°C . The plate was washed and HRP-labelled goat-anti-human IgG antibodies ( Fc-specific; JacksonImmunoResearch , diluted 1∶800 ) were added and incubated for 2 hours at 37°C . Plates were washed and developed using ABTS solution ( Roche Applied Science , Mannheim , Germany ) . Following 15 min incubation , the O . D . 405 nm was determined using 650 nm as reference wave length . T cell activation experiments using the system of T cell stimulator cells were done as previously described [15] . Briefly , irradiated ( 6000 rad ) T cell stimulator cells ( 2×104/well ) were co-cultured with 1×105 T cells , and following 48 h of stimulation , cell culture supernatant was harvested for cytokine measurement and methyl-3[H]-thymidine was added to the cultures . Expression of membrane-bound anti-CD3 antibody fragments on the T cell stimulator cells was detected via DyLight-649-conjugated goat antibodies to mouse IgG-H+L specific ( Jackson ImmunoResearch ) . Isolation of human T cells and monocytes and generation of immature and mature monocyte-derived DC was done as previously described [40] . For T cell proliferation assays with plate-bound antibodies , antibodies to CD3 ( #OKT3 , ADG , Kaumberg , Austria ) and CD28 ( #28 . 2 , Biolegend ) were immobilized for 4–5 h 37°C at the indicated final concentrations . Following two washing steps , 1×105 T cells/well were added and cultured for 48 h in the presence of soluble TIM-3 antibodies or isotype control antibodies ( both Biolegend , final concentration: 5 µg/ml ) as previously described [20] . For Th1 polarisation MNCs were activated with antiCD3/antiCD28 coated beads ( Dynabeads , Invitrogen ) and IL-2 ( 300 U/ml; PeproTech Inc . Rocky Hill , NJ ) for two days . Subsequently , cells were washed and IL-12 ( 25 ng/ml; PeproTech Inc ) and IL-2 ( 100 U/ml ) were added for 3 days . For DC stimulation assays human T cells ( 1×105/well ) were co-cultured with allogeneic immature or mature human DC at the indicated cell numbers for 5 days in presence of the indicated antibodies ( final concentration 10 µg/ml ) . To assess T cell proliferation methyl-3[H]-thymidine ( final concentration: 0 . 025 mCi; Perkin Elmer/New England Nuclear Corporation , Wellesley , MA ) was added for the last 18 hours of culture . Methyl-3[H]-thymidine uptake was measured as described [41] . Blood samples from patients infected with HIV-1 were obtained during routine check-ups at the outpatient clinic of the HIV-unit of the Department of Dermatology , Division of Immunology , Allergy and Infectious Diseases , Medical University of Vienna , Austria . Patients were referred to as either virologically suppressed when on ART ( antiretroviral therapy ) with a virus load below the limit of quantification ( 20 copies/ml ) for at least two consecutive controls or at least 3 months or viremic ( ART naïve; viral load above 1000 copies/ml ) . Mean duration of ART was 61 ( ±52 ) months . Patients with ongoing opportunistic infections or tumours or any other obvious acute medical conditions were excluded from the analysis . For analysis of TIM-3 and PD-1 expression on T cells of HIV-1-infected and healthy donors , whole blood samples were stained with TIM-3-PE , PD-1-PE or a PE-labelled isotype control ( all from Biolegend ) . For TIM-3/PD-1 costaining , a PD-1-APC from BD Pharmingen was used . Cells were counterstained with CD8-APC , CD4-PerCP , and CD45RA-FITC or CD57-FITC ( all from BD Pharmingen ) . Subsequently , cells were washed and treated with ADG lysis ( Kaumberg , Austria ) . PBMC of HIV-1-infected individuals were isolated using density gradient centrifugation and stimulated for 7 days in triplicates in flat-bottom 96-well plates with overlapping 15-mer HIV-1 Clade B Gag and Nef peptide pools ( generously provided by the National Institutes of Health AIDS Research and Reference Reagent Program; final concentration 20 µg/ml ) . All stimulation experiments were performed in absence of antibodies or in presence of isotype control mAb ( #MOPC-21 ) , a combination of PD-L1 mAb ( #29E . 2A3 ) and PD-L2 mAb ( # 24F . 10C12 ) , or TIM-3 mAb ( #F38-2E2 ) . mAbs ( purchased from Biolegend in functional grade quality ) were used at a final concentration of 10 µg/ml . Following 6 days of stimulation , culture supernatants were harvested for INF-γ measurement and subsequently methyl-3[H]-thymidine ( final concentration: 0 . 025 mCi ) was added and methyl-3[H]-thymidine incorporation was determined as described above . In an additional set of experiments T cell proliferation was assessed by CFSE staining . CFSE labelling was performed as described previously [42] and stimulation of CFSE-labelled PBMC was done for 7 days as described above with the exception that 24 well plates were used . Cells were harvested , stained for CD8 expression and analyzed by FACS as described . In both types of experiments mock stimulation cultures without HIV-1 peptides were set up in parallel for each donor to assess specific response to Gag/Nef peptides ( at least 2-fold greater than mock treated samples ) . Supernatants from T cell activation experiments were harvested prior to addition of methyl-3[H]-thymidine , and pooled from triplicate wells and used for measurement of IL-2 , IL-10 , IL-13 , IL-17 and IFN-γ . Supernatants from stimulation cultures of T cells derived from HIV-1-infected individuals were harvested on day 6 and used for measurement of IFN-γ . All measurements were performed in duplicates using the Luminex System 100 ( Luminex , Texas , USA ) . Two-tailed Student's-t test was used to assess significance . Error bars indicate the SD . | Inhibitory costimulatory receptors are a hallmark of exhausted T cells , which accumulate during chronic infection with viruses like HIV-1 . Recently , TIM-3 was described as functional receptor on exhausted human T cells . Galectin-9 was reported as an inhibitory ligand for TIM-3 on murine T cells , but it was not known whether galectin-9 has a role in human T cell activation processes . We have found that the activation of human T cells is not affected by the presence of galectin-9 or antibodies to TIM-3 . Furthermore , we demonstrate that galectin-9 does not serve as a ligand of human or murine TIM-3 . Analysis of T cells of HIV-infected individuals regarding the expression of TIM-3 and PD-1 demonstrates that TIM-3 is expressed on fewer cells and also at much lower levels . In fact , TIM-3 expression characterizes a T cell population that is distinct from the PD-1 expressing exhausted T cells . Our results indicate that PD-1 , but not TIM-3 , is a promising target to ameliorate T cell exhaustion . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"infectious",
"diseases",
"hiv",
"immunology",
"biology",
"viral",
"diseases",
"immunomodulation",
"immune",
"response"
]
| 2013 | TIM-3 Does Not Act as a Receptor for Galectin-9 |
Trypanosoma vivax is one of the most common parasites responsible for animal trypanosomosis , and although this disease is widespread in Africa and Latin America , very few studies have been conducted on the parasite's biology . This is in part due to the fact that no reproducible experimental methods had been developed to maintain the different evolutive forms of this trypanosome under laboratory conditions . Appropriate protocols were developed in the 1990s for the axenic maintenance of three major animal Trypanosoma species: T . b . brucei , T . congolense and T . vivax . These pioneer studies rapidly led to the successful genetic manipulation of T . b . brucei and T . congolense . Advances were made in the understanding of these parasites' biology and virulence , and new drug targets were identified . By contrast , challenging in vitro conditions have been developed for T . vivax in the past , and this per se has contributed to defer both its genetic manipulation and subsequent gene function studies . Here we report on the optimization of non-infective T . vivax epimastigote axenic cultures and on the process of parasite in vitro differentiation into metacyclic infective forms . We have also constructed the first T . vivax specific expression vector that drives constitutive expression of the luciferase reporter gene . This vector was then used to establish and optimize epimastigote transfection . We then developed highly reproducible conditions that can be used to obtain and select stably transfected mutants that continue metacyclogenesis and are infectious in immunocompetent rodents .
Trypanosoma vivax and Trypanosoma congolense are the main parasite species responsible for Animal African Trypanosomosis ( AAT ) or Nagana . This disease causes about 3 million deaths annually and has a marked impact on agriculture in sub-Saharan and South American endemic countries , leading to annual livestock production losses of about 1 . 2 billion US dollars [1]–[3] . T . vivax accounts for up to half of total AAT prevalence in West Africa where it is considered a predominant pathogen for domestic animals [2] , [3] . The main symptoms in cattle correspond to weight loss , high abortion rates , decreased milk production , and reduced draught power and endurance [2] , [3] . T . vivax presents a short and simple life cycle in contrast to T . brucei [4] and to a lesser extend to T . congolense . In tsetse flies , T . vivax development takes place in the proboscis where bloodstream forms ( BSF ) evolve to epimastigotes , a non infective , replicative form . After a multiplication phase , these epimastigotes undergo metacyclogenesis and transform into metacyclic infective forms , and here it is noteworthy that Glossina spp . are the only vectors in which T . vivax is able to multiply and pursue its differentiation into metacyclic forms . West African T . vivax populations have been introduced into South American countries - devoid of the tsetse fly - where they are now a real threat since they can be efficiently transmitted across vertebrate hosts by other hematophagous insects , including tabanids . In this case the parasites are transmitted mechanically between vertebrate hosts in a noncyclical manner , i . e . with no growth or multiplication in the insects [5] , [6] . This simpler lifecycle enables T . vivax to adapt to different vectors and hosts and may explain why it has emerged so rapidly in South America . Despite the fact that T . vivax has a major impact on emerging economies , limited efforts have gone into its study during the last decade . For our part , we have recently developed in vivo laboratory models of T . vivax infection , we initiated a detailed assessement of its infectious processes and characterized some of the key players in the immunopathology of experimental trypanosomosis [7] , [8] . Our work showed that sustained and reproducible infections can easily be obtained using C57BL/6 , BALB/c and Outbred mice that reproduce the parasitological , histological and pathological parameters of the livestock infection found in the field . These experimental in vivo models are useful in work conducted to explore the immunobiology of T . vivax infection and are essential in efforts made to elucidate , for instance , the function of some virulence factors in vivo [9] , [10] . Over the last decade , recombinant gene technology has expanded our ability to investigate gene expression and function in trypanosomatids . However , transgenesis and the selection of recombinant mutants depend on our ability to maintain and grow trypanosomes in axenic cultures . The growth of insect forms of T . vivax in vitro was firstly described by Trager in 1959 and in the mid 1970s , in the presence of tsetse tissues [10] , but the cultures were not stable and parasites did not survive for more than 18 days . Later , Isoun and Isoun took T . vivax BSF from infected cattle and managed to transform these into epimastigote forms without using insect or mammalian tissues . Unfortunately , dividing parasites were unable to withstand subculturing [11] . New methods initially dependent on feeder layer cells and subsequently adapted for the axenic cultivation of epimastigote and metacyclic forms of T . vivax were later proposed by several groups in the eighties and the nineties [12]–[18] . But presently , a general consensus among T . vivax researchers involves the difficulties to maintain the parasite in culture using the principles described in these pioneer reports [19] . This raises some concerns about the composition of the culture media described . Furthermore , this lack of a robust and efficient method for maintaining the parasite in vitro may readily explain the total absence of any genetic tools for engineering T . vivax , and this in turn has made it difficult to analyze parasite gene expression and function . We describe herein the successful development and standardization of in vitro axenic cultures of epimastigote forms of T . vivax obtained from BSF of the IL 1392 parasite strain stably kept in vivo [8] . This West African stock of T . vivax is derived from the Nigerian isolate Zaria Y486 which is infective for rodents and can be cyclically and/or mechanically transmitted [20] , [21] . Cultured epimastigote forms continue their differentiation in vitro into metacyclic parasites and thus acquire infectious properties in mice . In addition , we describe the first integrative expression vector for T . vivax , designed to constitutively express foreign gene products and bearing the neomycin phosphotransferase ( NeoR ) selectable marker which confers resistance to G418 . This expression system also harbors a long ribosomal promoter region of T . vivax to drive transcription of the reporter and NeoR genes and thus improve gene expression and permit recombinant selection . We used this vector to establish conditions conducive to the efficient and highly reproducible transfection and selection of T . vivax epimastigote mutants . We show here that the pTvLrDNA-Luc plasmid is appropriately integrated and that the product of the reporter gene is expressed at detectable levels . Finally , the culture protocols described herein were used successfully for the in vitro selection , growth and development of all the evolutive forms of genetically engineered T . vivax that are infectious to immunocompetent mice .
All mice were housed in our animal care facilities in compliance with European animal welfare regulations . Institut Pasteur is a member of Committee #1 of the Comité Régional d'Ethique pour l'Expérimentation Animale ( CREEA ) , Ile de France . Animal housing conditions and the protocols used in the work described herein were approved by the “Direction des Transports et de la Protection du Public , Sous-Direction de la Protection Sanitaire et de l'Environnement , Police Sanitaire des Animaux” under number B 75-15-28 , in accordance with the Ethics Charter of animal experimentation that includes appropriate procedures to minimize pain and animal suffering . PM is authorized to perform experiments on vertebrate animals ( license #75–846 issued by the Paris Department of Veterinary Services , DDSV ) and is responsible for all the experiments conducted personally or under her supervision as governed by the laws and regulations relating to the protection of animals . Trypanosoma ( Dutonella ) vivax IL 1392 was originally derived from the Zaria Y486 Nigerian isolate [8] , [22] . These parasites had recently been characterized and were maintained in the laboratory by continuous passage in mice , as previously described [8] . Seven to 10-week-old male Swiss Outbred ( CD-1 , RJOrl:SWISS ) or BALB/c mice ( Janvier , France ) were used in all experiments . Mice were injected intraperitoneally with bloodstream forms of T . vivax ( 103 parasites/mice ) or with cells derived from axenic cultures ( 2×106 metacyclic-like trypomastigotes ) . Parasitemia was determined as previously described [9] . All animal work was conducted in accordance with relevant national and international guidelines ( see above ) . Epimastigote cultures were initiated with the blood of infected mice once parasitemia reached at least 5 . 108 parasites/ml . Blood was collected by cardiac puncture onto heparin ( 2500 IU/kg ) , and was then diluted 1 ∶ 8 ( v/v ) with PBS −0 . 5% glucose to 5 . 107 parasites/ml . Parasites were separated from red blood cells by differential centrifugation using a swingout rotor ( Jouan GR412 , Fisher Bioblock Scientific , Strasbourg , France ) . This technique offered a higher index of recovery of viable BSF ( 4 . 0–4 . 5×108 BSF , corresponding to 80–90% recovery ) than classic ion-exchanged chromatography using DEAE-cellulose based methods . Briefly , diluted blood was processed by one first round of centrifugation ( 5 minutes at 200 g ) and the supernatant withdrawn with a Pasteur pipette without disturbing the red blood cell layer and the thin interface containing the white blood cells . Parasite-enriched suspension was submitted to a second round of centrifugation ( 10 minutes at 300 g ) . Supernatant was then centrifuged 10 minutes at 1800 g and BSF - containing pellets devoid of host cells used to inoculate culture flasks containing different culture medium to a final concentration of 106 to 107 parasites per ml . These were then incubated at 27°C in an atmosphere devoid of CO2 ( see Table 1 for details ) . Parasite adhesion was checked by visual inspection after 4 to 5 days when half the media had to be changed . Cultures were maintained in 25 cm2 polystyrene flasks ( T25 ) ( Corning , Bagneaux-sur-Loin ) by changing 3 ml of medium every 2 or 3 days . The TV1–5 media used in this study were based on D-MEM ( Dulbecco's Modified Eagle's Medium , Invitrogen ) or IMDM ( Iscove's Modified Dulbecco's Medium , Invitrogen ) . These media were supplemented with 0–0 . 4% glucose , 0–20% heat-inactivated fetal calf serum ( FBS , MP Biomedicals or Invitrogen ) and/or 0–20% heat-inactivated goat serum ( GS , Invitrogen ) , 0 . 03 mM bathocuproinedisulfonic acid , 0 . 45 mM L-cysteine , 0 . 2 mM hypoxanthine , 0 . 14 mM ß-mercaptoethanol , 0 . 4–6 mM L-proline , 0 . 05 mM thymidine , and 25 mM HEPES pH7 . 4 , as indicated in Table 1 . All supplements were obtained from Sigma Aldrich except HEPES ( Invitrogen , Cergy Pontoise ) . Conditioned medium consisted of 1 volume of centrifuged ( 10 minutes at 1800 g ) and filtered supernatant from 2- to 3-week-old cultures , diluted with 2 volumes of fresh medium . Parasites from the supernatant or from the adherent layer were collected , washed in PBS and resuspended at 5 . 107/mL . 40 µL of the various suspensions were spotted onto coverlips and allowed to settle for 10 min before being fixed for 15 s in cold methanol . Slides were incubated for 1 h at 37°C with mouse monoclonal antibodies directed against paraflagellar rod protein 2 ( anti-PFR2 L8C4 ) [23] . They were then washed 5 times with PBS and incubated for 45 min with a goat anti-mouse IgG secondary antibody labeled with Alexa Fluor 488 ( MolecularProbes , France ) . DNA was stained with 3 µg/mL 4′ , 6′-diaminido-2-phenylindole ( DAPI , Sigma-Aldrich ) for 10 min at room temperature and the slides were washed 5 times and finally mounted in Fluoromount G ( Interchim , Montluçon ) . Parasite forms were examined under an Olympus immunofluorescence multifilters BH-2 UV ( Zeiss ) or DMR ( Leica ) microscope . Images were captured , for instance using a CoolSnap HQ camera ( Roper Scientifique ) . Several steps were required to construct the first T . vivax specific vector ( see Table 2 for primer sequences ) . Initially , TvPRAC 5′UTR sequence containing the Spliced leader Acceptor Site ( p . -582 to p . -1 ) was amplified from BSF T . vivax genomic DNA using SLasF 5′ and SLasRmcs multiple cloning site primers . The amplified product ( 617 bp ) was subcloned into pCR Blunt topo vector ( Invitrogen ) ; this construct was submitted to nested PCR using SLasKpnI-F and McsSacI-R primers to introduce specific KpnI and SacI sites . A 616 bp fragment was then obtained after KpnI and SacI digestions and appropriately inserted into pBlueScript KS to create pTv5′UTRa . The T . vivax intergenic region between α and β tubulin ( Tvtubαβ ) was amplified from BSF genomic DNA using TvTubαβBam-F and TvTubαβAsc-R primers . The fragment obtained ( 506 bp ) was digested with BamHI and AscI and inserted into BamHI and AscI sites of the pTv5′UTRa vector . The neomycin resistance gene cassette ( NeoR ) was excised from pXS2-GFP [24] , by digestion with AscI/PacI and the 802 bp Neo-fragment further inserted downstream of TvTubαβ to produce vector pTv5′UTRb . In order to provide a putative 3′ polyadenylation signal for the NeoR gene , the 330 bp intergenic region located between TvTub β and TvTub α was amplified by PCR using TvTubβαF and TvTubβαSac-R and the resulting fragment was digested with PacI/SacI and subcloned into the PacI/SacI sites of the pTv5′UTRb digested vector . Firefly luciferase reporter gene was purified from Trypanosoma brucei pLEW100 vector [25] , digested with HindIII and BamHI and cloned among the TvPRAC 5′UTR and TvTubαβ sequences of pTv5′UTRb to produce pTv-LUC . Finally , 2 derivatives of pTv-LUC were constructed containing a long ( 1 . 8 kb , LrDNA ) or a short ( 1 . 2 kb , SrDNA ) upstream of the 18S rDNA sequence . These putative RNA PolI promotor regions were amplified with TvrDNAK-F and TvrDNAK-R primers and further inserted into the KpnI site of pTv-Luc to produce the final expression vectors pTvLrDNA-Luc and pTvSrDNA-Luc . All steps in these constructions were validated by sequencing to check that the different fragments were in the correct location and orientation . The GFP cassette was excized from the vector pXS2-GFP [24] by HindIII and EcoRI digestion . The resulting gene fragment ( 714 bp ) was used to replace the firefly luciferase reporter gene of the pTvLrDNA-Luc vector between the TvPRAC 5′UTR sequence ( HindIII site ) and the intergenic region Tvtubαβ ( EcoRI site ) . As here above , appropriate replacement was validated by sequencing . Parasites were recovered from the flasks after 15 to 20 days of culture . In order to recover adherent parasites without causing physical damage , the flasks were washed twice with PBS −0 . 5% glucose and adherent cells were detached from the surface of the plastic using a cell scraper in the presence of PBS . For transfection using the Gene Pulser system ( Biorad , Marnes-la-Coquette ) , parasites were washed and resuspended in Cytomix at 0 . 5–1 . 5×108 cells/ml then 500 µl suspensions were mixed with 5–20 µg vector DNA and electroporated in a 4 mm gap cuvette using two consecutive pulses of 1 . 2–1 . 8 kV , 200 Ω resistance and 50 µF capacitance . For the Amaxa nucleofections , pellets containing 0 . 5–1 . 5×108 parasites were resuspended in 100 µl of Human T Cell solution ( Lonza , Levallois Perret ) , mixed with 20 µg of circular or linearized plasmids and subjected to nucleofection using the 5 different Amaxa programs . The total volume of transfections was adjusted to 3 ml with TV3 medium and incubated at 27°C . Forty eight hours after transfections , G418 ( Invitrogen , Life Technologies , Villebon sur Yvette ) was added to the cultures at a final concentration of 0 . 5 µg/ml to allow selection of recombinant T . vivax . Genetically engineered parasites , whose resistance to G418 was confered by the Neo gene , were selected over time when a massive cell death was observed in the cultures leaving colonies of stably transformed T . vivax behind ( generally after 10 days ) . Total parasite genomic DNA was prepared from in vitro cultures with pure link genomic DNA ( Invitrogen , Life Technologies , Villebon sur Yvette ) . Correct plasmid integrations were checked by PCR using standard techniques and upFrProm/upRrProm or downFrProm/downRrProm oligonucleotides pairs and Dream Taq polymerase ( Fermentas , Villebon sur Yvette , France ) . A luciferase assay kit ( Roche Molecular Biochemicals; Mannhein , Germany ) was used to monitor luciferase expression . Serial dilutions of parasite suspensions were washed in PBS and pellets were resuspended in 150 µl of cell lysis buffer . Debris was removed by centrifugation . The lysates were then transferred into white , 96-well microplates ( Dynex Technologies , Chantilly , France ) . Light emission was initiated by adding the luciferin-containing reagent , in accordance with manufacturer instructions . The plates were immediately transferred to the luminometer ( Berthold XS3 LB960 ) and light emission measured for 0 . 1 s . Luminescence was expressed as Relative Light Units ( RLU ) . Wild type or TvGFP parasites were recovered from 14 days axenic cultures in TV3 medium . Adherent and supernatant cell populations were washed and resuspended in PBS −0 . 5% glucose balanced salt solution ( 2×106 cells/ml ) containing 1 µg/ml of propidium iodide . Two-color acquisition was carried out with a FACScalibur cytofluorometer ( Becton Dickinson ) . Dead cells were excluded from the analysis by gating out propidium iodide-stained cells . Parasites were gated on forward-light scatter/side-light scatter combined gate , and 40000 events were acquired . Results were analyzed by FlowJo software ( Tree Star , Inc ) .
We started T . vivax adaptation to axenic cultures using parasites previously adapted in vivo in mice [8] . More specifically , we began by inoculating HMI107 or B media with BSF purified from infected mice and incubating the preparations at 27°C , as described by Hirumi et al . in 1991 or by Gumm , in 1991 , respectively [14] , [15] . Various protocols were tested , e . g . different BSF levels , different concentrations of the various amino acids , different pH values , temperatures , reducing agents , and type of flask , but none of the cultures developed . Since the absence of glucose had previously been described as a factor triggering T . brucei BSF differentiation into procyclic forms [26] , we used purified bloodstream forms of T . vivax to inoculate TV1 and TV2 media that varied in composition mainly in terms of serum nature and/or the presence of glucose ( see Table 1 for details ) . BSF parasites were observed to be highly mobile for the first 3 days of cultivation in both TV1 and TV2 media . By day 4 , some parasites started to attach to the surface of the plastic and showed some of the morphological changes commonly seen in BSF that are differentiating into epimastigote forms , as previously described by Gumm [14] . Differentiating parasites replaced the prominent undulating membrane by a flagellum that emerges from the anterior portion of a shorter body and an anterior kinetoplast . However , the parasites were still unable to divide , and this in both TV1 and TV2 media , and they died after 7 days or 14 days , respectively . When grown in TV3 medium - that contains an equivalent mixture of complete IMDM and DMEM media ( vol/vol ) - BSF attached to the plastic flask after 4–5 days , suggesting that they were engaged in the process of differentiation into epimastigote cells . Three days later ( day 7/8 of culture ) , some parasites were seen to have shortened , indicating that they had differentiated into epimastigotes . Even more importantly , they started to multiply by forming small clusters ( Figure 1 A–C ) . These clusters increased in number and size and 3 to 4 weeks later had covered the entire surface of the culture flask . At this stage , both rosettes and free-swimming cells ( 1 . 5 . 107 cells/flask ) became abundant in the supernatant and these parasites could then be used to inoculate new flasks . In order to determine whether the mixture of the two sera and/or medium composition was critical for growth , parasites obtained from the TV3 culture supernatant were used to inoculate fresh flasks containing IMDM or DMEM supplemented with 10% FCS , 10% GS and different concentrations of glucose ( TV4 and TV5 media , respectively ) , comparable to those in TV1 and TV2 media . Only IMDM-based medium ( TV5 ) supported T . vivax growth at similar kinetics to TV3 medium , and this regardless of the glucose concentration . Different combinations ( vol/vol ) of five batches of fetal calf sera and three batches of goat sera were able to support growth without any significant differences ( data not shown ) , indicating that the positive effect of goat serum on T . vivax growth is not an artifact due to serum batch heterogeneity . Conversely , parasites were unable to grow in media supplemented solely with 20% FCS or 20% GS . Medium TV3 was therefore chosen for all subsequent cultures and experiments . Parasite growth kinetics and metacyclogenesis were investigated to characterize the parasite stages observed during axenic culture . As can be seen in Figure 2 , TV3 medium was inoculated with 1 . 5 . 107 cultured parasites and their development was monitored for three weeks . The parasites attached to the surface of the plastic within 2 hours ( Figure 2A ) and formed micro-colonies after 7 days ( Figure 2B ) . Conspicuous parasite multiplication was then observed for the following week and cells completely covered the entire surface of the plastic between days 14 and 21 ( Figure 2C and 2D ) . Parasite cell numbers were determined in the supernatant and the number of adherent cells was evaluated after scraping . As shown in Figure 2E , the number of cells in the supernatant increased with time and in proportion to the density of the adherent cell layer . A confluent 25 cm2 flask was able to produce 5 . 107 to 1 . 108 parasites in the supernatant every two days . With appropriate care and a medium amendment ( see below ) , it was possible to conserve parasite viability in a single flask for 6 to 10 weeks . In addition , the supernatants provided a sufficient number of parasites to initiate new cultures and thus support regular in vitro passages weekly or every 2–3 weeks . In efforts to determine whether established culture conditions were suitable for the differentiation of epimastigotes into infective metacyclics , we monitored the proportions of the different parasite forms in ongoing cultures . The relative positions of nuclei and kinetoplasts were evaluated by immunofluorescence after DNA staining with DAPI , and flagellum length and position were determined using antibodies to label paraflagellar rod protein 2 ( PFR2 ) [23] . As shown in Figure 3A , the kinetoplast in epimastigote forms was located between the nucleus and the anterior part of the cell body , whereas metacyclic-like trypomastigotes had a longer body-attached flagellum and a kinetoplast posterior to the nucleus . Numerous epimastigotes in the cultures were also observed to be dividing . Changes in parasite forms present in the supernatant and in the adherent layer were monitored throughout the culture period and the populations in each developmental stage were quantified on culture days 7 , 14 and 21 . We observed that the different populations that made up the adherent layer did not change in proportion over time , with the vast majority ( about 74% ) of cells consisting of epimastigotes throughout the plastic colonization period ( Figure 3B ) . The total proportion of epimastigotes was stable throughout the culture and only 24 to 32% were actively dividing cells . Some metacyclic-like cells were also observed in the population of attached cells , but accounted for only a small and invariant proportion ( about 5% ) . Conversely , changes in the trypomastigote population in the supernatant suggested that an active process of epimastigote differentiation into metacyclic cells ( metacyclogenesis ) was taking place . For instance , the parasite population in the supernatant after 7 days was similar to that observed in the adherent layer . A substantial change then occurred from day 14 with a dramatic increase in the proportion of metacyclic-like parasites ( 3 to 19% ) . This was accompanied by a considerable decrease in the number of epimastigotes and abnormal cells in the supernatant . The population of metacyclic-like cells peaked at this point then decreased by day 21 . The virulence of these axenic trypomastigotes was assessed by first collecting parasite populations from ongoing culture supernatants and analyzing these by immunofluorescence . The results showed that 2% , 19% and 16% of metacyclic-like cells were present on days 7 , 14 and 21 , respectively . Equivalent numbers of metacyclic-like cells ( 2×106 ) were then injected intraperitoneally in 3 groups of 5 Balb/c mice and parasitemia was measured over a period of 28 days ( Figure 3C ) . The results showed that 40% , 80% and 20% of individual blood smears contained BSF parasites ( >104 parasites/ml ) between days 12 and 21 . Since equivalent numbers of metacyclic-like cells were injected into the mice , our results indicated that the metacyclic-like cells present in the cultures were at different stages in their maturation and that the axenic differentiation process is not synchronous . Marked alkalinization of the TV3 medium was observed less than 30 minutes after the epimastigotes became attached to the plastic surface ( the pH increased from 7 . 4 to 8 . 6 ) . When the parasites were left at this high pH , they adopted a round shape and died . This phenomenon was prevented by increasing the HEPES final concentration to 100 mM and in this manner the pH was held at 7 . 4 during the initial growth process . We observed that adherent parasites could not be easily removed from the plastic surface and attempts to use a cell scraper were unfruitful , causing death in most of the parasites . Since the pH appears to be critical for adhesion , we analyzed whether pH variations impacted on parasite attachment . Confluent flasks were washed twice with PBS at pH 7 . 4 , pH 6 or pH 8 . 5 . Each of the three washing conditions led to partial cell detachment ( around 108 cells ) but the adherent cells that remained ( approximately 2×108 ) could then be easily scraped off the plastic surface without cell damage . The removal of medium ( and probably serum ) , rendered the attached cells less cohesive and loosened cells that presented higher levels of viability . Such a procedure was employed to provide 3 . 108 cells in a T25 confluent flask , and these were used to reinoculate fresh flasks or to carry out further experiments . Since L-proline is well known to be an important source of carbon for trypanosomes [27] , T . vivax epimastigote growth was also estimated at different proline concentrations ( 1 mM , 2 mM , 4 mM ) in TV3 medium . 107 cells were used to inoculate T25 flasks and the time required to obtain a confluent layer of cells was determined . As the proline concentration in the flask increased , the time required to obtain confluence decreased , from 3–4 weeks ( 1 mM proline ) to 14 days ( 4 mM proline ) . Since variations in glucose concentrations did not affect parasite growth , the results here indicate that proline is a key player in the growth of T . vivax epimastigotes . Our data showed that 4 mM L-proline was the optimal concentration in TV3 medium . Higher concentrations did not significantly improve culture conditions or parasite growth and maintenance . The effects of epimastigote density on cultivation were studied by performing a limiting dilution assay using fresh or conditioned media [28]–[30] . Here , 107 to 104 parasites were used to inoculate flasks containing TV3 media . The density of adherent cells and the presence of micro-colonies on the plastic surface were scored after 3 weeks of culturing at 27°C , with regular changes of the media . Epimastigotes used to inoculate fresh TV3 media , at concentrations of less than 106/per T25 flask became round , were unable to multiply and died after a week . By contrast , when using conditioned media containing 30% supernatant from former T . vivax cultures , flasks inoculated with only 105 parasites gave rise to dense parasite clusters after 4 weeks . Micro-colonies 2 mm in diameter were also observed after 4 weeks in flasks inoculated with 104 cells , and these reached confluence by 6 weeks . No genetic manipulation of T . vivax has ever been described in the literature , nor any expression of transgenes . We therefore began by constructing plasmids containing the luciferase reporter gene ( see Material and Methods ) in efforts to determine appropriate conditions for reproducible transfection of T . vivax . Figure 4A schematically represents pTv-Luc vector that was specially designed in order for T . vivax to express the luciferase gene and neomycin phosphotransferase ( NeoR ) , which confers resistance to G418 . Upstream of the reporter gene we cloned the 5′UTR of the T . vivax proline racemase gene ( PRAC ) that contains an efficient spliced donor acceptor site [9] . Work in T . congolense and T . brucei has previously identified RNA pol I promoter elements in regions spanning 2 kb upstream of the 18 S element of the rDNA gene cluster [31] , [32] . But , taking T . cruzi specific vectors as an example [33] , where poor gene expression is observed in the absence of such sequences , we aimed to provide the T . vivax vector with a hyperexpression cassette to regulate gene transcription . In order to better define the region with a putative T . vivax promoter , we constructed two different plasmids harboring respectively a long 1 . 8 kb fragment ( pTvLrDNA-luc ) and a short 1 . 2 kb fragment ( pTvSrDNA-luc ) upstream the 18 S rDNA gene ( see Figure 4B ) . Axenic T . vivax epimastigotes were then transiently transfected with the pTv-LUC , pTvLrDNA-luc or pTvSrDNA-luc plasmids using a Gene Pulser electroporator under high voltage conditions . Luciferase activity was measured in the different cell lines 48 h after transfection to check for the presence of the putative pol I promoter in selected sequences . Interestingly , luciferase activity was 10 times higher after transfection with all plasmids containing the putative rDNA promoter sequence compared to transfection with the pTvLUC plasmid . Comparable levels of luciferase activity were obtained with the pTvLrDNA-luc and pTvSrDNA-luc plasmids , suggesting that the rDNA promoter region in T . vivax is located in a 1 . 2 kb region directly upstream of the 18 S ribosomal DNA gene ( Figure 4B ) . However , given that the pTvLrDNA-luc plasmid has a better potential for recombination , subsequent work conducted to optimize T . vivax transfection conditions was conducted using this vector . Different concentrations of circular plasmid molecules were tested , and 20 µg was observed to be the optimal DNA concentration for T . vivax transfection studies , as based on the RLU obtained and as illustrated in Figure 5A . Different numbers of axenic parasites were then electroporated with 20 µg of circular plasmid . The results obtained showed that 1 . 5 . 108 parasite cells/per transfection gave the highest RLU in the supernatant ( Figure 5B ) . Then , in order to check that an acceptable success rate was obtained with T . vivax transfection , the relative efficiency of the gene transfer was measured by comparing transfection using a Gene Pulser system and Amaxa nucleofection . With 20 µg of circular plasmid used in each of the transfer conditions , the epimastigotes were subjected to 4 different Gene Pulser system voltage conditions and to 5 different nucleofection Amaxa programs . Examination of the cells after Gene Pulser electroporation showed massive rates of mortality at all voltages used ( not shown ) . Moreover , the parasites were unable to adhere to the plastic flasks , and this precluded any growth in TV3 medium . By contrast , as described previously for T . brucei and T . congolense [34] , [35] , the Amaxa nucleofection method greatly enhanced transfection efficiency and additionally , T . vivax showed better adhesion to the plastic surface and increased survival rates . Thus , as can be seen by the expression of luciferase monitored 24 h after gene transfer , the transfection efficiency obtained with Amaxa program “×006” was 25 fold higher than that obtained with the best Gene Pulser conditions ( two 1 . 6 kV pulses ) ( Figure 5C ) . This program was therefore used in all subsequent experiments . Efficient selection conditions were then determined conducive to obtaining stably pTvLrDNA-luc transfected parasites . 107 parasites were used to inoculate fresh flasks containing various concentrations of G418 ( 10 to 0 . 25 µg/ml ) , and TV3 medium was changed every 2 or 3 days . A G418 concentration of 0 . 5 µg/ml was sufficient to kill all the cells after 10 days of culture . In order to maintain the requirement for ‘quorum sensing’ on transfectant growth , cells were maintained in TV3 conditioned media in the presence of G418 for at least 4 weeks . In order to obtain stably transfected parasites , we targeted the T . vivax ribosomal region by using AfeI linearized pTvLrDNA-luc ( see Figure 4B ) to transfect parasite cells and compared the results with transfections using the circular plasmid . Following nucleofection , the cells were used to inoculate 2 independent flasks . After 24 h , one of the flasks was assayed for luciferase activity . Cells transfected with linear DNA showed half the luciferase activity of those transfected with circular DNA ( Figure 6A ) . But after several weeks of selection with G418 , the cells transfected with circular DNA started to decay and eventually died whereas those cells transfected with linear DNA formed small clusters and 5 to 10 micro-colonies per flask after 4 weeks of selection . Transfection efficiency using linear DNA was then estimated as 1 . 5–3 . 0×10−7 . The parasites reached confluence 4 weeks later and the supernatants could be transferred into fresh flasks for selection in the presence of G418 . After selection , parasite genomic DNA was prepared from two independent cultures stably transfected with pTvLrDNA-luc and submitted to PCR using two primer pairs to ascertain whether homologous recombination had occurred , as indicated in the illustration in Figure 6B . Consistently with integration of the pTvLrDNA-luc plasmid into the 18 S rDNA region , fragments of the expected size ( 1 . 8 kb ) were obtained after amplification for both culture DNA ( Figure 6C , upper and lower pannels ) , indicating that homologous recombination had occurred upstream the TvPRAC 5′ UTR ( Figure 6C , upper pannel ) and downstream the Tvtubαβ ( Figure 6C , lower pannel ) regions . The parasites were maintained in axenic TV3 medium for several weeks and luciferase light emission per parasite was shown to be stable over time for at least 24 weeks ( Figure 6D ) and correlated linearly with the number of parasites on a wide range ( >than 4 logs , Figure 6E ) . Since wild type ( WT ) metacyclic-like parasites produced in axenic cultures can successfully infect mice , we examined whether this was also the case for the stable metacyclic-like forms developed in vitro from the TvLrDNA-luc parasite line . Initially , the equivalent to 2×106 TvLrDNA-luc metacyclic-like forms obtained from 14-day cultures ( see Figure 3C ) were used to infect BALB/c mice . At onset of BALB/c parasitemia , a corresponding number of BSF ( 102 ) from TvLrDNA-luc or WT-infected mice were injected into groups of 4–6 Outbred mice and the parasitemias obtained were compared every 5 days , as previously described [8] . Figure 7 presents the number of BSF recorded during the infection , as well as relative survival in the two groups of mice . It can be seen that parasitemias and survival rates were similar and not significantly different in the mice infected with TvLrDNA-luc and those infected with WT parasites . Therefore , after only one passage in mice , the parasitemia kinetics obtained with cultured parasites were similar to those observed with BSF maintained exclusively by serial weekly passages in mice . Similar results were obtained after 2 months and after 12 months of axenic growth , indicating that both WT and mutant T . vivax maintained their infectivity after subculturing in vitro . Ongoing experiments have been performed to further validate the transfection procedures described here above by introducing into the T . vivax specific vector another reporter gene . For this aim , pTvLrDNA-GFP was constructed by replacing the luciferase gene among the TvPRAC 5′UTR and TvTubαβ vector sequences by a Green Fluorescent Protein ( GFP ) cassette excized from the pXS2-GFP ( [24] , see Methods and Figure 8A ) . T . vivax epimastigotes were transfected with pTvLrDNA-GFP using the Amaxa program “×006” and further cultured in TV3 medium at 27°C . Forty eight hours after transfection , G418 ( 0 . 5 µg/ml ) was added to the cultures to select stable transfectants . Similarly to the parasites transfected with pTvLrDNA-luc , massive mortality was observed up to 10 days with the simultaneous growth of stably ( neor ) transformed T . vivax that reached confluence four weeks later . The integration of the contruction was validated by PCR using upFrProm/upRrProm and downFrProm/downRrProm couple of primers , as decribed for the luciferase vector ( not shown ) . Figure 8B shows the microscopic observation of recombinant GFP - expressing parasites ( TvGFP ) obtained from 14 days axenic subcultures . Adherent and supernatant cell populations from two independent cultures were washed in PBS −0 . 5% glucose and analyzed by cytofluorometry . The FACS analysis shows that both adherent and supernatant TvGFP populations express highly homogenously the GFP ( app . 95% ) , as compared to the absence of fluorescence of WT parasites ( Figure 8C ) . Additional experiments are in progress to evaluate the behavior of TvGFP in vivo .
While human African trypanosomosis has drawn the attention of many research groups over the last few decades , less consideration has been given to AAT despite its considerable impact on livestock development and fertility and the economic hardship it causes in several countries . Major breakthroughs have recently been made in the study of T . congolense , namely the development and standardization of axenic cultures and the development of transfection techniques [35] , [36] . Researchers had showed growing interest in T . vivax adaptation to experimental animals from more than twenty years after the encouraging studies of Leeflang in the 1970s , and the description being made of parasite isolates that were infective to rodents [22] , [37] , [38] . Much attention had also been paid for several years to the development of short-term axenic cultures [14]–[16] , [39] . However , few studies conducted over the next two decades quoted these reports on T . vivax in vitro growth or parasites obtained in culture , possibly due to methodology inconsistencies and/or difficulties reproducing the complexity of parasite interactions with its host environment [19] . But T . vivax still remains a threat for livestock in Africa and South America where the disease is considered as emergent and is consequently the subject of numerous outbreak reports [3] , [40] . To gain better insights into the biology of T . vivax and its interactions with its mammalian hosts , we recently undertook a detailed study of a pathogenic strain of the parasite that had been isolated in West Africa and stored frozen for several decades . We used this IL 1392 strain to establish novel mouse models of experimental infection and immunopathology [7] , [8] . Today , we report herein on how we managed to overcome the lack of genetic tools for analyzing T . vivax gene expression and function by standardizing the axenic conditions for epimastigote cultivation of the IL 1392 strain and its in vitro differentiation into infectious forms . We also constructed specific vectors appropriate for parasite transgenesis , developed suitable conditions for T . vivax transfection and for further selection of transfectants . Moreover , transfection procedures were further validated by the engineering of green fluorescent parasites that stably express the GFP reporter gene . Finally , we carried out an in vitro and in vivo adaptation of a transgenic TvLrDNA-luc parasite line that constitutively expresses the luciferase reporter gene . Our data show that the TvLrDNA-luc mutant went through all the T . vivax developmental stages in vitro , in the same manner as WT parasites , and that metacyclic-like forms of the mutant are infective to immunocompetent mice . In order to overcome the difficulties found to reproduce culture conditions described in the past we worked to optimize axenic protocols to develop standard conditions of T . vivax maintenance and growth . For instance , although fetal calf ( FCS ) or goat ( GS ) sera had previously been presented as a essential medium components for trypanosome sustenance and growth in vitro [14] , [16] , our experiments showed that T . vivax was unable to grow in culture media supplemented only with FCS or GS . Successful parasite axenic cultures were only possible when a mixture of FCS and goat serum ( GS ) was used to complement the media , and this resulted in significant BSF attachment , in significant differentiation into epimastigotes and in further parasite growth . Moreover , in the presence of GS , any batch of FCS could be used without affecting parasite differentiation and development . We nonetheless noted that the number of parasites loaded into the cultures had an impact on axenic epimastigote cultivation . For instance , at low densities ( <106 cells/ml ) , T . vivax was unable to pursue its developement and the epimastigotes died in a few days without undergoing differentiation or further divisions . Interestingly , and in agreement with previous observations in T . brucei [28]–[30] , this restriction can be overcome by ensuring that up to one third of the conditioned medium consists of supernatants from former parasite cultures , and in this manner fewer epimastigotes can initially be loaded ( i . e . 104 parasites/ml ) . This observation suggests that conditioned medium contains signaling compounds or growth factors that are released by T . vivax in culture and these stimulate and support the proliferation and development of new cells or at least assist in their maintenance under axenic conditions . It is noteworthy that the axenic culture conditions described herein are suitable for epimastigote differentiation and for the continuous production of metacyclic-like forms . T . vivax from the adherent layer differentiate into infective forms without requiring any special medium adaptation , thus mimicking the gradual process of metacyclogenesis in culture . Interestingly , and in contrast to T . congolense [35] , T . vivax parasites that underwent metacyclogenesis in vitro from epimastigotes and were then conserved by regular axenic passages for more than one year , retained their infectiveness in immunocompetent mice . And it is of note that the factors involved in metacyclogenesis per se are not yet known in nature or described in the literature , for T . vivax or T . congolense . We cannot foresee whether the optimization of the axenic parasite cultures developed here can be automatically extrapolated to different wild type or other cloned T . vivax strains . But , a recent report using T . congolense showed that the time period to achieve the adaptation in axenic culture of different parasite strains and derivatives is strain dependant [35] . Nevertheless , previous molecular analysis of IL 1392 has proven the common origin of this and the South American and Asian isolates that phylogenetically pertain to the same clade [41] . It is possible that similarly to IL 1392 , strains from the same clade can be cultured using the protocols described herein . The development in the 1990s of Trypanosomatid transfection using Gene Pulser systems [42]–[45] was then adapted to other species [36] , [46] , [47] . Despite a number of similar transfection parameters shared by all Kinetoplastidae , recombination efficiencies and susceptibility to drug selection diverged . Thus , crucial adjustments were necessary to ensure appropriate transfection rates , such as the use of specific parasite regulatory sequences . The vectors we constructed to establish T . vivax transfection are based on classical models where foreign genes are placed under the control of species-specific 5′ and 3′ UTRs containing the regulatory sequences required for appropriate gene expression . Consequently , the ribosomal promoter-containing sequence was localized and inserted upstream of the selected transgenes to promote their expression and genomic recombination and thus obtain stably modified transfectants . For purposes of validating our T . vivax-specific overexpressing vector , we compared the archetypal Gene Pulser transfection system with Amaxa nucleofection technology . Amaxa technology has been reported , with T . brucei , to greatly increase the number of transfectants obtained [34] , [48] . In line with this , an initial series of experiments showed that Amaxa protocols had two major advantages over Gene Pulser conditions: they improved transient transfection efficiency and increased parasite viability and adhesion to the surface of the culture flask . Subsequent experiments showed that transfection of the T . vivax-derived circular vector was unable to generate stable transfectants . This may suggest that T . vivax is not able to maintain episomal DNA , unlike T . brucei and T . congolense that carry out particular sequences ( i . e . parp ) that promote episomal mantainance [36] , [48] . Alternatively , T . vivax circular DNA may be integrated but less efficiently than linear plasmids , resulting in a smaller number of live parasites that do not survive culture conditions . By contrast , when the linearized vector was used for the transfections , this generated stable transfectants that showed appropriate integration of the foreign gene into the ribosomal promoter region . This result shows that the linearized vector facilitates integration , as already shown for other kinetoplastids [33] , [49] . In conclusion , the T . vivax strain generated in our studies and stably expressing a luciferase reporter gene will be very useful for characterizing the in vivo infectious process and for validating the effectiveness of drug candidates in medium or high-throughput screening tests . Bioluminescent T . vivax will also certainly prove useful in enhancing our knowledge of the different aspects of parasite development , the acquisition of virulence and the triggering of pathology . However , cellular trafficking and localization of a given stage specific parasite protein may be promptly assessed by parasite engineered with fluorescent specific vector described here , where the current GFP cassette would be replaced by a gene of interest fused to the GFP reporter . Under the same promoter elements , transgenic parasites should express the fusion protein linked to GFP and easily identified . The work described herein has therefore developed the first specific genetic tools for the study of T . vivax biology and opens up new possibilities for the study of experimental Nagana , particularly the expression and regulation of critical genes implicated in the parasite's evasion of the host immune system . Additionaly , our work also paves the way for the development of more sophisticated tools to reduce the expression of parasite genes by inducible RNAi or by conventional gene knockout based on homologous recombination . | Trypanosoma vivax is a major parasite of domestic animals in Africa and Americas . Most studies on this parasite have focused on gathering epidemiological data in the field . Studies on its biology , metabolism and interaction with the host immune system have been hindered by a lack of suitable tools for its maintenance in vitro and its genetic engineering . The work presented herein focused on determining axenic conditions for culturing and growing insect ( epimastigote ) forms of T . vivax and prompting their differentiation into metacyclic forms that are infectious for the mammalian host . In addition , we describe the development of appropriate vectors for parasite transgenesis and selection in vitro and their use in analyzing genetically modified parasite lines . Finally , we report on the construction of the first T . vivax recombinant strain that stably expresses a foreign gene that maintains its infectivity in immunocompetent mice . Our work is a significant breakthrough in the field as it should lead , in the future , to the identification of parasite genes that are relevant to its biology and fate , and to work that may shed light on the intricacies of T . vivax–host interactions . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"parastic",
"protozoans",
"trypanosoma",
"genetics",
"molecular",
"genetics",
"biology",
"microbiology",
"protozoology",
"genetics",
"and",
"genomics"
]
| 2011 | Genetic Engineering of Trypanosoma (Dutonella) vivax and In Vitro Differentiation under Axenic Conditions |
Identifying and eliminating snail habitats is the key measure for schistosomiasis control , critical for the nationwide strategy of eliminating schistosomiasis in China . Here , our aim was to construct a new analytical framework to predict high-risk snail habitats based on a large sample field survey for Oncomelania hupensis , providing guidance for schistosomiasis control and prevention . Ten ecological models were constructed based on the occurrence data of Oncomelania hupensis and a range of variables in the Poyang Lake region of China , including four presence-only models ( Maximum Entropy Models , Genetic Algorithm for rule-set Production , Bioclim and Domain ) and six presence-absence models ( Generalized Linear Models , Multivariate Adaptive Regression Splines , Flexible Discriminant Analysis , as well as machine algorithmic models–Random Forest , Classification Tree Analysis , Generalized Boosted Model ) , to predict high-risk snail habitats . Based on overall predictive performance , we found Presence-absence models outperformed the presence-only models and the models based on machine learning algorithms of classification trees showed the highest accuracy . The highest risk was located in the watershed of the River Fu in Yugan County , as well as the watershed of the River Gan and the River Xiu in Xingzi County , covering an area of 52 . 3 km2 . The other high-risk areas for both snail habitats and schistosomiasis were mainly concentrated at the confluence of Poyang Lake and its five main tributaries . This study developed a new distribution map of snail habitats in the Poyang Lake region , and demonstrated the critical role of ecological models in risk assessment to directing local field investigation of Oncomelania hupensis . Moreover , this study could also contribute to the development of effective strategies to prevent further spread of schistosomiasis from endemic areas to non-endemic areas .
Schistosomiasis japonica , a water-borne parasitic disease , has long been prevalent in 12 provinces located along the Yangtze River in China . It is caused by infection with the parasite Schistosoma japonicum , which has a unique intermediate host , Oncomelania hupensis . During the past 60 years , China has made remarkable progress towards eliminating schistosomiasis japonica . Much of this progress was made during the World Bank Loan Project ( 1992–2001 ) , e . g . , prevalence was reduced by more than 50%[1] . Additional control activities were initiated in 2005 based on replacement of draft cattle with mechanisation and increased health education , which resulted in further gains in controlling this disease . Up to 2015 , 12 endemic provinces located along the Yangtze River had reached the stage of transmission being interrupted or controlled[2] . Despite declining schistosomiasis prevalence year-by-year and an overall low prevalence , this disease remains a serious problem in China . Transmission risk remains high in some regional areas , especially the lake and marshland regions . It is alarming that the area of snail habitat in some regions ( e . g . Hubei , Hunan , Jiangxi and Anhui ) are still extensive and new or infective snail habitats are still being discovered , especially in these lake and marshland regions ( e . g . Dongting Lake and Poyang Lake regions ) [3] . This represents a serious challenge for preventing a rebound in prevalence and spread of schistosomiasis in China . Therefore , how to predict high-risk areas accurately and allocate the limited health resources effectively has become the key and future focus of schistosomiasis prevention and control activities . It is widely acknowledged that the epidemiology of S . japonicum infections has a particular spatial characteristic because it depends on the presence of its sole intermediate host snail , O . hupensis , whose reproduction is governed by specific climatic and environmental condition[4] . It is certain that where schistosomiasis occurs ( excluding imported cases ) , O . hupensis exist . Therefore , controlling the spread of O . hupensis , especially infected hosts , will lead to long-term effective schistosomiasis control since killing and eliminating snail habitats is the most sustainable measure for controlling this disease . However , this is challenging from the perspective of a national control strategy because of the massive investment of manpower , physical and financial resources , and potential environmental pollution . Knowledge about a species’ ecological and geographical distribution is of great concern in fields such as ecological conservation and species biodiversity . To assess the spatial distribution of O . hupensis , field investigations have been widely applied . However , for many regions ( including mountains and swamp ) , detailed and accurate data on spatial distribution of snail habitats are not available because collecting such data via snail surveys is labour-intensive and costly , and some places are difficult to access . Consequently , with the advent of techniques including remote sensing ( RS ) , global positioning system ( GPS ) and geographical information system ( GIS ) , many researches are increasingly relying on the classification of RS images as a means to predict snail distributions . Previous studies have used indices–such as land surface temperature ( LST ) , normalized difference vegetation index ( NDVI ) , and normalized difference water index ( NDWI ) –extracted directly from RS images as predictors of distributions pattern[5] . These methods can assist in understanding spatial patterns of snail habitats at a broad scale . However , because of poor accuracy and high misclassification error associated with those methods , exploring a variety of new predictive models based on statistical algorithms to predict patterns of snail distribution has been growing rapidly[6 , 7] . Ecological niche models ( ENMs ) is a powerful method for producing predictive risk maps of species distribution and has been widely applied to many species in various fields , especially invasive species[8 , 9] . The concept of “ecological niche” was first proposed by Joseph Grinnell in 1917 to define species distribution patterns ( excluding migration ) [10] . Based on this concept , ENMs are developed to predict the geographical distribution of species through analysing the statistical relationships between known localities of the species of interest and risk variables . How to compare the relative performance of different models and make the final selection of best modelling strategy remains a challenge[11 , 12] . Four lake and marshland provinces along the Yangtze River ( Hubei , Hunan , Jiangxi and Anhui ) are high-prevalence regions for schistosomiasis japonica , accounting for approximately 94% of the infected persons in China[13] . Within this region , Dongting Lake and Poyang Lake regions account for 86% of infections[3] . Therefore , swamp and lake areas are the current major focus for schistosomiasis prevention and control . As the largest freshwater lake in China , Poyang Lake constitutes the largest endemic area for schistosomiasis and undoubtedly is a priority for schistosomiasis control . Poyang Lake is located in northern Jiangxi Province and near the southern bank of the middle and lower reaches of the Yangtze River ( Fig 1 ) . Five main tributaries , the rivers Gan , Fu , Xiu , Rao and Xin , flow into Poyang Lake , which empties each year into the Yangtze River via Hukou[14] . During flooding ( April to June ) , water from these five rivers recharge Poyang Lake . The water level peaks from July to September . Thus , flood plains develop during the period April to September . Between November and early March , the water subsides and this area becomes a landscape of marshlands[15] . As a result , Poyang Lake is a typical seasonal water-land transition lake with dramatic fluctuations in water level , historically ranging from 12 m to 17 . 2 m measured by Xingzi hydrometric station; this corresponds to a change in water surface from about 500 km2 in the dry season to up to approximately 3 700 km2 ( 7-fold increase ) in the rainy season . Seasonal variations in the water level , vast marshlands as well as an extensive grassland plain , provides an excellent environment for the growth and development of O . hupensis . To date , of 12 administrative counties ( cities or districts ) located along Poyang Lake , four ( Jinxian , Hukou , Lushan and Gongqing ) have reached the national criterion for interruption of Schistosomiasis japonica transmission ( schistosoma japonicum infection rate below 1% in humans and bovines , and no acute schistosomiasis in humans found and no schistosome-infected snails detected for two consecutive years ) , while seven counties ( Duchang , Yongxiu , Yugan , Xingzi , Xinjian , Nanchang and Poyang ) have reached the national criterion for transmission control ( no infected schistosomiasis in humans and bovines , as well as no infected snails found for five consecutive years ) [16] . Therefore , Poyang Lake remains a key area of endemic schistosomiasis in China . In this study , we aimed to identify high-risk snail habitats in the Poyang Lake region using an integrated ecological niche modelling approach , with both presence-absence ( PA ) and presence-only ( PO ) models discussed . This study was based on O . hupensis occurrence data and a range of variables , including remote sensing derived data , biological climate data , soil-related data and economic-related data , to develop a new technical framework for predicting snail habitats . These results can also provide guidance on–and a theoretical basis for–the national surveillance and control of schistosomiasis .
The extent of potential snail habitats in Poyang Lake depends on the areas with fluctuation of water level . Previous studies have shown that about 95% of snail habitats are distributed in the plains ( elevation 14 to 17 m ) ; at elevation ≤13 or >17 m , areas are not suitable as snail habitat[17] . The highest ( 17 . 16 m elevation on July 25 , 2007 ) and lowest ( 12 . 58 m elevation on March 24 , 2003 ) water levels have been measured at Xingzi hydrometric station in Jiangxi Province , which were the corresponding date for the highest and lowest elevation used in this study . The study area covers about 3 , 636 km2 . All maps were rescaled to spatial resolution of 30m to conform to the data structure using the resampling technique in ArcGIS 10 . 0 . The nearest neighbor method was used to resample categorical variables and the approach of bilinear interpolation was applied instead for continuous variables . In addition , the masking technique was also applied in ArcGIS 10 . 0 to extract exactly the same geographic extent of all maps . Considering suitable elevation for snail habitat ( 14–17m ) , we selected the geographic boundary of Poyang Lake on July 25 , 2007 with elevation of 17 . 16m as the mask to obtain the largest and potential snail habitat areas . The number of ENMs available for predicting species distribution is immense and can be classified into two categories based on the data types used for model construction . Predictive maps produced only based on species’ occurrence data are categorized as presence-only models ( hereafter “PO” models ) and those requiring both presence and absence data are classified as presence-absence models ( hereafter “PA” models ) . Four PO models ( Maximum Entropy Models ( MAXENT ) , Genetic Algorithm for rule-set Production ( GARP ) , Bioclim and Domain ) and six PA models ( Generalized Linear Models ( GLMs ) , Multivariate Adaptive Regression Splines ( MARS ) , Flexible Discriminant Analysis ( FDA ) , as well as machine algorithmic models Random Forest ( RF ) , Classification Tree Analysis ( CTA ) , Generalized Boosted Model ( GBM ) ) were constructed . See S1 Text for details on using these models . Predictions from each model were compared to each other based on three indices , receiver-operating characteristic curve ( ROC curve ) , Kappa and calibrating plot . Firstly , ROC graphing was applied to evaluate the discrimination performance of these models . Discrimination performance is the ability of a model to correctly distinguish occupied from unoccupied sites , which can be measured by setting a threshold and predicting the species to be present or absent at a site based on whether the model prediction is above or below the threshold . The most common measure of discrimination is the area under the ROC curve ( AUC ) . ROC curve is a two-dimensional graph , with true positive rate ( sensitivity ) as Y axis and false positive rate ( 1–specificity ) as X axis; it can be used to illustrate the relationship between sensitivity and specificity . Secondly , Cohen’s kappa , which is also called consistency test , was adopted to visualize a model’s consistency . The kappa statistic defines the accuracy that might have resulted by chance alone . It ranges from –1 to +1 , where +1 indicates perfect agreement between prediction and observation , and –1 indicates complete disagreement . In addition , values of 0 or less indicate agreement no better than random classification . To enable a standardized evaluation of all models , threshold values ( Maximized value of the kappa; Maxkappa ) were applied to transform all predictions to binary ( presence-absence ) predictive maps . Finally , our calibration plot compares the agreement between predicted probability of occurrence and observed proportions of sites occupied . It shows the ability of the model to make unbiased estimates of the probability of the outcome[19] . The calibration plot can be developed by breaking the predicted probabilities up into several intervals , and plotting the proportion of evaluation sites that are observed to be occupied within each of these intervals against the median predicted value of each interval . When the points lie closely along a 45° line then the model is well calibrated and has good agreement[20] . Based on the comprehensive assessment of results , the model with an ROC curve value >0 . 8 , kappa value >0 . 5 and relative good calibration was defined as superior models . Superior models were used to build the following ensemble model and the variable importance for detecting snail habitat were further evaluated . The importance of each variable was estimated using a permutation procedure . Models with superior predictive accuracy were combined together to construct an ensemble model to generate the final predicted risk map of snail habitats through the method of AUC-based weighted average[21 , 22] .
ROC curves indicated that the ten model’s performance varied markedly from each other in their predictive ability ( Table 2 and Fig 2 ) . Discriminative ability of RF ( AUC = 0 . 96 ) was best among current models and its variance was lowest ( AUC . sd = 0 . 004 ) . Discriminative ability was relative better for MARS , GBM , CTA and Domain ( AUC = 0 . 81 , 0 . 85 , 0 . 88 and 0 . 88 , respectively ) . However , MAXENT and Bioclim models showed poorest discrimination with AUC<0 . 7 . The average AUC for the six PA models was 0 . 84 , outperforming the four PO models ( AUCmean = 0 . 74 ) . The trends observed in kappa were generally similar to those assessed with AUC ( Table 2 ) . RF–which had the highest discriminative ability as assessed by AUC–also had the highest kappa scores ( 0 . 82 ) . Consistency as assessed by Kappa was generally higher for CTA and Domain than GBM , MARS , FDA , GLM as well as GARP , and poorest for MAXENT and Bioclim among current models . The mean kappa value for PA models ( 0 . 58 ) was also higher than that for PO models ( 0 . 38 ) . For the results of calibration plots and goodness-of-fit tests ( Fig 3 ) , models including GLM , FDA , CTA and GBM were relatively better calibrated–calibration plots nearing the 45° line . In addition , MAXENT , GARP , Bioclim and Domain models were poorly calibrated . Taken together , RF , CTA , GBM and MARS showed superior accuracy among these ten Ecological Niche Models constructed in this study . Four models ( RF , CTA , MARS and GBM ) with superior accuracy were used to construct the final ensemble model and assess the importance of variables . To evaluate the final ensemble model , another small dataset , including 822 absence records and 178 presence records , was manually collected in the field during 2016 for external validation . Its AUC was 0 . 89 , sensitivity was 0 . 79 and specificity was 0 . 82 . As shown in Fig 4 , the vertical bar indicates the importance rank of each variable contributing to these four models , and the horizontal bar shows the average importance rank of each variable . The smaller the latter value , the more important is the variable . LST , DEM and NDVI were found to be the most important factors influencing these four models , making the highest contribution . Other environmental factors that significantly influenced the model were Bio17 , Bio15 , Bio4 and landuse , which were amongst the top ten important determinants of the geographical distribution of snail habitats . It was noted that the level of GDP also showed a relatively high contribution to these predictive models . The predictive risk map for snail habitats in Poyang Lake region generated by the above ensemble model was expressed as a probability ( Fig 5 ) . Probability of the presence of snail habitats ranged from 0 to 1 , with a color closer to red indicating climatic , environment and economic factors that are more suitable for the growth or reproduction of snails . The distribution risk of potential snail habitats in Poyang Lake region were classified into five grades ( very low , low , moderate , high and very high ) with areas of 2 096 . 1 km2 , 640 . 1 km2 , 371 . 2 km2 , 322 . 5 km2 and 206 . 9 km2 , respectively , where population at risk were 136 300 , 53 900 , 30 600 , 11 700 and 3 800 , respectively . The area most suitable for the presence of snail habitats are located in: ( 1 ) middle part of Xingzi County through which the River Xiu and the River Gan flow; Yugan County ( the River Fu and the River Xin ) ; Poyang County ( the River Rao ) and the north-east part of Poyang County; ( 2 ) intersection of Xinjian , Yongxiu and Duchang counties; ( 3 ) other high-risk areas scattered along the peripheral regions of Poyang Lake . Overlaid with our previous results on identified high-risk clusters of schistosomiasis [14] , the final high-risk snail habitat map was generated and the area was classified into four grades in terms of urgency for prevention and control actions ( Fig 5 and Table 3 ) . The highest risk level and most significant areas were very high risk snail habitat within schistosomiasis clusters , which are mainly located in Cluster 4 and Cluster 2 , as well as partly in Cluster 3 , covering an area of 52 . 3 km2 . Areas belonging to the first grade were mainly distributed in the watershed of the River Fu in Yugan County as well as the watershed of the River Gan and the River Xiu in Xingzi County . The second grade was high and moderate-high risk snail habitat within schistosomiasis clusters , which were located partly in Xingzi County and in most regions in the middle part of Yugan County , covering areas of 72 . 8 km2 and 79 . 7 km2 , respectively . The third grade was very high risk snail habitat outside schistosomiasis clusters , covering an area of 154 . 6 km2 . The fourth grade was high and moderate risk snail habitat outside schistosomiasis clusters , covering areas of 249 . 7 km2 and 291 . 5 km2 , respectively .
As the largest freshwater lake and typical marshland areas , Poyang Lake region is widely known as a suitable site for endemic schistosomiasis and is a major concern for schistosomiasis control in China . This study has attempted to predict high-risk snail habitats in the Poyang Lake region by ecological niche modelling to provide assistance for future targeted monitoring and control . This study presents predictive maps of the spatial distribution of Oncomelania hupensis , the key intermediate host snail of schistosomiasis , which is vital for long-term sustainable effective schistosomiasis control in the Poyang Lake region . The MAXENT model showed the worst predictive performance in our study , which is different from previous reports by Scholte et al[6] , who predicted the geographical distribution of Schistosomiasis mansoni’s intermediate host and claimed that MAXENT has high accuracy . The MAXENT model was sensitive to sampling errors , which might easily introduce a bias to modeling . The GARP model can be greatly influenced by selection of the training data set and has large uncertainties because of different results for each computation . Predictive results of the GARP model are always a discrete value of 0 and 1 , hence showed a weak predictive accuracy . The Bioclim model could be considered as the simplest model in our study since it just assumes a rectilinear environment envelope; it does not include correlations or interactions between factors[12] . This study showed that the Bioclim model showed relatively poor performance . Similar results have been reported in many studies predicting the geographical distribution of multi-species using Bioclim and those studies have reported poor results[12] . Domain model was found to have excellent discrimination ability but weak calibration , which agrees with previous studies[23] . Similar to Bioclim , Domain is also based on an environment envelope , but it has the ability to include discontinuity of the species’ records . Therefore , excellent discrimination ability of Domain model might be due to intense spatial auto-correlation in snails’ presence records . However , Domain model also has a major limitation: for each potential site , only the nearest neighbor point within the envelope is used to determine its suitability for species , which might result in weak calibration . Most parameters for the four PO models were maintained at their default settings because of a lack of reasonable prior information , which might have introduced some subjectivity and thus bias . Of the six presence-absence models used in this study , GBM , CTA and RF were based on machine learning algorithms of classification trees , desirable to handle unordered , nonlinear and multidimensional data . They produced relatively higher accuracy compared to other models used in our study . RF , especially , showed the best predictive performance and could predict the geographical distribution of snail habitats accurately . We also found that these six PA models differed from each other in their complexity . RF can be considered the most complex: it fits many classification trees to a data set , and then combines the predictions from all the trees . In our study , it showed the highest predictive accuracy . CTA and GBM are also relatively less complex than the other models . Hence , it seems that increased model complexity may contribute to better predictive performance . For MARS , it showed relatively good accuracy , indicating that there might be non-linear correlation between snail habitats and risk variables , and the variables might have complicated associations among each other , which explains why classical statistical models such as GLMs only have normal predictive performance . With regard to the prediction of snail habitats , we found that the overall predictive performances of PA models–especially machine learning algorithm based approaches–outperformed PO models . This is consistent with the results obtained by Brotons[24] et al . , who proposed that absence data could help to identify low suitability areas that might have otherwise been classified as good habitats if only presence data were used . This study also confirmed that snail habitats are related to multiple factors including several ecological environment variables and social-economic variables . LST , DEM and NDVI were the top three variables affecting snails’ geographical distribution . Similar results have also been reported in many previous studies[25 , 26] . Meanwhile , the indexes of precipitation–including Bio17 and Bio15 –also showed relatively higher importance . These important environment factors ultimately affect soil moisture content and vegetation density to influence the growth and reproduction of snails . In addition , social-economic index of GDP was also shown to have high contribution to the predictive ability of the model . The economic development of Poyang Lake region has been quite unbalanced; undeveloped counties-including Jiujiang , Hukou , Pengze , Xingzi , Poyang , Duchang and Yugan-belong to traditional agricultural regions with quite a low level of mechanization . Most local residents in these regions live by farming , fishing or grazing livestock . Considering that livestock from free-range farming is an important infection source for schistosomiasis , areas with a traditional pattern of farming and grazing–especially in economically undeveloped regions–should be a focus for schistosomiasis control . The high-risk areas of both snail habitats and schistosomiasis were found to be mainly concentrated in the confluence of Poyang Lake and five main tributaries , which have a high risk of schistosomiasis epidemics and require significant attention . One potential explanation is that these areas appear frequently to alternate between land and water in different seasons , providing an ideal environment for snail growth and reproduction . Furthermore , most of the local residents in these areas live by farming or fishing , which increases their contact with contaminated water . Xingzi , in particular , is an undeveloped county , and the majority of infected individuals in this county are farmers or fishermen . They become infected mostly due to the behaviors of grazing or fishing[14] . The classification of prevention and control priority levels for high-risk snail habitats , as shown in Fig 5 , could provide precise guidance for local authorities to formulate an effective targeted control strategy . The primary and most urgent for prevention and control of both schistosomiasis and snails is the first grade , which is mainly distributed in the watershed of the River Fu in Yugan County as well as the watershed of the River Gan and the River Xiu in Xingzi County . This region has a very high risk of both schistosomiasis and snails; hence it has a high prevalence of infected snails . The epidemic conditions for infection and transmission in this region might be very easily to meet , therefore it has a high spread risk of schistosomiasis , even outbreaks . Therefore , areas of first grade should be prioritized for schistosomiasis prevention and control . Meanwhile , other grades also should be paid attention based on the available health resources . However , it should be noted that so-called “high-risk sites” in our study represent “suitable locations for the snail habitats”; that does not mean the snails will always be present at the predicted sites . One issue deserves a discussion . Data resolutions were different in the original variables used , which might affect predictive accuracy somewhat , especially variable importance due to the scale transformation . It would be informative to explore whether and in what degree different scale and scale changes impact the predictive accuracy . In addition , variables used to construct the models in our study came from different years , which might be a limitation for this kind of study of habitat identification . However , the spatial distribution of variables e . g . land use and vegetation type did not show large changes during shorter time periods in Poyang Lake area , and therefore was unlikely to have a great influence on the study results . Furthermore , all the absence records for the process of modelling building were generated from background areas of the study extent without field validation although we checked them via the techniques of visual interpretation of remote sensing images , which might introduce some uncertainties . But we did validate the final model’s performance with field investigated data , including both the presence and absence records of snail habitats . In summary , this study depicted the spatial distribution of snail habitats in the Poyang Lake region through diverse ecological niche models and compared their performance in terms of predictive accuracies . The results showed that the overall predictive performance of presence-absence models outperformed presence-only models in prediction of snail habitats . Furthermore , increasing model complexity might contribute to predictive accuracy and models based on machine learning algorithms of classification trees present higher accuracy than others . Prediction distribution of snail habitats can provide precise guidance and a theoretical basis for the targeted surveillance and control of schistosomiasis in the Poyang Lake area . The high-risk areas of both snail habitats and schistosomiasis are mainly concentrated in the confluence of Poyang Lake and five main tributaries , which have a high risk of schistosomiasis epidemics and require significant attention . Among these , the most serious and significant areas were distributed in the watershed of River Fu in Yugan County as well as the watershed of River Gan and River Xiu in Xingzi County . | Oncomelania hupensis is the sole intermediate host of Schistosoma japonicum and it is critical for the long-term sustainable control and elimination of schistosomiasis . We suggest a new analytical framework to predict high-risk snail habitats based on ecological models . Their predictive performance were compared via three indices ( receiver-operating characteristic [ROC] curve , Kappa value and calibrating plot ) and further they were classified into three grades with different discrimination ability . The four best models were chosen and combined to identify snail habitats , facilitating classification of the snail habitats in the Poyang Lake region into four risk grades , which should be prioritized for interruption of the snail life cycle , with control intensity and scope dependent on the available health resources . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"invertebrates",
"ecology",
"and",
"environmental",
"sciences",
"medicine",
"and",
"health",
"sciences",
"malacology",
"china",
"population",
"dynamics",
"tropical",
"diseases",
"geographical",
"locations",
"parasitic",
"diseases",
"animals",
"oncomelania",
"gastropods",
"aquatic",
"environments",
"habitats",
"bodies",
"of",
"water",
"neglected",
"tropical",
"diseases",
"population",
"biology",
"snails",
"zoology",
"lakes",
"marine",
"and",
"aquatic",
"sciences",
"molluscs",
"people",
"and",
"places",
"helminth",
"infections",
"schistosomiasis",
"eukaryota",
"freshwater",
"environments",
"asia",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences",
"organisms",
"geographic",
"distribution"
]
| 2019 | Identification of high-risk habitats of Oncomelania hupensis, the intermediate host of schistosoma japonium in the Poyang Lake region, China: A spatial and ecological analysis |
The cellular endosomal sorting complex required for transport ( ESCRT ) machinery participates in membrane scission and cytoplasmic budding of many RNA viruses . Here , we found that expression of dominant negative ESCRT proteins caused a blockade of Epstein-Barr virus ( EBV ) release and retention of viral BFRF1 at the nuclear envelope . The ESCRT adaptor protein Alix was redistributed and partially colocalized with BFRF1 at the nuclear rim of virus replicating cells . Following transient transfection , BFRF1 associated with ESCRT proteins , reorganized the nuclear membrane and induced perinuclear vesicle formation . Multiple domains within BFRF1 mediated vesicle formation and Alix recruitment , whereas both Bro and PRR domains of Alix interacted with BFRF1 . Inhibition of ESCRT machinery abolished BFRF1-induced vesicle formation , leading to the accumulation of viral DNA and capsid proteins in the nucleus of EBV-replicating cells . Overall , data here suggest that BFRF1 recruits the ESCRT components to modulate nuclear envelope for the nuclear egress of EBV .
The endosomal sorting complex required for transport ( ESCRT ) machinery is conserved evolutionarily and involved in catalyzing the scission of membrane necks in endosome sorting , biogenesis of multivesicular bodies ( MVBs ) , cytokinesis and release of enveloped virions . In contrast to the cellular membrane-scission protein dynamin family , which cleaves membrane necks by constricting them from the outside , membrane scission mediated by the ESCRT machinery is from inside the neck ( reviewed in [1] , [2] ) . The ESCRT components ( also known as class E proteins ) consist of five multiprotein complexes , ESCRT-0 , -I , -II , -III , Vps4 ( vacuolar protein sorting-4 ) ATPase , and several ESCRT-associated proteins [3] , [4] . ESCRT-0 , -I , and -II are soluble complexes that shuttle between cytosolic and membrane-bound forms . These components sequentially coordinate together to bud the membrane and recruit ESCRT-III for the scission of membrane neck . ESCRT-III proteins belong to the Chmp family and are soluble monomers that assemble on membranes to form tight filamentary spirals and are released from the membranes at the final stage with other ESCRT proteins by the transient ATP-dependent reaction of Vps4 . In addition to the regular composition , cellular ESCRT-I protein TSG101 ( tumor susceptibility gene 101 ) alternatively activates the spiral assembly of ESCRT-III through bridging by the ESCRT associated protein apoptosis linked gene-2 interacting protein X ( Alix ) [5] . Because these class E proteins are recruited sequentially and assembled for their functions , the interaction-disrupted mutants of Alix and Chmps , as well as the ATPase activity defective Vps4 ( e . g . Vps4AE228Q ) , are useful tools to investigate the involvement of the ESCRT machinery in various biological processes [6]–[9] . In addition to physiological functions , components of the ESCRT machinery are used by many enveloped viruses for budding and release from cells . By sequence comparison , late-budding ( L ) domains have been identified extensively in the structural proteins of these viruses by three conserved tetrapeptide motifs [Y ( L ) XXL , PT/SAP and PPXY] that mediate the recruitment and interaction of class E proteins to facilitate virus budding [10] , [11] . Among the L domains , Y ( L ) XXL- , PT/SAP- , and PPXY-type motifs interact specifically with ESCRT associated Alix , TSG101 and Nedd4-like E3 ubiquitin ligases ( e . g . Trp-Trp-domain-containing protein-1 ) , respectively . Substitutions in the interacting motifs of the structural proteins also lead to defects in virion maturation and release [6] . The dynamics of ESCRT protein recruitment in retroviruses were found to be extremely transient ( ∼1–3 min ) and sufficient for their functions on the membrane for virus release [12] . In contrast to enveloped RNA viruses , the contribution of ESCRT machinery to the maturation of enveloped herpesviruses remains to be explored . Herpesviruses are large DNA viruses associated with human and animal diseases . After viral DNA replication , the newly synthesized genomes are packaged into pre-assembled intranuclear capsids . Based on the current envelopment-deenvelopment-reenvelopment model , large-sized herpesviral nucleocapsids ( 115–130 nm ) begin budding through a transient envelopment process with the nuclear envelope . This is first mediated by the viral protein kinase and nuclear membrane associated proteins at the inner nuclear membrane ( INM ) for the local disassembly of compact nuclear lamina for primary envelopment . After release from the nuclear envelope derived structures , the nucleocapsids subsequently become associated with viral tegument proteins and glycoproteins at cytoplasmic apparatuses for final maturation of virions ( reviewed in [13] , [14] ) . So far , the involvement of ESCRT in virion release , and cytoplasmic reenvelopment of herpes simplex type 1 ( HSV-1 ) and human cytomegalovirus ( HCMV ) have been characterized using dominant negative ( DN ) inhibitors of ESCRT and siRNA strategies [7]–[9] , [15] . By immunofluorescence analysis , the cytoplasmic nucleocapsids and envelope components are associated with MVB and colocalized with endosomal markers in infected cells [16] , [17] . As observed by electron microscopy ( EM ) analysis , HHV-6 also induces MVB formation and cytoplasmic egress through an exosomal release pathway [18] , suggesting that herpesviruses use the ESCRT machinery for their membrane-dependent maturation in the cytoplasm . Knowledge regarding the nuclear egress of nucleocapsids has emerged only recently . Several viral proteins have been characterized and shown to regulate the primary envelopment of herpesviruses from the nuclear membrane , in particular the herpesvirus conserved homologs of UL34 and UL31 . The homologs of HSV-1 UL34 are type II integral membrane proteins that localize predominantly to the INM , outer nuclear membrane ( ONM ) and ER , whereas UL31 homologs are nuclear matrix-associated phosphoproteins [19]–[23] . The homolog pairs UL34/UL31 are conserved among herpesviruses and codependent for their localization to the nuclear rim . They also share the ability to interact with nuclear lamin proteins . Transient overexpression of the UL34/UL31 homologs of HSV-1 or HCMV induces subtle alterations of the nuclear lamina , which are distinguishable from the dramatic redistribution of lamin proteins in cells replicating the virus [24]–[26] . This suggests that the homologs of UL34 and UL31 potentially regulate the structure of the nuclear membrane , but coordination with other viral proteins is required for the fine-tuning of nuclear egress . In addition to viral products , the preformed UL34/UL31 homolog complexes in alpha- and beta-herpesviruses can recruit cellular factors , such as PKCs and lamin B receptor ( LBR ) , and assemble into complexes to facilitate the nuclear egress of nucleocapsids [27]–[30] . However , the cellular factors contributing to the nuclear membrane budding ( primary envelopment ) of nucleocapsids remain elusive . Epstein-Barr virus ( EBV ) is a gammaherpesvirus that infects most of the human population . During lytic infection , EBV encodes several gene products that modulate the cellular environment and facilitate virion maturation [14] . The study of the nuclear egress of EBV was hampered particularly by the lack of an efficient replication system in vitro . In previous studies , we found that EBV BGLF4 kinase regulates the structure of nuclear lamina to facilitate the initiation of nucleocapsid egress [31] , [32] . In addition to BGLF4 kinase , the gene products of BFRF1 and BFLF2 , the UL34 and UL31 positional homologs of HSV-1 , were shown to regulate the primary egress of nucleocapsids [33] . By an unclear mechanism , transient expression of BFRF1 in 293 cells induces the formation of multiplied nuclear membranes and cytoplasmic cisternal membrane structures , suggesting the potential contribution of EBV BFRF1 to membrane modification during nuclear egress [33] . More than 30 years ago , a pioneering ultrastructural study found that EBV lytic replication in transformed B cells induced alterations of the nuclear membrane consisting of deep enfolding or multilayered structures , accompanied by some irregular vacuoles with electron dense material in the cytoplasm ( Figure 3 in [34] ) . Recently , using advanced electron tomography , special nuclear invagination structures of both nuclear membranes , containing multiple viral nucleocapsids , also were seen in cells infected with the replication-efficient murine gammaherpesvirus 68 ( MHV-68 ) [35] . In order to explore the cellular machineries involved in EBV maturation , we set out to study the contribution of the ESCRT machinery to the EBV maturation process .
To determine whether the ESCRT machinery participates in EBV maturation , we used the EBV converted nasopharyngeal carcinoma ( NPC ) cell line NA , in which lytic replication can be induced by the expression of the viral immediate early protein Rta [36] . Among the ESCRT proteins , Chmp proteins mediate the formation of functional membrane scission complexes , and Vps4 controls the release of ESCRT complexes from membranes [2] , [4] , [37] . Inhibition of Chmps using GFP-tagged Chmp proteins , or Vps4 using catalytically inactivated Vps4-DN , traps the class E complexes and inhibits virion release by various viruses [8] , [9] , [12] , [38] . Therefore , NA cells were transfected with Rta accompanied by the dominant negative ( DN ) forms of class E protein GFP-Chmp4b or catalytically inactivated Vps4AE228Q mutant ( Vps4-DN ) plasmid . At 96 h post transfection , the cellular DNA and culture supernatant were harvested to test for EBV virion release . With about 50% transfection efficiency , we found the expression of GFP-Chmp4b or Vps4-DN reduced the amounts of secreted virions ( ∼44% and ∼22% of that of GFP transfected cells ) as detected by quantitative PCR of the BamHI W fragment of the EBV genome ( Figure 1A ) . Even not very significant , there was a slight decrease of intracellular viral DNA in GFP-Chmp4 expressing cells and a slight increase of intracellular viral DNA in Vps4-DN expressing cells . In immunobloting , we found Rta transfection induced the expression of the EBV immediate-early protein Zta , early proteins BMRF1 , BFRF1 , BFLF2 , and BGLF4 , and major capsid structural proteins BcLF1 ( VCA ) and glycoprotein BLLF1 ( gp350 ) , and the expression was not affected by the coexpression of GFP-Chmp4b or Vps4-DN ( Figure S1A ) . This indicates that inhibition of the ESCRT machinery interferes with EBV replication or virion release . To investigate the contribution of ESCRT proteins in EBV maturation , we examined the distribution of the ESCRT proteins in EBV replicating NA cells . Slide-cultured NA cells were transfected with plasmid expressing GFP or GFP-Chmp4b and Rta plasmid to induce the lytic cycle for 48 h . The subcellular localization of the viral nuclear envelope protein BFRF1 and the dominant negative GFP-Chmp4 were observed first . We found the BFRF1 protein showed a cytoplasmic punctate and perinuclear distribution in GFP-expressing Rta-induced NA cells ( Figure 1B ) . With a moderate ability to block virion release in Figure 1A , the GFP-Chmp4b proteins were redistributed from a diffuse punctate pattern in vector plasmid transfected cells into clumping structures and colocalized with BFRF1 at the perinuclear region in cells replicating EBV ( Figure 1C ) . Different from Chmp4b , Chmp1b is an accessory protein for MVB formation and dispensable for HIV budding [39] , [40] . Here Chmp1b-GFP showed a homogenous distribution in vector transfected NA cells ( Figure S1B , upper panels ) , whereas EBV reactivation by Rta transfection redistributed Chmp1b-GFP to cytoplasmic puncta , and colocalized with viral BFRF1 ( Figure S1B , lower panels ) , suggesting Chmp1b is also involved in EBV reactivation-induced nuclear envelope modification . Among the ESCRT proteins , bridging protein Alix is progressively recruited by membrane anchoring Gag for retroviral assembly [12] , [39] . Chmps are subsequently activated by Alix in vivo [5] . Here we found in vector- or Rta-expressing EBV negative NPC-TW01 cells , the parental cell line of NA cells , Alix was localized in the cytoplasm in a diffuse pattern , similar to that of vector transfected cells ( Figure S1C ) . In confocal images , EBV reactivation for 48 h apparently enhanced intranuclear and nuclear rim distribution of Alix ( Figure 1D ) . The nuclear envelope morphology was affected by virus replication , and Alix was partially colocalized with viral BFRF1 protein at the nuclear rim ( in 65% reactivated cells , merge ) . In cells with punctate BFRF1 containing structures , Alix showed nuclear rim or cytoplasmic distribution ( in 35% reactivated cells ) , suggesting Alix may be released into cytoplasm after the formation of BFRF1 puncta . Because the ESCRT components function through a very transient interaction , it is reasonable that only small amounts of Alix were coimmunoprecipitated with HA-BFRF1 protein by anti-HA antibody in EBV reactivated NA cell lysates ( Figure 1E , lane 4 ) . Alix antibody precipitated the specific protein but without coimmunoprecipitating HA-BFRF1 , possibly because of epitope blockage or instability of the immunocomplexes ( Figure 1E , lane 2 ) . Data here suggest Alix , as well as other ESCRT proteins , are redistributed to nuclear envelope associated compartments and associated with viral BFRF1 in cells replicating EBV . While membrane modulating-ESCRT proteins were relocated near to the nucleus , the nuclear envelope structure of Rta-transfected cells was then monitored through immunostaining of viral and cellular nuclear membrane-associated proteins . We found that viral BFRF1 and BFLF2 were colocalized at the nuclear rim in EBV reactivated NA cells . Oriented punctate structures with perinuclear BFRF1 or BFLF2 staining were observed in about 90% of NA cells , especially in the concave region of the kidney shaped nucleus ( Figure S1D , inset ) or in the cytoplasm ( Figure S1D , arrowhead ) . With Rta transfection , we found INM-associated lamin A/C and emerin were redistributed into the nuclear periphery or cytoplasmic region of the BFRF1-positive punctate structure in NA cells , but not in NPC-TW01 cells ( Figure 1F and 1G ) . Cellular DNA was also detected in a proportion of cytoplasmic punctate structures with lamin and BFRF1 staining ( in about 15–20% of cells expressing BFRF1 , Figure 1F , arrowhead ) , suggesting that these cytoplasmic vesicles are derived from the reorganized nuclear membranes . Collectively , data here indicate that EBV reactivation redistributes ESCRT proteins , reorganizes the nuclear envelope and induces cytoplasmic nuclear envelope-derived structures . Because viral BFRF1 and BFLF2 colocalized with nuclear membrane proteins in cytoplasmic punctate structures , we determined whether BFRF1 or BFLF2 is responsible for the nuclear envelope reorganization using a transient expression system . Expression of HA-BFRF1 alone showed nuclear rim and cytoplasmic distribution in transfected HeLa cells ( Figure 2A and 2B ) , whereas Flag-BFLF2 was predominantly distributed in the nucleus ( Figure 2C and 2D ) . Different from the clumped perinuclear patterns or oriented punctate structures observed in cells replicating virus ( Figure 1C , 1F , 1G and S1D ) , HA-BFRF1 induced dispersive vesicles in the cytoplasm of transfected cells ( Figure 2A ) . Compared with the typical nuclear rim staining pattern of lamin A/C and emerin in vector transfected cells ( Figure 2A and 2B , upper panel ) , cytoplasm-redistributed nuclear emerin with vesicle structure was found colocalized with HA-BFRF1 at the perinuclear region of 45∼55% cells with BFRF1 expression ( Figure 2B , Merge ) . In contrast , the nuclear rim staining of lamin A/C and emerin was not affected by the expression of Flag-BFLF2 ( Figure 2C and 2D ) . In addition to INM-anchoring proteins , we also detected the distribution of nuclear pore complexes by FG repeat-specific mAb414 . FG repeat-containing nucleoporins showed a nuclear rim staining in control plasmid transfected cells ( Figure 2E ) . Expression of HA-BFRF1 induced the redistribution of FG-repeat containing nucleoporins into perinuclear and intranuclear puncta , partially associated with perinuclear HA-BFRF1 containing vesicles . Because BFRF1 and BFLF2 function together for the nuclear egress of EBV , the distribution of both proteins was detected in cotransfected HeLa cells ( Figure 2F ) . Similar to cells expressing HA-BFRF1 alone , HA-BFRF1 distributed at the nuclear rim and in dispersive cytoplasmic vesicles in cells co-expressing HA-BFRF1 and Flag-BFLF2 . Flag-BFLF2 either showed nuclear rim and cytoplasmic colocalization patterns with HA-BFRF1 or was expressed in the nucleus of co-transfected cells . This suggests that BFRF1 alone is capable of inducing nuclear membrane associated vesicles , whilst BFRF1 can further recruit the nuclear distributed BFLF2 into these cytoplasmic vesicles in co-transfected cells . Because parts of BFRF1-induced vesicles were located in the perinuclear compartment , we wondered whether these vesicles associate with particular cellular organelles . Therefore HeLa cells were cotransfected with plasmids expressing BFRF1 and YFP-tagged organelle retention sequences , including for the ER , Golgi , and endosomes . Compared with the specific localization of the various organelles in vector transfected cells , expression of HA-BFRF1 redistributed organelle markers into an uneven and slightly clumped pattern ( Figure S2A and S2B ) . Without a conventional organelle targeting sequence , BFRF1 was predominantly colocalized with ER markers at the nuclear rim and nucleus-extended structure . A small portion of BFRF1 was colocalized with the Golgi markers in the cytoplasm and with endosome markers at the nuclear periphery ( Figure S2B , merge ) , suggesting that BFRF1-induced vesicles may fuse progressively with other cytoplasmic organelles . To determine directly the effect of BFRF1 expression on nuclear membrane alteration and vesicle formation , we visualized the subcellular structures in HeLa cells expressing BFRF1 by transmission electron microscopy ( TEM ) . Compared to the smooth nuclear membrane and homogenous cytoplasm in vector transfected cells ( Figure 3A ) , multiple irregular cytoplasmic vesicles and altered nuclear membranes were observed in more than 30 HA-BFRF1 expressing cells ( Figure 3B ) . Both single- and multiple-layer membranes ( arrowheads ) were found in these irregular cytoplasmic vesicles ( Figure 3C ) . The nuclear membrane showed an irregular wavy structure ( Figure 3D ) , with cisternal multilayered membranes ( Figure 3E , a ) or the disappearance of margin integrity ( Figure 3E , star ) . Notably , a budding structure , similar to retroviral budding through plasma membranes , was also observed on the modified nuclear membrane of HA-BFRF1 transfected cells ( Figure 3E , b ) , indicating that expression of EBV BFRF1 alone is sufficient to induce vesicle formation through dramatic modulation of the nuclear envelope . Because the ESCRT components Chmp4b and Alix showed partial colocalization with BFRF1 at the nuclear periphery or rim of cells replicating EBV ( Figure 1 ) , we explored further the contribution of the cellular ESCRT machinery to BFRF1-mediated vesicle formation . We found wild-type GFP-Vps4A displayed diffuse cytoplasmic and nuclear fluorescence in vector transfected HeLa cells , whereas GFP-Vps4-DN accumulated as aggregates in the perinuclear space , as the characteristic class E compartment [12] ( Figure 4A and 4B , upper panels ) . Expression of HA-BFRF1 or CFP-BFRF1 caused the redistribution of GFP-Vps4A into cytoplasmic puncta and enhanced nuclear membrane associated structures with a partial colocalization pattern ( Figure 4A and S2C ) . Remarkably , expression of Vps4-DN abolished the BFRF1-induced vesicle formation in the cytoplasm ( Figure 4B and S2D ) . Vps4-DN colocalized with HA-BFRF1 on the nuclear membrane as connected froth-like structures or at the nuclear periphery as clumps , suggesting that functional ESCRT machinery is required for BFRF1-induced vesicle formation . Next , we explored the participation of ESCRT components in BFRF1 mediated vesicle formation . The distribution of the ESCRT-I protein TSG101 was first observed in HeLa cells co-expressing HA-BFRF1 and Flag-BFLF2 . Using two different antibodies ( r654 prefers the native epitopes and 4A10 detects a . a . 167–374 of TSG101 ) , we found that TSG101 showed a homogenous cytoplasmic distribution in vector transfected cells and was slightly enhanced to the nuclear rim with HA-BFRF1 and Flag-BFLF2 in co-expressing cells ( Figure S2E ) . A coimmunoprecipitation assay also supported the interactions between TSG101 and HA-BFRF1/Flag-BFLF2 complexes ( data not shown ) , implying that TSG101 may participate in BFRF1/BFLF2-mediated functions during virus maturation . In contrast to TSG101 , immunofluorescence staining showed the ESCRT adaptor Alix was expressed predominantly in a diffuse cytoplasmic pattern in vector transfected cells ( Figure 4C ) . With the expression of HA-BFRF1 , Alix was redistributed partially and colocalized with HA-BFRF1on the nuclear rim , similar to the patterns observed in virus replicating cells ( Figure 1D ) . Because Alix was found to bridge viral proteins to ESCRT in many cases , we suspected Alix may serve as the factor bridging BFRF1 and the cellular ESCRT machinery . We found that only small amounts of HA-BFRF1 and Alix were coimmunoprecipitated with GFP-Chmp4b by GFP antibody , even in the presence of Vps4-DN ( Figure 4D , lane 2 and 3 ) . Also , small amounts of Alix were coimmunoprecipitated with HA-BFRF1 by anti-HA antibody ( Figure 4D , lane 4 ) . Nevertheless , coexpression of Vps4-DN and GFP-Chmp4b , which presumably accumulate as unresolved class E complexes , significantly enhanced the interaction between BFRF1 and Alix ( Figure 4D , lane 6 ) , suggesting the ESCRT machinery participates in BFRF1 induced vesicle formation . To prove the role of Alix in bridging BFRF1 and the ESCRT machinery , a highly specific Alix siRNA [40] was then added to transiently transfected cells . Compared with control siRNA , double treatments with specific siRNA significantly reduced the expression of Alix at 48 , 72 and 96 h post transfection ( Figure 4E ) . In Alix knockdown cells , BFRF1 lost the ability to induce vesicle formation and showed a nuclear rim or cytoplasmic reticular pattern ( Figure 4 ) . Nevertheless , vesicles were still observed in cells with residual Alix signals , suggesting that a small amount of Alix is sufficient for the function of the ESCRT machinery ( data not shown ) . Interestingly , we noticed that the cytoplasmic accumulation of BFRF1 in Alix knocked down cells is different from the nuclear rim retention of BFRF1 in cells expressing Vps4-DN or GFP-Chmp4 , suggesting Alix may contribute to the nuclear targeting of BFRF1 . To determine whether the ability of BFRF1 to induce vesicle formation is conserved among other herpesviral homologs , HSV-1 HA-UL34 and Flag-UL31 were expressed in HeLa cells . As described previously [24] , HA-UL34 expression alone showed a perinuclear and cytoplasmic reticular pattern without vesicle formation in the cytoplasm , whereas Flag-UL31 was expressed predominantly in the nuclei of transfected cells ( Figure S3A ) . HA-UL34 partially redistributed the nuclear emerin to the perinuclear region , whereas emerin showed a relatively typical nuclear rim distribution in cells expressing Flag-UL31 . Co-expression of HSV-1 HA-UL34 and Flag-UL31 produced a colocalization pattern at the perinuclear puncta and thickened nuclear rim ( Figure S3B ) , which is considered to result from nuclear envelope alteration induced by UL34/UL31 complexes [41] . In a sequence alignment , unique regions were found in EBV BFRF1 and HCMV UL50 , in addition to the conserved domains ( Figure S3 ) . Altogether , data here suggest other BFRF1 homologs also modulate the architecture of the nuclear envelope , while the vesicle forming ability may not be conserved in BFRF1 homologs . Here , BFRF1 appears to be the first nuclear envelope associated protein using the ESCRT machinery through interaction with Alix . To characterize the region responsible for interaction with Alix , vesicle formation and membrane anchoring , we analyzed the BFRF1 protein sequence through motif search and alignment with other herpesviral homologs . No conventional L motifs ( PTAP , PPXY and YXXL ) were identified in BFRF1 , whereas two putative L-like motifs ( 62YKFL65 and 74YPSSP78 ) were found within the amino terminal putative late domain 1 ( LD1 , a . a . 8–65 ) and LD2 ( a . a . 74–134 ) of BFRF1 . In addition , a putative BFLF2 interacting domain ( ID , a . a . 135–179 ) and a transmembrane region ( TM , a . a . 314–336 ) were identified in the BFRF1 protein by multiple alignments with herpesviral homologs and protein secondary structure prediction ( Figure 5A and S3C ) . An EBV specific region ( ESR ) also was identified in the region a . a . 180–313 of BFRF1 . A series of HA-tagged BFRF1 deletion mutants was subsequently generated to test the functions of various regions . In confocal images , Alix located predominantly in the cytoplasm of vector transfected HeLa cells . Expression of BFRF1 slightly redistributed Alix close to the nuclear rim and BFRF1 was partially associated with Alix in nuclear peripheral vesicles ( Figure 5B , merge ) . The BFRF1ΔLD1 , ΔLD2 and ΔID proteins showed fragmented , aggregated patterns around the nucleus or were embedded in the nuclear membrane . Partially Alix-colocalized signals were observed in the nuclei of cells expressing BFRF1ΔLD2 and at the nuclear rim of cells expressing BFRF1ΔLD1 or ΔID . In contrast , BFRF1ΔESR was colocalized with Alix at the nuclear rim in a continuous smooth pattern , whereas BFRF1ΔTM is evenly distributed with Alix in the cytoplasm . It was notable that BFRF1-induced vesicles were abolished in all mutants ( Figure 5B ) , suggesting multiple regions of BFRF1 are required for vesicle formation . The EBV specific region of BFRF1 ( a . a . 160–313 ) may mediate the membrane budding and the putative TM is responsible for its nuclear envelope targeting . In confirming the BFRF1-Alix interaction domain , slow migrating bands in addition to the predicted molecular weight , which may resemble post translational modification , were seen clearly for ΔLD1 , ΔLD2 and ΔID of BFRF1 . Compared with the WT , reduced amounts of BFRF1ΔLD1 , ΔLD2 or ΔID were coimmunoprecipitated with Alix in both directions ( Figure 5C , lanes 9 to 12 ) , suggesting the region between LD1 and ID may mediate the interaction with Alix . Interestingly , deletion of ESR or TM enhanced the interaction between BFRF1 and Alix ( Figure 5C , lane 13 and 14 ) , suggesting without the sequential recruitment of the ESCRT proteins for vesicle formation , the abrogation of the membrane modulating or anchoring function of BFRF1 is likely to stabilize the interaction between BFRF1 and Alix . The ability of BFRF1 mutants to reorganize the nuclear membrane was also investigated using the INM-anchoring emerin as a marker in confocal analysis ( Figure S4A ) . Expression of WT BFRF1 induced the redistribution of emerin into small cytoplasmic puncta with partial colocalization with BFRF1 . Expression of BFRF1ΔLD1 , ΔID and ΔESR partially redistributed emerin to perinuclear clumps or into the nucleus with dispersed staining , whereas BFRF1ΔLD2 induced the BFRF1-emerin clumping in the nucleus . Interestingly , BFRF1ΔTM caused partial redistribution of emerin into the cytoplasm , suggesting BFRF1ΔTM can still disturb the integrity of the nuclear membrane . It is possible that BFRF1 modified the membrane architecture through direct or indirect interactions with other membrane-anchoring proteins . Next , the BFRF1-interacting regions in Alix were revealed . So far , three domains were identified in Alix , the amino terminal Bro domain ( a . a . 1–358 ) , the central V domain ( a . a . 362–705 ) and the carboxyl-terminal Proline-rich region ( PRR , a . a . 717–868 ) ( Figure 5D ) [42] . Using Flag-tagged functional fragments of Alix , we found that , with a small amount of the WT , the Bro and PRR domains , but not V , were coimmunoprecipitated with HA-BFRF1 ( Figure 5E , lanes 7 to 10 ) , suggesting there could be more than one contact region between BFRF1 and Alix . Again , the Bro and PRR domains alone appeared to form a more stable complex with BFRF1 , compared to that of WT Alix . In the current model of membrane morphogenesis , the intermolecular interactions of membrane anchoring proteins provide the force to curve the cellular membrane [43] , [44] . To determine the potential of BFRF1 to induce membrane curving , we detected the possible intramolecular interaction of BFRF1 by immunoprecipitation . In reciprocal immunoprecipitation of HA-BFRF1 and GFP-BFRF1 , we showed that BFRF1 can form dimers ( Figure S4B , lane 8 ) , potentially to force the membrane curving . We determined next whether the recruitment of ESCRT machinery by BFRF1 contributes to the nuclear egress of EBV nucleocapsids . In our subcellular fractionation analysis , PARP and α-Tubulin were clearly separated into the nuclear and cytosolic fractions ( Figure 6B , lanes 1 to 3 ) . Nucleoporin Nup62 was separated into both nuclear and cytoplasmic fraction as described [45] , indicating the protocol is suitable for identifying the components of different subcellular compartments . Simultaneously , the amounts of viral genome were first detected by quantitative PCR in reactivated NA cells in the presence of Alix siRNA or a plasmid expressing Vps4-DN ( Figure 6A ) . At 72 h post transfection , EBV reactivation enhanced the intracellular viral genome content by 7 . 5 fold ( total ) and 5 fold in the nuclear fraction . Treatment with Alix siRNA did not change the amount of intracellular viral genomes significantly but this was enhanced slightly by the expression of Vps4-DN , similar to the observation in Figure 1A . Remarkably , treatment with siAlix increased intranuclear viral genome content of Rta transfected NA cells by more than 2 fold . Expression of Vps4-DN also increased the intranuclear viral genome content ( ∼1 . 3 fold ) , suggesting that Alix and functional ESCRT machinery contribute to the nucleus to cytosol translocation of the EBV genome . As well as the viral genome , we also analyzed protein distribution in these cells by subcellular fractionation ( Figure 6B ) . Compared to Rta plasmid transfection alone ( Figure 6B , lanes 4 to 6 ) , additional treatment with siAlix led to significant accumulation of the major capsid protein BcLF1 in the nuclear fraction ( Figure 6B , lane 8 ) . With a 50∼60% transfection percentage , expression of Vps4-DN also increased the detection of BcLF1 in the nuclear fraction ( Figure 6B , lanes 13 and 15 ) , suggesting that functional ESCRT machinery regulates the nucleus to cytosol translocation of the major capsid protein . Because the replication efficiency of EBV is much lower than that of alpha- or beta-herpesviruses , we thought it would be difficult to quantify the intranuclear viral capsids using EM . Therefore , antibodies against viral capsid proteins were used to locate the viral capsids in cells replicating EBV . To this end , EBV-positive NA cells were cotransfected with Alix siRNA or plasmids expressing Vps4-DN and Rta and stained for Alix , BFRF1 , viral capsid component ( s ) and cellular DNA . In the immunofluorescence images , BFRF1 located predominantly at the nuclear rim surrounding the cellular DNA , with punctate structures in the cytoplasm of NA cells at 72 h post Rta transfection ( Figure S5A , vector ) . In the presence of siRNA control , the major capsid protein BcLF1 colocalized with BFRF1 in the cytoplasmic punctate structures ( Figure 6C , right panels ) . Knockdown of Alix caused the accumulation of BcLF1 in the nucleus . In addition , expression of Vps4-DN restricted BFRF1 distribution to the nuclear membrane and enlarged perinuclear structures ( Figure S5 , mCherry-Vps4-DN ) which may be derived from the extended nuclear envelope , such as the intranuclear nuclear reticulum or perinuclear ER compartment [46] . The viral capsid proteins , including BcLF1 and triplex proteins BORF1 and BDLF1 , also were detected to indicate the likely distributions of viral nucleocapsids . In cells expressing control DsRed protein , we found that viral capsid proteins were detected predominantly in the cytoplasm or at the margin of Rta-transfected NA cells at 72 h post transfection ( Figure 6D ) , suggesting the nucleocapsids may translocate through the cytoplasm for virion release . Remarkably , expression of Vps4-DN significantly promoted the accumulation of the viral capsid components BcLF1 , BORF1 and BDLF1 in the nucleus ( Figure 6E , overlaid with light green signals ) , suggesting that the ESCRT machinery regulates EBV maturation , very likely at the nuclear egress of nucleocapsids . Finally , because the ESR region is unique in EBV BFRF1 and BFRF1ΔESR shows strong colocalization with Alix at the nuclear rim , without vesicle formation ( Figure 5B ) , we sought to determine whether expression of the BFRF1ΔESR mutant interferes with virus maturation . We found that coexpression of BFRF1ΔESR with Rta in NA cells did not affect the amount of intracellular EBV genomes ( Figure 6A , last columns ) . Notably , BFRF1ΔESR expression increased the content of intranuclear viral genomes . In protein analysis , BFRF1ΔESR was co-fractionated with Nup62 , EBV Rta , BFRF1 and BcLF1 in the transfected cells ( Figure 6B , lanes 10 to 12 ) , implying that BFRF1ΔESR may trap viral DNA , nuclear envelope components and viral capsids in the cytoplasm . Consistently , BFRF1ΔESR not only induced the accumulations of Alix and emerin but also colocalized with capsid BcLF1 protein in a distinct perinuclear compartment , without cytoplasmic vesicle-like structures ( Figure S5B and S5C ) . In contrast , expression of WT BFRF1 seemed to enhance vesicle formation and the cytoplasmic BcLF1 staining pattern in reactivated NA cells ( Figure 6F and S5C ) . Overall , we assumed the coordination of multiple domains within BFRF1 is critical for recruiting the ESCRT machinery during viral nuclear egress; therefore BFRF1ΔESR may function as a dominant negative mutant for EBV maturation .
In this study , we aimed to explore the contribution of the ESCRT machinery to the maturation process of EBV . We provide evidence that EBV BFRF1 is the first identified nuclear envelope-associated protein that employs the cellular ESCRT machinery to induce nuclear envelope-derived cytoplasmic vesicles . Evidence here also suggests that a similar mechanism is used for the nuclear egress of EBV nucleocapsids . Using dominant negative mutants , we showed that the ESCRT machinery is important for the maturation of EBV virions ( Figure 1A ) . In cells replicating the virus , the ESCRT proteins were redistributed and partially colocalized with the viral BFRF1 protein at nuclear rim . Interestingly , the INM-associated proteins , lamin A/C and emerin , and some cellular DNA , were detected simultaneously within cytoplasmic BFRF1 associated punctate structures ( Figure 1F and 1G ) . In the subsequent immuno-staining and transmission EM analysis , we found that expression of BFRF1 alone is sufficient to modulate nuclear membrane structure and induce vesicle formation ( Figure 2 and 3 ) . However , the dispersed distribution of BFRF1-induced vesicles in HeLa cells was different from the polar distribution of BFRF1-containg puncta near the nuclear concave of EBV reactivated cells , suggesting that other viral proteins also may regulate the BFRF1-contaning structures during virus replication . Remarkably , BFRF1 alone was sufficient to recruit and interact with the ESCRT proteins Alix , Chmp4b and Vps4 ( Figure 4 ) . Expression of Alix siRNA or the dominant negative form Vps4 abolished BFRF1-induced vesicle formation from the nucleus-associated membrane . Because of the low replication efficiency of EBV in the epithelial cells , we were not able to quantify nucleocapsids directly under EM . However , Alix siRNA , Vps4-DN and the BFRF1ΔESR mutant accumulated viral DNA and major capsid components in the nuclei of cells replicating the virus ( Figure 6 ) . A hypothetical model of the coordinated action of EBV BFRF1 and cellular ESCRT components in modulating the nuclear membrane is proposed ( Figure 7 ) according to our observations and the electron tomography ( ET ) images from MHV-68 ( Figure 6 in [35] ) . We postulate that , after EBV reactivation , the membrane-anchoring BFRF1 may be translocated from the ER membrane to the nuclear membrane and subsequently initiate the protrusion of the nucleus associated membranes . BFRF1 then cooperates with the ESCRT proteins to form vesicles , which may be derived from double- ( Figure 3C and 7A ) and single-layered nuclear membrane ( Figure 3B , 3C and 7B ) . In one way , the vesicles budding from the INM may be subsequently fused with the ONM as the scenario proposed for HSV-1 [13] , [14] . The vesicles derived from the double-layered nuclear envelope or INM may contain some cellular nuclear components , such as lamin , emerin and even cellular DNA fragments . BFRF1 expression also induces the amplification of nuclear membranes , which potentially may provide more membranous materials for viral maturation ( Figures 3E and 7C ) . In cells replicating EBV , other viral factors are believed to cooperate with BFRF1 , regulate the function or subcellular localization of ESCRT components and aid packaging of nucleocapsids into nuclear membrane derived vesicles for nuclear egress . In our model , the mechanism through which BFRF1 and Alix translocate into the nucleus remains unclear . Recently , multiple pathways , such as nuclear pore complex ( NPC ) /Ran-GTP dependent pathway , a diffusion-retention mechanism and nuclear envelope trafficking have been shown to mediate the trafficking of various cellular integral and peripheral INM proteins [47] . Based on EM observations of the nuclear egress of HSV-1 , a putative NPC-independent vesicular trafficking pathway was proposed for the newly synthesized ONM-associated viral proteins to travel through the nuclear envelope lumen and subsequently fuse to INM [47] , [48] . In our study , we found Alix was distributed in the cytoplasm in cells not replicating EBV , while EBV reactivation or expression of BFRF1ΔESR redistributed a portion of Alix into the nucleus ( Figure 1D and 5B ) . Alix siRNA caused the accumulation of BFRF1 on the nuclear membrane or ER-associated apparatus in a reticular pattern in the cytoplasm , implying that Alix and BFRF1 may regulate their nuclear transport reciprocally ( Figure 6C ) . One possibility is that newly synthesized BFRF1 can oligomerize on the ONM and form a structure protruding towards the nuclear envelope lumen , then small vesicles may be formed through the recruitment of ESCRT components into the nuclear envelope lumen , and subsequently fuse with the inner nuclear membrane . Once BFRF1 is translocated onto the INM , the recruited Alix would then be released into the nucleus . There could be a reinitiated assembly of BFRF1 , Alix and other ESCRT components to promote the large vesicle formation towards the cytoplasm . This scenario is supported partly by the observation of large double membrane invaginations , containing multiple nucleocapsids and nuclear pore complexes , in the ET images of cells replicating MHV-68 ( Figure 6 in [35] ) . Further studies are needed to solve the various questions , including how virus controls the budding from single- or double-layered nuclear membranes , how the transport direction is controlled in cells replicating the virus , and whether the same ESCRT components are involved in both directions of vesicle formation . In addition , according to our organelle maker images , the BFRF1-induced nuclear envelope-derived vesicles may fuse with the cytoplasmic apparatus to change the membrane composition and promote tegumentation and , finally , viral maturation . Techniques such as immunogold-labeling , together with high-resolution imaging , may determine whether BFRF1containg cytoplasmic vesicles can use the ESCRT machinery to modulate cytoplasmic organelles . Regarding the functional domains , we found that the Bro- and PRR- regions of Alix showed stronger interactions with BFRF1 than that of the wild type Alix ( Figure 5E ) . Because the wild type Alix is involved in a very transient mode in coordinated association/dissociation among different domains for its function [49] , we suspect the interaction of a single domain of Alix with BFRF1 tends to be stabilized . In the functional domain mapping , LD1 ( a . a . 8–65 ) , LD2 ( a . a . 74–134 ) and ID ( a . a . 135–179 ) of BFRF1 are required for vesicle formation ( Figure 5B ) . Alix and emerin showed distinct subcellular distribution in the presence of BFRF1ΔLD1 , ΔLD2 or ΔID ( Figure 5B and S4A ) , suggesting that these regions are required for functional recruitment of ESCRT proteins and nuclear membrane remodeling . Further characterization of the functional motifs LD1 , LD2 and ID ( a . a . 8–179 ) of BFRF1 may suggest possible mechanisms involved in the vesicle formation process . However , unlike ΔLD1 , ΔLD2 and ΔID , ΔESR was strongly co-stained with Alix on the nuclear rim and induced intranuclear distribution of emerin ( Figure 5B and S4A ) . Collectively , these data indicate that different domains of BFRF1 may function at different stages of the vesicle formation process . In cells with BFRF1 homolog expression or herpesviruses replication , multilayered nuclear membranes , packaged cisternal and annulate lamellae structures are observed commonly ( Figure 7C ) and [18] , [22] , [50] . We speculate that uncoordinated interactions between the ESCRT proteins and membrane anchoring BFRF1 homologs may induce the formation of multilayered class E compartment-like structures . Further EM studies may be required to reveal how BFRF1 derived vesicles traffic in the cytoplasm to facilitate the cytoplasmic maturation of EBV . Interestingly , the ability to induce vesicle formation was not found in the HSV-1 BFRF1 homolog , pUL34 , in this study . Other studies also showed that the coexpression of pUL34 and pUL31 of pseudorebies virus or ORF67 and ORF69 of Kaposi's sarcoma-associated herpesvirus ( KSHV ) in transfection system is required for the formation of vesicles at the nuclear margins , resembling the primary envelopment without nucleocapsids [41] , [51] . According to our sequence alignment , the functional domains LD1 , LD2 and ID for ESCRT protein recruitment , and the TM domain for INM anchoring of BFRF1 , are relatively well conserved among the other BFRF1 homologs ( Figure S3C ) , suggesting that other BFRF1 homologs may share the abilities of recruiting ESCRT components and targeting to INM to facilitate nucleocapsid egress . Although HCMV pUL50 is more closely related to HSV-1 pUL34 in the phylogenic analysis ( Figure S3D ) , a region of similar length in HCMV pUL50 and that matched the ESR of BFRF1 was identified in the alignment . Whether pUL50 also shares a similar function with BFRF1 remains to be studied . There are some differences in the process of membrane budding of RNA viruses and the nuclear egress of herpesviruses . Most RNA viruses bud from the single lipid-bilayer cytoplasmic membrane [12] , [52] , [53] . In contrast , the nuclear egress of herpesviruses involves passing through double lipid-bilayer membranes and the underlying nuclear lamina network . In this study , we provide evidence that EBV redistributes several ESCRT proteins to perinuclear puncta or clumping structures ( Figure 1C and S1B ) to modify the nuclear membrane . Inhibition of recruited ESCRT machinery caused the obvious accumulation of viral DNA and capsid proteins in the nucleus of cells replicating EBV ( Figure 6 ) , suggesting that the cellular ESCRT machinery is required for the nuclear egress of EBV . The ESCRT machinery was shown to regulate the secondary envelopment of HSV-1 and HCMV virions in the cytoplasmic compartment [7]–[9] , [16] , [54] . It will be interesting to determine whether the ESCRT machinery is also involved in the nuclear egress of alpha- and beta-herpesviruses . In terms of component usage , different herpesviruses show various dependencies on ESCRT components for their maturation . The production of HSV-1 requires multiple Chmps proteins and Vps4 , but not TSG101 and Alix proteins [7] , [8] . Similarly , Chmp1A and Vps4 , but not TSG101 and Alix , are important for HCMV production [9] . In EBV , expression of dominant negative Chmp4b and Vps4 also reduced virion release into the culture supernatant ( Figure 1A ) . In confocal analysis , Alix , Chmp4b , Chmp1b and Vps4 were recruited and partially colocalized with BFRF1 at the nuclear membrane , suggesting that functional Chmp proteins and Vps4 ATPase are commonly used for the maturation of herpesviruses . Different herpesviruses show varying preferences in using class E components . Here we found that BFRF1 alone was sufficient to recruit the class E proteins Alix , Chmp4 and Vps4 ( Figure 4 ) . However , when BFRF1 was co-expressed with BFLF2 , only small amounts of BFRF1 were coimmunoprecipitated by TSG101 antibody ( data not shown ) and this interaction was not seen in BFRF1 expressing cells , suggesting that the cooperation of BFRF1 with other viral proteins may lead to highly-ordered complexes for recruiting encapsidated nucleocapsids to the budding site . Taken together , we show that the cellular ESCRT machinery is recruited by EBV BFRF1 and participates in the scission of the nucleus-associated membrane , a mechanism used for the nuclear egress of EBV . Although the detailed coordination of ESCRT components with viral factors in cells replicating the virus needs to be studied further , this analysis provides not only information for EBV maturation , but also novel insights into the involvement of ESCRT machinery in regulating nuclear envelope architecture .
HeLa cells were derived from human cervical epithelial cells ( ATCC #CCL-2 ) . The EBV negative cell line NPC-TW01 was established from a Taiwanese nasopharyngeal carcinoma [55] and NA is a recombinant Akata EBV converted NPC-TW01 cell line [36] . All cells were cultured in Dulbecco's Modified Eagle's Medium ( HyClone ) and supplemented with 10% fetal calf serum , penicillin ( 100 U/ml ) and streptomycin ( 100 µg/ml ) at 37°C with 5% CO2 . For EBV induction , NA cells were transfected with plasmid pRTS15 expressing Rta [56] , using Lipofectamine ( Invitrogen ) in OptiMEM medium ( GIBCO-BRL ) according to the manufacturer's instruction . For Alix siRNA treatment , Alix-specific siRNA ( 100 nM final ) with 2′-O-methyl ribosyl modifications , which reduce off-target transcript silencing [40] , and a sense sequence 5′-GAAGGAUGCUUUCGAUAAAUU-3′ [8] or control siRNA were synthesized by Dharmacon Research Inc . ( Lafayette , CO ) and transfected twice into slide-cultured cells for 48 , 72 or 96 h . EBV positive NA cells were transfected with vector pSG5 or Rta expressing plasmid pRTS15 accompanied by a plasmid expressing GFP-Chmp4b or Vps4-DN . At 96 h post transfection , the culture supernatants were collected , subjected to centrifugation at 10 , 000× g for 30 min at 4°C and filtered through a 0 . 45 µm nylon membrane ( MCE membrane , Millipore ) to remove cell debris . To harvest the secreted viral particles and extract the viral genome from the virions , the supernatant was incubated with DNase I to eliminate contamination with cellular DNA , as described previously [57] . The viral genome was concentrated using a QIAmp MinElute Virus Spin Kit ( QIAGEN ) according to the manufacturer's instructions . The samples were confirmed for the absence of glyceraldehyde-3-phosphate dehydrogenase DNA to rule out the possibility of DNA contamination from cell debris . The cells in the plates were lysed with DNA extraction buffer ( 10 mM Tris-HCl , pH 8 . 0 , 2 . 5 mM MgCl2 , 1% Tween 20 , 1% NP-40 , 1 mg/ml proteinase K ) at 50°C for at least 4 h . To measure the viral DNA content of the transfected cells or culture supernatants , qPCR using SYBR green I dye ( Invitrogen ) and the iCycler iQ Detection System ( Bio-Rad ) was used to detect the EBV BamHI W fragment and beta microglobulin ( Bm ) gene in the human genome ( modified from [58] ) . Briefly , the amplification of the Bm gene was used to determine the input cellular DNA in serial dilutions ( 104 , 103 , 102 , 101 and 1 cells ) of standard EBV positive H2B4 cells and each sample . A BamHI W calibration curve also was obtained by the amplification of serial dilutions ( in water ) of H2B4 DNA containing 104 , 103 , 102 , 101 and 1 BamHI W copies ( assuming diploid H2B4 cells carry a single copy of the EBV genome ) . A linear calibration curve was then generated by plotting Ct values ( y-axis ) against Log10 BamHI W copy number ( X-axis ) , from which the number of EBV genomes in the individual samples was determined . All standards and samples , together with EBV-positive and EBV negative controls , were analyzed in duplicate . The primers for Bm are 5′-GGTTGGCCAATCTACTCCCAGG-3′ and 5′-GCTCACTCAGTGGCAAAG-3′ . Rta-transfected NA cells with were lysed with radioimmunoprecipitation assay ( RIPA ) buffer [50 mM Tris/HCl , pH 7 . 5 , 150 mM NaCl , 1% Nonidet P-40 , 0 . 5% sodium deoxycholate , 0 . 1% SDS , complete protease inhibitor cocktail ( Roche ) and 1 mM Na3VO4] , disrupted in SDS-sample buffer and displayed by 10% SDS-PAGE for immunoblotting detection . To detect the EBV proteins , the lab-made anti-Zta 4F10 , anti-Rta 467 , anti-BMRF1 88A9 , anti-BGLF4 2616 , anti-BcLF1 L2 , and anti-BLLF1 201 were used as described previously [59] . The anti-BFRF1 and anti-BFLF2 antibodies were kindly provided by Dr . Alberto Faggioni ( Università La Sapienza , Italy ) . For lamin A , GFP tagged proteins , mCherry-Vps4-DN or nucleoporin Nup62 detections , anti-lamin A/C 636 ( Santa Cruz ) , anti-GFP JL-8 ( Clontech ) , anti-Vps4 H-165 ( Santa Cruz ) and FG-repeat specific mAb414 ( Abcam ) were used as instructed . HA-BFRF1 and Flag-BFLF2 were generated by cloning XhoI-HA-BFRF1-NotI or BamHI-Flag-BFLF2-NotI into pcDNA3 . 0 ( Clontech ) and were kindly provided by Dr . Hsiu-Ming Shih ( Academia Sinica , Taiwan ) . GFP-BFRF1 , CFP-BFRF1 and YFP-BFLF2 were also cloned using the same strategy into pEGFP-C1 , pECFP-C1 and pEYFP-C1 ( Clontech ) . pCR3 . 1-GFP–Chmp4b , pCR3 . 1-GFP–Vps4A and pCR3 . 1-Vps4-DN–mCherry [12] , [60] are gifts from Dr . Paul D . Bieniasz ( Rockefeller University , US ) . Cellular organelle marker plasmids , including pEYFP-ER , pEYFP-Golgi and pEYFP-Endo were provided by Dr . King-Song Jeng ( Academia Sinica , Taiwan ) . For HSV-1 UL34 and UL31 expression , HA-UL34 and 3×Flag-UL31 were generated by cloning EcoRV-UL34-XhoI or HindIII-UL31-EcoRI from the strain F HSV-1 ( ATCC #VR-733 ) infected A549 cells DNA extract into pcDNA3 . 0-HA or p3×FLAG-CMV-7 . 1 ( Sigma-Aldrich ) , respectively . Plasmids expressing Flag-Alix , Flag-Alix_Bro , Flag-Alix_V and Flag-Alix_PRR were generated by PCR cloning of Alix gene fragments into pCAGGS/MCS vector [10] . All HA-BFRF1 mutants including HA-BFRF1d ( 8–65 ) , HA-BFRF1d ( 74–134 ) , HA-BFRF1d ( 135–179 ) , HA-BFRF1d ( 180–313 ) and HA-BFRF1d ( 314–336 ) were generated by a single primer based site-directed mutagenesis strategy [61] with pcDNA3 . 0-HA-BFRF1 template and the primers specified in Table S1 . For the detection of HA-BFRF1 , Flag-BFLF2 , BGLF4 , lamin A/C , emerin , TSG101 or Alix , Slide-cultured HeLa cells were transfected with plasmids expressing HA-BFRF1 , Flag-BFLF2 or vector pcDNA3 . 0 . The slides were fixed with 4% paraformaldehyde in PBS at 24 h post transfection at RT for 20 min , washed with PBS ( 145 mM NaCl , 1 . 56 mM Na2HPO4 , 1 mM KH2PO4 , pH 7 . 2 ) and permeabilized with 0 . 1% Triton X-100 at RT for 5 min . The slides were then incubated with anti-HA ( Covance or GeneTex ) , anti-Flag ( Sigma-Aldrich or Viogene ) , rabbit anti-BGLF4 serum , anti-lamin A/C ( Santa Cruz ) , anti-emerin ( Santa Cruz ) , anti-TSG101 ( GeneTex ) , or rabbit anti-Alix serum [62] at 37°C for 1 . 5 h . For the detection of major viral capsid components BcLF1 , BORF1 or BDLF1 , fixed NA cells were permeabilized with 0 . 1% Triton X-100 and incubated with anti-BcLF1 L2 , or rabbit polyclonal anti-BORF1 and anti-BDLF1 antibodies ( which were gifts from Dr . Shih-Tung Liu , Chang Gung University , Taiwan ) , at 37°C for 1 . 5 h . After washing with PBS for 5 min three times , slides were incubated with Rhodamine- , FITC- or AMCA ( amino-methyl-coumarin-acetate ) -conjugated anti-mouse or rabbit Ig antibodies ( CAPPEL ) at 37°C for 1 h . DNA was stained with Hoechst 33258 at RT for 30 sec . The staining patterns were observed under fluorescence or confocal microscopy ( Ziess ) . To observe cell morphology , the margin of cells was detected by MetaMorph software ( Molecular Devices ) and indicated by white outline in some figures . A total of 3×105 HeLa cells were transfected with plasmid expressing HA-BFRF1 or control vector pcDNA3 . 0 for 24 h and processed for TEM analysis . Briefly , the cells were trypsinized , pelleted and washed in 0 . 1 M phosphate buffer ( pH 7 . 4 ) . Pellets were rinsed in 0 . 1 M phosphate buffer and fixed in 4% paraformaldehyde for 30 min at 4°C . Cells were washed and postfixed in 1% osmium tetroxide for 10 min at room temperature . Sample were dehydrated with increasing concentrations of ethanol from 70 to 100% and then infiltrated with propylene oxide for 1 h , propylene oxide: Epon = 1∶1 for 1 h and then pure Epon for 2 h . Sample were embedded by curing at 40°C for 24 h , followed by 60°C for 48 h prior to sectioning for TEM . Embedded samples were cut into 65-nm-thick sections and stained with uranyl acetate and lead citrate . Samples were imaged using a HITACHI H-7100 transmission electron microscopy , and images were acquired using AMT camera system . EBV positive NA or HeLa cells ( 1×107 ) were transfected with plasmids expressing Rta , HA-BFRF1 , Flag-BFLF2 , GFP-Chmp4b , Vps4A-DN or relative vector control pSG5 , pcDNA 3 . 0 ( Invitrogen ) , pEGFP-C1 ( Clontech ) or pDsRed-Mono ( Clontech ) to match equal total DNA amounts . At 24 h post transfection , cells were harvested and disrupted in RIPA buffer . Cell lysates were centrifuged for 10 min at 16 , 000× g , to remove the insoluble fraction . Before immunoprecipitation , the lysate was pre-cleared with 125 µl of 20% protein A-Sepharose beads ( Pharmacia ) for 1 h at 4°C . To immunoprecipitate HA-BFRF1 , Flag-BFLF2 , TSG101 , Alix or GFP-Chmp4b , lysates were incubated with anti-HA ( Covance ) or anti-Flag ( 1 . 5 µg , Sigma-Aldrich ) , anti-TSG101 ( 3 µg , GeneTex ) , anti-GFP antibody ( 2 µg , Clontech ) or rabbit anti-Alix serum ( 1 µl ) at 4°C for 1 h . Protein A-Sepharose beads ( 125 µl at 20% ) were then added to pull down the immunocomplexes with rotation for 1 . 5 h at 4°C . The immunocomplexes were then washed extensively with RIPA lysis buffer and cold PBS , disrupted in SDS-sample buffer and displayed in 10% SDS-PAGE for immunoblotting . Subcellular fractionation was basically performed as described previously [63] . Transfected NA cells ( 5×106 ) were treated with hypotonic buffer ( 10 mM Tris , pH8 . 0 , 60 mM KCl , 1 . 5 mM MgCl2 , 0 . 5% NP-40 , 1 mM PMSF ) on ice for 1 h . After centrifugation at 200× g for 5 min , the supernatant was harvested as the cytosolic fraction . The pelleted nuclear fraction was then washed twice with hypotonic buffer and PBS at 4°C and resuspended separately in DNA extraction buffer ( 1% SDS , 10 mM Tris-HCl , pH7 . 6 , 10 mM EDTA , 400 mM NaCl , 100 µg/mL RNase A , 200 µg/mL proteinase K ) at 55°C overnight or RIPA lysis buffer . For viral genome detection , the viral DNA was further purified by phenol-chloroform extraction and analyzed by qPCR . For protein expression analysis , the total cell lysate , nuclear or cytosolic fraction derived from equal cell numbers was applied to SDS-PAGE for immunoblotting . PARP and α-Tubulin were detected as nuclear and cytosolic markers , respectively . | Herpesviruses are large DNA viruses associated with human and animal diseases . After viral DNA replication , the herpesviral nucleocapsids egress through the nuclear membrane for subsequent cytoplasmic virion maturation . However , the mechanism by which the virus regulates the nuclear membrane and cellular machinery involved in this process remained elusive . The cellular endosomal sorting complex required for transport ( ESCRT ) machinery is known to participate in the biogenesis of multivesicular bodies , cytokinesis and the release of enveloped viruses from cytoplasmic membranes . Here , we show that functional ESCRT machinery is required for the maturation of Epstein-Barr virus ( EBV ) . ESCRT proteins are redistributed close to the nucleus-associated membrane through interaction with the viral BFRF1 protein , leading to vesicle formation and structural changes of the nuclear membrane . Remarkably , inhibition of ESCRT machinery abolishes BFRF1-induced vesicle formation , and leads to the accumulation of viral DNA and capsid proteins in the nucleus . Specific interactions between BFRF1 and Alix are required for BFRF1-derived vesicle formation and crucial for the nuclear egress of EBV . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"viral",
"transmission",
"and",
"infection",
"microbiology",
"host-pathogen",
"interaction",
"viral",
"structure",
"viruslike",
"particles",
"membranes",
"and",
"sorting",
"biology",
"host",
"cells",
"cell",
"biology",
"virology",
"molecular",
"cell",
"biology",
"nucleocapsid"
]
| 2012 | The ESCRT Machinery Is Recruited by the Viral BFRF1 Protein to the Nucleus-Associated Membrane for the Maturation of Epstein-Barr Virus |
The fundamental aim of structural analyses in biophysics is to reveal a mutual relation between a molecule’s dynamic structure and its physiological function . Small-angle X-ray scattering ( SAXS ) is an experimental technique for structural characterization of macromolecules in solution and enables time-resolved analysis of conformational changes under physiological conditions . As such experiments measure spatially averaged low-resolution scattering intensities only , the sparse information obtained is not sufficient to uniquely reconstruct a three-dimensional atomistic model . Here , we integrate the information from SAXS into molecular dynamics simulations using computationally efficient native structure-based models . Dynamically fitting an initial structure towards a scattering intensity , such simulations produce atomistic models in agreement with the target data . In this way , SAXS data can be rapidly interpreted while retaining physico-chemical knowledge and sampling power of the underlying force field . We demonstrate our method’s performance using the example of three protein systems . Simulations are faster than full molecular dynamics approaches by more than two orders of magnitude and consistently achieve comparable accuracy . Computational demands are reduced sufficiently to run the simulations on commodity desktop computers instead of high-performance computing systems . These results underline that scattering-guided structure-based simulations provide a suitable framework for rapid early-stage refinement of structures towards SAXS data with particular focus on minimal computational resources and time .
We investigated structural transitions in three two-state protein systems , where the target structure was initially known .
For the VHP-based polypeptide system ( Fig 1A inset ) , proof-of-concept simulations aimed at both transitions from bent to elongated and elongated to bent conformation . Target scattering data were calculated via the Debye equation ( Eq 1 ) . We analyzed the backbone’s elongatedness by extracting distances between N-terminal and C-terminal Cα atoms for both free and scattering-guided SBM simulations . The distance distributions ( Fig 1A ) show a clear shift towards the target structure’s end-to-end distance . This confirms that , as intended , conformations which are not in accordance with the target curve are avoided in scattering-guided simulations . To show the degree of similarity between typical X-ray scattering patterns from the refinement and the target data , Debye intensities of simulated structures are illustrated in Fig 1B . The curves converge to a certain extent , but do not show perfect agreement . This is due to the fact that the refinement is not only steered by the scattering bias , but also by the physico-geometrical SBM , so that an equilibrium between these two contributions settles in . Computation time scaled as 1 . 4 to 1 for scattering-guided and free SBM simulations . As the Debye summation is an O ( N 2 ) problem ( see Eq 1 ) , the ratio of scattering-guided and free computation times will substantially increase with system size . In this context , rapid evaluation of SAXS profiles from structural models becomes even more important for dynamic refinement procedures such as scattering-guided biomolecular simulations . Time-dependent RMSD curves and bias potential are depicted in Fig 1C and Panel A of S4 Fig for elongated-to-bent and bent-to-elongated transition , respectively . Guiding the simulations towards the target scattering data obviously causes the structural transition to bidirectionally occur back and forth . Minimum target RMSDs are 0 . 15 nm ( Fig 1D ) and 0 . 19 nm ( Panel B in S4 Fig ) , respectively . This implies that structure-opening conformational changes from rather compact to more spacious structures are more difficult to sample than structure-closing ones . Free SBM simulations of bent and elongated conformation yielded average Cα RMSDs of 0 . 17 nm and 0 . 18 nm , respectively . In light of this , scattering-guided SBM simulations were capable of reproducing each target structure with the method’s inherent best possible accuracy . Despite the drastic change in secondary structure , they could model the conformational transition in both directions properly and persistently sample physically reasonable structures near the target conformation . Result parameters are summarized in Table 1 along with the values from analogous explicit-solvent MD simulations . Considering computation times τ 0 . 2 comp , the structure-based method turned out to be faster by two orders of magnitude than the full-MD approach in terms of wall-clock time . Detailed explicit-solvent MD results can be found in S5 and S6 Figs . The bias energy was analyzed as a function of the trajectory’s target RMSD . As displayed in Fig 4A , low bias potential is principally associated with low target RMSD . A Pearson correlation ρ of 0 . 44 indicates that they are in fact positively correlated . However , we find a considerably spread ensemble of distinct structures at equal bias potential levels . With a bias potential of 0 . 88 ε , the minimum target RMSD structure is not exactly in the energetic minimum . Results for the reverse transition are presented in Panels C and D in S4 Fig . Analogous explicit-solvent MD results can be found in Panels C and D in S5 and S6 Figs . To examine the influence of temperature and bias weight , we conducted grid-search variational studies . Results are depicted in Fig 5A and 5B for elongated-to-bent and bent-to-elongated transition , respectively . As soon as the initially increasing V XS av ( k χ ) drops down ( T = 50 , 70 ) or plateaus ( T = 90 , 110 ) , RMSD target min converges towards the average value of related free simulations ( see Table 1 ) . Average χ2 dissimilarity of simulated scattering curves with respect to the target data minimizes accordingly . Near these turnaround points labeled k χ * hereafter , structure-based potential and scattering bias are assumed to be thoroughly balanced . This promotes rapid conformational transitions according to the target data in due consideration of the physico-geometrical model , but prevents the data from being overfitted . In SBMs , the bias potential has to be weighted in such a manner as to introduce a distinct competing minimum to the original single-basin energy funnel . We set k χ ∼ O ( k χ * ) to ensure occurrence of a clear transition , whilst modifying the underlying regular potential as little as possible . The elongated-to-bent transition yielded a smaller k χ * ≈ 1 · 10 - 8 ε ( see Fig 5A ) compared to the bent-to-elongated case with k χ * ≈ 5 · 10 - 8 ε ( see Fig 5B ) . This behavior confirms our previous finding of structure-closing transitions to be favored over structure-opening ones . For both transitions , we find the effect of gradually increasing the coupling constant to be less pronounced at higher temperatures ( Fig 5 ) . It is conspicuous that , independent of kχ , all simulations at T = 110 could—at least temporarily—sample conformations near the target . The increased thermal energy allows to overcome potential barriers in the energy landscape , resulting in greater protein flexibility and sampling power . However , these thermal structural fluctuations by itself are isotropic in conformational space and not directed towards any particular conformation as is the case with a major scattering bias . In contrast , at lower temperatures T = 50 and 70 , RMSD target av values almost double with kχ increasing up to the order of 10−8 ε , before they significantly decline as well . With less thermal energy being available and an increased coupling of simulations to scattering intensities attaching more relative importance to structural information from SAXS , this is in accordance with the expectations . Remember that the global change in orientation of secondary structure elements with respect to each other substantially affects the polypeptide’s overall shape . As a consequence , this system required large temperature ( and bias weight ) to ensure sufficient global conformational flexibility and stably reach a conformation near the target . Though a basic trend should be maintained , these findings cannot be directly translated to other protein systems . The optimal combination of temperature T and bias weight kχ depends on the individual system and should be determined by grid search or other systematic parameter optimization methods . In SBMs , the overall contact and dihedral energy is set equal to the number of atoms in the system [49] . This choice yields folding temperatures near 1 in the structure-based reduced units , corresponding to approximately 120 reduced GROMACS temperature units , and ensures a consistent parameterization . Thus , model-inherent absolute energies are highly system-specific and not comparable among different systems . Not only differ biomolecular systems in general and thus their respective absolute energies , but also the nature of their individual conformational transitions each associated with a specific energy barrier of different ( unknown ) height . Due to the high diversity among biomolecular systems , different systems require different bias weights and temperatures to suitably impact the underlying structure-based potential and provide sufficient thermal energy to induce or accelerate the conformational transition of interest . These parameters are not transferable and have to be determined separately for each system . Modeling the large-scale structural transition between open and closed conformation based on artificial difference data gives a theoretically constructed test example of a real protein movement . Simulations started from open and closed state and aimed at closed and open state , respectively . We computed artificial target difference data ( Fig 2B ) using the Debye equation on amino-acid level . RMSD and bias energy curves of open-to-closed and closed-to-open transition are shown in Fig 2C and Panel A in S7 Fig , respectively , illustrating how the structural similarity to initial and target state develops over the course of the simulations . Both refinements showed one clear transition from initial to target conformation in form of an immediate intercept of RMSD curves . VXS instantaneously minimized accordingly . Subsequently , the target RMSD curves proceeded near respective average free RMSD values . Best structures as measured by target RMSD and bias energy are shown in Fig 2D and Panel B in S7 Fig , respectively . Minimum target RMSD is 0 . 14 nm for both directions of the conformational transition . Detailed results of analogous explicit-solvent MD simulations can be found in S8 and S9 Figs . All result parameters are summarized in Table 2 . Considering computation times τ 0 . 2 comp , structure-based refinements turned out to be faster by almost two orders of magnitude than the full-MD approach , while yielding more accurate structures in terms of minimum target RMSD . We analyzed the bias energy as a function of target RMSD ( Fig 6A and Panel C in S7 Fig ) , which revealed positive Pearson correlations throughout . As a result of the almost instantaneous structural transitions , numerous similar conformations with small bias energy and target RMSD less than 0 . 2 nm do exist , yielding a dense cluster of fluctuating points ( RMSDtarget , VXS ) in this area . This behavior disrupts a potential linear relationship between these quantities as assumed in the Pearson correlation analysis , causing rather small but certainly positive values for ρ . We conducted grid-search variational studies for both open-to-closed ( see Fig 7A ) and closed-to-open ( see Fig 7B ) transition . As indicated by the regions of undefined bias potential in Fig 7 , simulations using a bias weight kχ greater than 10−8 ε apparently blew up . Depending on χ2 , an immoderate bias weight may produce a very large bias potential . This generates an unacceptably large force , which eventually results in a failure of the integrator . For both directions of the conformational transition , the turnaround bias weight k χ * is in the order of 10−10 ε . At this point , the bias potential clearly exhibits its global minimum and average χ2 dissimilarity significantly drops down accordingly . In contrast to the VHP polypeptide , lower temperatures were sufficient to stably sample conformations near the target structure . This is due to the fact that the structural transition of AKE does not induce a drastic overall change in its molecular shape . Upon binding lysine , LAO protein ( Fig 3A ) experiences major structural change [56] . Modeling this domain motion based on artificial difference data gives another test case of a real protein movement . Starting from the crystal structure of the unliganded holo state , these simulations aimed at the unbound apo state and vice versa . Reference and target scattering were calculated from the crystal structures with CRYSOL [57] and thus implicitly include hydration shell contributions . We generated artificial difference data ( Fig 3B ) by subtracting the initial solution scattering from the target solution scattering . Time-dependent initial and target RMSD as well as bias potential are shown in Fig 3C and Panel A in S10 Fig for structure-based holo-to-apo and apo-to-holo simulations , respectively . Biasing simulations towards theoretical difference data resulted in the transitions to readily occur . The bias potential minimized almost instantaneously according to the trajectory’s convergence towards the target state . The final target RMSD of approx . 0 . 2 nm was consistent with corresponding free simulations . For both directions of the conformational transition , best structures exhibit a minimum target RMSD of 0 . 09 nm ( Fig 3D and Panel B in S10 Fig ) . Structure-based refinements were capable of producing structures in full agreement with the target state . Provided equal computing resources , they required only a small fraction of computing time by comparison with analogous explicit-solvent MD simulations . Detailed full-MD results can be found in S11 and S12 Figs . Considering the holo-to-apo transition , the explicit-solvent refinement ( tsim = 10 ns ) lasted for 4 d 15 h 5 min 35 s , whereas the SBM run ( tsim = 2000 arb ) spanned 4 h 11 min 55 s . According to computation times τ 0 . 2 comp related to the trajectory approaching a state with a target RMSD less than 0 . 2 nm , the SBM proved to be up to ten times faster in terms of wall-clock time . All result parameters are summarized in Table 3 . As in the other test systems , bias energy and target RMSD exhibit positive Pearson correlations throughout . According to Fig 8A and Panel C in S11 Fig , the structural diversity at equal bias potential levels is similar for SBM and explicit-solvent MD . Though the best structure cannot definitely be identified from a trajectory on the basis of VXS on its own , the bias potential can serve as a primary indicator for a simulation’s current state and eventual success or failure . Grid-search variational studies for both holo-to-apo and apo-to-holo transition revealed a similar behavior as for AKE . As highlighted by the regions of undefined bias potential in Fig 9 , simulations applying a bias weight kχ greater than 6 ⋅ 10−8 ε failed due to excessively large scattering-related forces . For both directions of the structural transition , the turnaround bias weight k χ * is approx . 10−10 ε . Average χ2 dissimilarity clearly minimizes here ( Fig 9A and 9B , bottom ) , whereas the bias potential does not have a distinct minimum as is the case for AKE but starts to monotonically increase as a function of kχ ( Fig 9A and 9B , middle ) . Again , the evolution of minimum target RMSD indicates lower temperatures to be sufficient to stably reach the target conformation ( Fig 9A and 9B , top ) . This is due to the fact that the conformational transition corresponds to a relative movement of subdomains in the structure so that the molecular shape does not experience a drastic overall change . To assess the method’s robustness towards errors in the scattering data , we conducted a structure-based refinement of LAO protein’s holo-to-apo transition towards noisy artificial difference data . Theoretical absolute scattering curves of reference and target structure were blurred according to a random Gaussian noise . For each q point , mean and standard deviation were modeled as the related clean intensity value and its square root , respectively . Details are described in S3 Appendix . We calculated noisy difference data by subtracting the blurred reference intensity from the blurred target intensity ( Fig 10 ) . Using usual error propagation , errors were calculated as the sum of the Gaussians’ absolute standard deviations and used to individually weight the q points in the simulation . We applied the same parameters as in the refinement towards clean data . Although the bias potential levels off at a considerably higher value , which is to be expected , the simulation could produce equal-quality structures and thus proved to be robust against errors , at least for the level of noise assumed here ( see S13 Fig ) . Note once more that scattering-guided SBM simulations dispense with computationally expensive solvent effects . In view of these results , we did not find a need for explicit solvation in refinement simulations comprising small-angle scattering data up to a maximum momentum transfer of 5 nm−1 . The fact that SBM simulations coupled to small-angle difference scattering data could reproduce each target state with high accuracy indicates that such curves hold sufficient information to guide the simulation towards the correct conformation , at least for the systems studied here . Regardless of their reduced level of complexity by comparison with explicit-solvent MD , scattering-guided SBM simulations produced equal-quality results in a small fraction of computing time . A fundamental paradigm in protein biophysics is the interdependency of macromolecular structure and function . In light of this , small-angle X-ray scattering has significantly gained in importance , especially for structural analyses of dissolved macromolecules . Accurate interpretation of resulting scattering intensities in terms of atomistic models is still a challenging task . By incorporating information from SAXS into structure-based models , we aimed at efficiently interpreting scattering data within computational simulations . Studying three different test systems , we have proven our method to be capable of effectively probing real protein transitions , based only on low-resolution scattering data . Giving results equivalent to those from analogous full-MD methods [20] , scattering-guided SBM simulations could expedite interpretation of intensities from biological SAXS by about two orders of magnitude . Such simulations benefit from extensive sampling as a result of their intrinsically accelerated dynamics . They could rapidly generate structural ensembles in accordance with the input data and provide a valuable alternative for efficient refinement of atomistic structures against SAXS data . Thus , they are particularly suitable for initial high-throughput analyses and can easily perform on usual commodity hardware . If desired , the resultant structure can still be given a final polish within a regular MD force field . As a result of technical advances in light sources and detectors , the wide-angle regime encoding local structural fluctuations has become increasingly accessible in the experiment . So as to level up with experimental resolution , increasingly fine-grained modeling may then be indicated at the cost of leaping computational demands [5 , 21] . Finally , it is important to note that some systems cannot be analyzed straightforward using SAXS . In all test systems , structural transitions could be modeled by a collective movement along one effective degree of freedom , which influences the protein’s shape and thus the difference curve at q ≤ 2 nm−1 crucially . As a consequence , structural fits were unique . However , at higher q values , multiple candidate structures can generate interfering features in the difference profiles . For example , the structural change in the cytoplasmic portion of a sensor histidine kinase protein ( PDB code 2C2A [58] ) induces a Cα RMSD shift of 1 . 25 nm . This conformational transition can effectively be described as a rotation of one subdomain around a helix bundle , but influences the overall molecular shape only marginally . Despite a substantial decrease in bias potential , refinements towards theoretical difference data did not converge to the target structure . This implies that structures exist , which adequately reproduce the difference data , but are not compatible with structural models obtained from crystallographic methods . These findings are in accordance with results presented in Ref [20] and due to a lack of information in the low-resolution experimental data , resulting in unaccomplishably high demands on the theoretical model . Having said this , the protocol for interpretation of SAXS data within SBMs established in this work can serve as a suitable starting point for further developments . These include e . g . expanding single-basin SBMs to multi-Gō models with several minima and testing other functional forms of the bias potential . Furthermore , we intend to directly interface the structure-based refinement framework with parameter optimization methods such as Bayesian inference . In addition , we see several possibilities to extend our hybrid framework to additionally account for information derived from other experimental techniques than SAXS . We plan to extend the framework by considering co-evolutionary contact information from biomolecular sequence data [36] , distance and angle information from NMR spectroscopy , and cryo-EM density maps [6] . Whereas co-evolutionary information can be considered by additional potential terms similar to usual SBM native contacts , NMR distance and angle information can be accounted for by implementing suitable spatial restraints . Provided cryo-EM data , another energetic term can be introduced to bias the structure towards the electron density map based on a spatial overlap . Performing simulations with such a hybrid structure-based/biased/restraint force field , the system can relax into configurations that are consistent with all these contributions .
As a starting point , all-atom SBMs were constructed from the considered system’s initial structure with eSBMTools [30] to obtain suitable coordinate and topology files . Debye scattering terms are encoded as a special type of bonded interaction in the topology file [20] . Scattering topology as well as related extended coordinate file were constructed with gmx genrestr . This command creates half a matrix of virtual-site type-3 pairs , i . e . Debye terms , for the input coordinate file . Amino-acid scatterers centered on virtual interaction sites at the respective residue’s center of mass were used . All residues were considered . The resulting topology include file was added to the system’s topology directly after the atoms section . The corresponding atom type ‘MW’ was manually appended to the atom types table . If Debye scattering data were used as a target , the initial scattering was calculated using gmx waxsdebye . Suitably adjusted run parameters for the SBM refinement are listed in S4 Appendix . Temperatures T and bias weights kχ were set as described in Results and discussion . Finally , SBMs were preprocessed with gmx grompp and run with gmx mdrun . The set-up of explicit-solvent MD simulations followed the common steps of adding hydrogen atoms , choosing potential and water model , neutralizing electric charge by adding an appropriate number of ions , minimizing energy , and equilibrating temperature and pressure . We used the CHARMM27 force field [60] , TIP3P water model [61] , Verlet cut-off scheme , and a constant temperature of 300 K . Electrostatics were treated with the Particle Mesh Ewald method . Parrinello-Rahman pressure coupling and V-rescale temperature coupling were applied . To obtain coordinate and topology file , initial models were preprocessed and protonated with gmx pdb2gmx . A periodic cubic box exceeding twice the longest inter-protein distance was constructed with gmx editconf . The structure was initially energy-minimized using the GROMACS preprocessor gmx grompp and simulation command gmx mdrun . After solvation and electric-charge neutralization , the structure was energy-minimized again . Subsequently , systems were equilibrated in the canonical and isothermal-isobaric ensemble until temperature and pressure converged . Non-hydrogen atoms were position-restrained to their initial positions . A half-matrix of Debye terms was constructed with gmx genrestr for the NPT-equilibrated structure , including all residues and using amino-acid scatterers . This created the scattering topology , which was manually included into the system’s topology . The initial reference scattering was generated with gmx waxsdebye . After preprocessing with gmx grompp , the scattering-guided MD simulation was performed using the gmx mdrun command . Results are shown for coupling strengths kχ optimized via grid-search variational studies comprising 16 simulations in total for each system . | Proteins are the molecular nanomachines in biological cells and thus vital to any known form of life . From the evolutionary perspective , viable protein structure emerges on the basis of a ‘form-follows-function’ principle . A protein’s designated function is inextricably linked to dynamic conformational changes , which can be observed by small-angle X-ray scattering . Intensities from SAXS contain low-resolution information on the protein’s shape at different steps of its functional cycle . We are interested in directly getting an atomistic model of this encoded structure . One powerful approach is to include the experimental data into computational simulations of the protein’s function-related physical motions . We combine scattering intensities with coarse-grained native structure-based models . These models are computationally highly efficient yet describe the system’s dynamics realistically . Here , we present our method for rapid interpretation of scattering intensities from SAXS to derive structural models , using minimal computational resources and time . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
]
| [
"molecular",
"dynamics",
"nmr",
"spectroscopy",
"simulation",
"and",
"modeling",
"protein",
"structure",
"thermodynamics",
"research",
"and",
"analysis",
"methods",
"proteins",
"chemistry",
"biophysics",
"molecular",
"biology",
"free",
"energy",
"physics",
"biochemistry",
"biochemical",
"simulations",
"scattering",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"chemistry",
"small-angle",
"scattering",
"computational",
"biology",
"spectrum",
"analysis",
"techniques",
"macromolecular",
"structure",
"analysis",
"biophysical",
"simulations"
]
| 2019 | Rapid interpretation of small-angle X-ray scattering data |
The explosive spread of Zika virus ( ZIKV ) and associated complications in flavivirus-endemic regions underscore the need for sensitive and specific serodiagnostic tests to distinguish ZIKV , dengue virus ( DENV ) and other flavivirus infections . Compared with traditional envelope protein-based assays , several nonstructural protein 1 ( NS1 ) -based assays showed improved specificity , however , none can detect and discriminate three flaviviruses in a single assay . Moreover , secondary DENV infection and ZIKV infection with previous DENV infection , both common in endemic regions , cannot be discriminated . In this study , we developed a high-throughput and multiplex IgG microsphere immunoassay ( MIA ) using the NS1 proteins of DENV1-DENV4 , ZIKV and West Nile virus ( WNV ) to test samples from reverse-transcription-polymerase-chain reaction-confirmed cases , including primary DENV1 , DENV2 , DENV3 , WNV and ZIKV infections , secondary DENV infection , and ZIKV infection with previous DENV infection . Combination of four DENV NS1 IgG MIAs revealed a sensitivity of 94 . 3% and specificity of 97 . 2% to detect DENV infection . The ZIKV and WNV NS1 IgG MIAs had a sensitivity/specificity of 100%/87 . 9% and 86 . 1%/78 . 4% , respectively . A positive correlation was found between the readouts of enzyme-linked immunosorbent assay and MIA for different NS1 tested . Based on the ratio of relative median fluorescence intensity of ZIKV NS1 to DENV1 NS1 , the IgG MIA can distinguish ZIKV infection with previous DENV infection and secondary DENV infection with a sensitivity of 88 . 9–90 . 0% and specificity of 91 . 7–100 . 0% . The multiplex and high-throughput assay could be applied to serodiagnosis and serosurveillance of DENV , ZIKV and WNV infections in endemic regions .
Despite a marked decrease of Zika virus ( ZIKV ) infection since late 2017 , the specter of congenital Zika syndrome ( CZS ) and its re-emergence in flavivirus-endemic regions highlight the need for sensitive and specific diagnostic tests [1–4] . Similar to the laboratory diagnosis for other flaviviruses , detection of nucleic acid as soon as possible post-symptom onset ( PSO ) is considered as the gold standard to confirm ZIKV infection , [5 , 6] . Since many individuals test for ZIKV infection beyond the period when RNA is detectable and most ( ~80% ) of ZIKV infections are asymptomatic , serological tests remain as a key component of ZIKV confirmation [5 , 6] . Furthermore , ZIKV can be transmitted sexually or following asymptomatic infection [7–9] . ZIKV is a member of the genus Flavivirus of the family Flaviviridae , which includes several pathogenic mosquito-borne viruses in different serocomplexes . The four serotypes of dengue virus ( DENV ) belong to the DENV serocomplex; West Nile virus ( WNV ) and Japanese encephalitis virus ( JEV ) to the JEV serocomplex; yellow fever virus ( YFV ) as a single member; and ZIKV10 . Given that the envelope ( E ) protein is the major target of antibody response after flavivirus infection , different E antigens such as recombinant E protein , inactivated virions or virus-like particles have been developed for serological tests [10–13] . Due to the presence of several highly conserved residues of flavivirus E proteins , anti-E antibodies in serum are commonly cross-reactive to different flaviviruses [13–17] . The guidelines of Centers for Disease Control and Prevention ( CDC ) recommend that positive or equivocal results of E protein-based IgM tests require further testing with time-consuming plaque reduction neutralization tests ( PRNT ) [5 , 6] . However , PRNT can confirm ZIKV-infected individuals who acquire ZIKV as the first flavivirus infection , known as primary ZIKV ( pZIKV ) infection , but often can only be interpreted as unspecified flavivirus infections for those who have experienced previous DENV or other flavivirus infections , limiting its application for ZIKV serodiagnosis in flavivirus-endemic regions . When 795 sera that were IgM positive for ZIKV antigen by ELISA were tested for flavivirus neutralizing antibodies by PRNT , 45% were positive for ZIKV and at least one other flavivirus [18] . This non-specificity may be an inherent property of the early post-infection response to ZIKV or reflect prior flavivirus experience . A large number of Americans ( 7 million ) have experienced a WNV infection since 1999 [19] and ~8 million traveled to yellow fever endemic countries in 2015 [20 , 21] . Thus , a sensitive , specific and multiplex serological test that can distinguish ZIKV and other flavivirus infections is needed in both U . S . and flavivirus-endemic countries [18] . Moreover , several studies have shown that anti-DENV or WNV antibodies can enhance ZIKV infection in vitro [22–26] and in small animals , in which administration of DENV-immune plasma resulted in increased viremia and mortality in stat2 knock out mice [27] . This is known as antibody-dependent enhancement , in which antibody at suboptimal concentration for neutralization can enhance DENV , ZIKV or other flavivirus entry and replication in Fcγ receptor-bearing cells such as monocytes and is believed to contribute to disease pathogenesis [28] . Despite ADE of ZIKV by previous DENV immunity was not supported by two studies in non-human primates [29 , 30] , more in-depth studies of DENV immunity on ZIKV disease outcome and complication in humans are warranted [31–33] . Thus , serological tests that can distinguish pZIKV infection ( p = primary ) from ZIKV infection with previous DENV ( ZIKVwprDENV , wpr = with previous ) infection are crucial to understand the pathogenesis of ZIKV and CZS in regions where ZIKV and DENV co-circulate . Compared with traditional E protein-based assays , several enzyme-linked immunosorbent assays ( ELISAs ) based on ZIKV nonstructural protein 1 ( NS1 ) , including a recently reported blockade of binding ELISA , have shown improved specificity [34–39] . However , secondary DENV ( sDENV ) and ZIKVwprDENV infections , of which both were common in endemic regions , cannot be discriminated [34–39] . Moreover , none can detect and distinguish ZIKV , DENV and other flavivirus in a single assay . With its high-throughput and multiplex ( up to 100-plex ) capacity , microsphere immunoassay ( MIA ) has been employed in the detection of cytokines , transplantation and transfusion antigens , and various bacterial and viral pathogens [40–43] . Previously , we reported that a combination of ELISAs based on the NS1 proteins of DENV and ZIKV can distinguish various DENV and ZIKV infections [44 , 45] . In this study , we developed a high-throughput and multiplex IgG MIA using NS1 proteins of DENV1 to DENV4 , ZIKV and WNV , and showed that the NS1 IgG MIA can detect and distinguish not only primary DENV , ZIKV and WNV infections but also sDENV and ZIKVwprDENV infections .
The Institutional Review Boards ( IRB ) of the University of Hawaii approved this study ( CHS #17568 , CHS#23786 ) . S1 Table summarizes the numbers , serotypes , sampling time and sources of different panels of serum or plasma samples , including those from primary DENV1 ( pDENV1 ) , primary DENV2 ( pDENV2 ) , primary DENV3 ( pDENV3 ) , primary WNV ( pWNV ) , pZIKV , sDENV and ZIKVwprDENV infections as well as flavivirus-naïve individuals . Samples collected <3 months or ≥3 months PSO were designated as convalescent- or post-convalescent-phase samples , respectively . Samples from reverse transcription-PCR ( RT-PCR ) confirmed Zika cases were from the Pediatric Dengue Cohort Study ( PDCS ) and the Pediatric Dengue Hospital-based Study in Managua , Nicaragua between July 2016 and March 2017 [46 , 47] . The Zika cases that were DENV-naïve or previously DENV-exposed were defined as pZIKV ( p = primary ) or ZIKVwprDENV ( wpr = with previous ) panels , respectively . The DENV-immune status was based on anti-DENV antibody testing by an inhibition ELISA at entry and annually of the PDCS [44–47] . Parents or legal guardians of all participants provided written informed consents , and participants ≥6-year old provided assents . These studies were approved by the IRBs of the University of California , Berkeley , and Nicaraguan Ministry of Health . Thirty-six plasma samples from blood donors , who were tested WNV-positive by the transcription-mediated amplification ( a sensitive nucleic acid detection method used in blood bank ) , IgM and IgG antibodies between 2006 and 2015 , designated as pWNV infection , were provided by the American Red Cross at Gaithersburg , Maryland [48] . Pre-2015-16 ZIKV epidemic convalescent- and post-convalescent-phase samples from RT-PCR confirmed cases with different primary DENV infections ( pDENV1 , pDENV2 , and pDENV3 ) or sDENV infection were from Taiwan , Hawaii and Nicaragua; 53 flavivirus-naïve samples from a seroprevalence study in Taiwan were included as control in this study [44 , 45 , 49–52] . Samples from cases with primary DENV4 infection were not available . Primary DENV or sDENV infection was determined by IgM/IgG ratio or focus-reduction neutralization tests as described previously [49–51] . The NS1 gene ( corresponding to amino acid residues 1–352 ) of ZIKV ( HPF2013 strain ) with a His-tag at the C-terminus was codon-optimized ( Integrated DNA Technologies , Skokie , IL ) and cloned into pMT-Bip vector to establish a Drosophila S2-cell stable clone [44] . ZIKV-NS1 protein from supernatants of the stable clone was purified by fast purification chromatography system ( AKTA Pure , GE Health Care Bio-Science , Pittsburg , PA ) [44] . Purified DENV1-4 and WNV NS1 proteins were purchased from The Native Antigen ( Oxford , UK ) . Ten μg each of the 6 purified NS1 proteins , bovine serum albumin ( BSA ) and PBS ( as negative antigen control ) were coupled individually onto 8 types of magnetic carboxylated miscrosphere beads ( 1 . 25 X 106 each ) containing different fluorophores ( MagPlexTM-C ) ( Luminex , TX , Austin ) using two-step carbodiimide process at room temperature [53 , 54] . The antigen-conjugated microspheres were stored in 250 uL PBN buffer ( PBS with 1% BSA and 0 . 05% sodium azide , Sigma Aldrich ) at 4°C until use . Eight types of microsphere beads coupled with different NS1 proteins , BSA or PBS were combined and diluted in PBS-1% BSA . Fifty μL of the mixture ( containing ~1250 beads of each type ) were added to each well of a flat-bottom 96-well plate , and incubated with 50 μL diluted serum or plasma ( 1:100 dilution in PBS-1% BSA ) at 37°C for 30 min in the dark , followed by wash with 200 μL of PBS-1% BSA twice , incubation with 50 μL of red phycoerythrin-conjugated anti-human or anti-mouse IgG ( Jackson Immune Research Laboratory , West Grove , PA ) at 37°C for 45 min in the dark , and wash with 200 μl of PBS-1% BSA twice [54] . Microspheres were then resuspended in 100 μl of PBS-1% BSA , incubated for 5 min and read by Luminex 200 machine ( Austin , TX ) . All incubations were performed on a plate shaker at 700 rpm and all wash steps used a 96-well magnetic plate separator ( Millipore Corp . , Billerica , MA ) [54] . Each plate includes two positive controls ( confirmed-ZIKV or DENV infection ) , four negative controls ( flavivirus-naïve samples ) , samples , and mouse anti-His mAb ( all in duplicates ) . The median fluorescence intensity ( MFI ) was determined for 100 microspheres for each well . The MFI values for each antigen were divided by the mean MFI value of one positive control ( MFI~104 ) and multiplied by 104 to calculate to rMFI for comparison between plates ( S1 Fig ) . The cutoff rMFI for each antigen was defined by the mean rMFI value of 19 flavivirus-naïve samples plus 5 standard deviations , which gave a confidence level higher than 99 . 9% from 4 negatives [55] . Each MIA was performed twice ( each in duplicate ) . New batch of conjugated antigens was tested with flavivirus-naïve panel to determine the cutoff rMFI . DENV1- , DENV2- , DENV3- , and ZIKV-NS1 IgG ELISAs have been described previously [44 , 45] . Briefly , purified NS1 proteins ( 16 ng for individual NS1 protein per well ) were coated on 96-well plates at 4°C overnight , followed by blocking ( StartingBlock blocking buffer , Thermo Scientific , Waltham , MA ) , incubation with primary antibody ( serum or plasma at 1:400 dilution ) and secondary antibody ( anti-human IgG conjugated with horseradish peroxidase , Jackson Immune Research Laboratory , West Grove , PA ) , and wash [44 , 45] . After adding tetramethylbenzidine substrate ( Thermo Scientific , Waltham , MA ) followed by stop solution , the optical density ( OD ) at 450 nm was read with a reference wavelength of 630 nm . Each ELISA plate included two positive controls ( confirmed-ZIKV or DENV infection ) , four negative controls ( flavivirus-naïve sample ) , and samples ( all in duplicate ) . The OD values were divided by the mean OD value of one positive control ( OD close to 1 ) in the same plate to calculate the relative OD ( rOD ) values for comparison between plates [44 , 45] . The cutoff rOD was defined by the mean rOD value of negatives plus 12 standard deviations , which gave a confidence level of 99 . 9% from 4 negatives [55] . Each ELISA was performed twice ( each in duplicate ) . Two-tailed Mann-Whitney test was used to determine the P values between two groups , the two-tailed Spearman correlation test the relationship between the rOD and rMFI values , and the receiver-operating characteristics ( ROC ) analysis the cutoffs of the rMFI and rOD ratios ( GraphPad Prism 6 ) . The 95% confidence interval ( CI ) was calculated by Excel .
We first employed the multiplex NS1 IgG MIA to test samples from primary DENV ( pDENV1 , pDENV2 and pDENV3 ) , pZIKV and pWNV infection panels . Compared with flavivirus-naïve panel , the pDENV1 panel recognized the NS1 proteins of DENV1 ( 100% ) and other DENV serotypes ( 33 . 3 to 61 . 9% ) , but not those of different serocomplexes ( ZIKV and WNV NS1 proteins ) ( Fig 1A and 1B ) . Similarly , the pDENV2 and pDENV3 panels recognized the NS1 protein of the homologous serotype ( DENV2 , DENV3 ) better than those of other serotypes ( Fig 1C and 1D ) , but did not recognize ZIKV or WNV NS1 protein except two samples ( recognizing WNV , 2/13 ) . The pZIKV panel recognized ZIKV NS1 protein but not those of WNV and DENV except two sample recognizing DENV2 ( 2/38 ) , whereas the pWNV panel recognized WNV proteins rather than those of ZIKV and DENV except one sample ( recognizing DENV4 , 1/36 ) ( Fig 1E and 1F ) . Taken together , these findings suggested that primary infection panels recognized the homologous ( infecting serotype ) NS1 protein better than other NS proteins within the same serocomplex , and in general did not recognize an NS protein of different serocomplexes ( Fig 1G ) . We next tested samples from sDENV and ZIKVwprDENV panels . For convalescent-phase samples , sDENV panel not only recognized NS1 proteins of DENV1-4 ( 66 . 7 to 100% ) but also those of ZIKV and WNV ( 45 . 8 to 54 . 2% ) ( Fig 2A ) . The ZIKVwprDENV panel recognized ZIKV NS1 protein ( 100% ) as well as DENV1-4 and WNV NS1 proteins ( 60 . 0 to 90 . 0% ) ( Fig 2B ) . A similar trend was observed for post-convalescent-phase samples ( Fig 2C and 2D ) . These findings were in agreement with our previous reports based on NS1 IgG ELISAs [44 , 45] , and suggested that after repeated flavivirus infections , such as sDENV and ZIKVwprDENV infections , anti-NS1 antibodies cross-reacted to multiple NS1 proteins , including those from prior exposure or sometimes those with no prior exposure . Previously we reported that sDENV panel not only recognized DENV1 NS1 protein but also ZIKV NS1 protein in IgG ELISA ( 95 . 8 and 66 . 7% , respectively ) ; similarly the ZIKVwprDENV panel recognized both ZIKV and DENV1 NS1 proteins ( 95 . 0 and 85 . 0% , respectively ) [44] . Using the rOD ratio of ZIKV NS1 to DENV1 NS1 with a cutoff at 0 . 24 , we can distinguish ZIKVwprDENV and sDENV panels . Since the same sDENV and ZIKVwprDENV panels recognized both DENV1 and ZIKV NS1 proteins in IgG MIA ( Fig 2A and 2B ) , we calculated the ratio of relative median fluorescence intensity ( rMFI ) of ZIKV NS1 to that of DENV1 NS1 and found that a cutoff of the rMFI ratio at 0 . 62 , as determined by ROC analysis , can distinguish these two panels with a sensitivity of 88 . 9% and specificity of 91 . 7% ( Fig 2E ) . Since both panels also recognized DENV2 NS1 protein , we further calculated the ratio of rMFI of ZIKV NS1 to DENV2 NS1; interestingly a cutoff of the rMFI ratio at 0 . 62 was able to distinguish these two panels with a sensitivity of 94 . 4% and specificity of 90 . 9% ( Fig 2F ) . Similar observations were found for post-convalescent-phase sDENV and ZIKVwprDENV panels; these two panels can be distinguished by a cutoff ( 0 . 62 ) of the rMFI ratio for ZIKV NS1 to DENV1 NS1 or DENV2 NS1 with a sensitivity/specificity of 90 . 0/100% or 83 . 3/100% , respectively ( Fig 2G and 2H ) . Since these panels have been tested with individual DENV1 to DENV4 and ZIKV NS1 IgG ELISAs previously [45] , we compared the detection rates for each NS1 protein between ELISA and MIA . For the pZIKV panel , ZIKV NS1 ELISA had a detection rate of 100% , comparable to that of MIA , for the post-convalescent-phase samples , but only 5% for the convalescent-phase samples , which was much lower than that of MIA ( 100% ) ( Fig 3A and 3B ) . Although 19 convalescent-phase pZIKV samples were tested negative by ZIKV NS1 IgG ELISA , the relative optical density ( rOD ) values were positively correlated with the rMFI values ( correlation coefficient r = 7464 , P = 0 . 0002 ) ( Fig 3C ) , suggesting that ZIKV NS1 MIA was more sensitive than ELISA . A positive correlation was also found between rOD and rMFI values for the post-convalescent-phase samples ( r = 8922 , P<0 . 0001 ) ( Fig 3D ) . For pDENV1 panel , DENV1 NS1 ELISA and MIA had comparable detection rates ( 100% ) for both convalescent and post-convalescent-phase samples ( Fig 3E and 3F ) . Similarly , a positive correlation was found between rOD and rMFI values ( Fig 3G and 3H ) . For ZIKVwprDENV panels , ZIKV NS1 IgG ELISA and MIA had comparable detection rates for both convalescent and post-convalescent-phase samples ( Fig 4A and 4B ) . A positive correlation was found between rOD and rMFI values for ZIKV NS1 as well as DENV1 , DENV2 , DENV3 and DENV4 NS1 tested ( Fig 4C–4E ) . Similar observations were found for sDENV panels ( S2 Fig ) . Table 1 summarizes the results of all samples tested with different NS1 proteins ( DENV1 , DENV2 , DENV3 , DENV4 , DENV1 , 2 , 3 or 4 , ZIKV and WNV ) in the IgG MIA . For statistical analysis comparing different panels , one sample from each participant was included ( S2 Table ) . The overall sensitivity of each DENV ( DENV1 , DENV2 , DENV3 ) NS1 IgG MIA to detect different DENV infections ranged from 73 . 6 to 90 . 1% and specificity from 98 . 1 to 100% ( Table 2 ) . Interestingly , combination of four DENV NS1 IgG MIA increased the sensitivity to 94 . 5% , while maintaining the specificity of 97 . 2% , suggesting that this multiplex assay can be applied to detect DENV infections rather than distinguish different DENV serotypes . For the ZIKV NS1 IgG MIA , the overall sensitivity was 100% and specificity 87 . 9% . For the WNV NS1 IgG MIA , the overall sensitivity was 86 . 1% and specificity 78 . 4% ( Table 2 ) .
In this study , we developed a high-throughput and multiplex IgG MIA using NS1 proteins of DENV1 to DENV4 , ZIKV and WNV to detect and distinguish various DENV , ZIKV and WNV infections . Based on the results , we propose an algorithm to discriminate primary DENV , pZIKV and pWNV infections , sDENV infection and ZIKVwprDENV infection ( Fig 5 ) . Previous studies of flavivirus serodiagnosis mainly focused on two flaviviruses . Compared with a recent study of IgG MIA containing ZIKV and DENV antigens , our multiplex IgG MIA consists of 6 antigens ( DENV1 to DENV4 , WNV and ZIKV NS1 proteins ) plus two controls ( BSA and PBS ) [56] . To our knowledge , this is the first report of a single serological test to detect three flavivirus infections . Our findings that combination of DENV1 to DENV4 NS1 IgG MIA increased the sensitivity to 94 . 3% while maintaining a specificity of 97 . 2% and that the rMFI ratio of ZIKV NS1 to DENV1 or DENV2 NS1 can distinguish ZIKVwprDENV and sDENV infections with a sensitivity of 83 . 3–94 . 4% and specificity of 90 . 9–100 . 0% have important applications to serodiagnosis and serosurveillance of DENV and ZIKV infections in regions where both viruses co-circulate . Generally in agreement with our recent study of individual DENV NS1 ELISAs [45] , we found that DENV1 and DENV3 NS1 IgG MIAs can detect primary DENV infection of the homologous serotype with a sensitivity ( 100% ) higher than that for heterologous serotypes ( 25 . 0 to 100% ) ( Table 2 ) . DENV1 , DENV2 and DENV3 NS1 IgG MIAs can detect secondary DENV infection with a sensitivity of 95 . 5 to 100% . This was also consistent with our previous study using Western blot analysis , in which anti-NS1 antibodies recognized NS1 protein predominantly of the infecting serotype after primary DENV infection and multiple NS1 proteins after secondary infection [13] . Taken together , due to the variable and extensive cross-reactivity of anti-NS1 antibodies after primary and secondary DENV infections , respectively , it is difficult to use a single NS1 IgG MIA or ELISA to identify the infecting DENV serotype . Notably , the combination of four DENV NS1 IgG MIA can detect different primary and secondary DENV infections with a sensitivity of 94 . 3% and specificity of 97 . 2% ( Table 2 ) , suggesting the feasibility and application of this multiplex NS1 IgG MIA to detect DENV infection rather than distinguish DENV serotypes . The overall sensitivity of the ZIKV NS1 IgG MIA was 100% and the specificity was 87 . 9% , primarily due to the cross-reactivity of the sDENV panel ( Table 2 ) . The sensitivity ( 100% ) was higher than or comparable with those previously reported ( 79 to 100% ) using the Euroimmun ZIKV NS1 IgG ELISA kit [34–37] . The ZIKV NS1 blockade of binding ELISA had an overall specificity of 91 . 4–92 . 6% , which reduced to 77 . 6–90 . 5% when comparing with sDENV panel [38 , 39] . A recently reported ZIKV NS1 IgG3 ELISA had a sensitivity of 97% based on samples from Salvador , but it reduced to 83% when comparing with samples outside of Salvador [32] . A previous study of multiplex IgG MIA including ZIKV NS1 reported a sensitivity of 100% and specificity of 78% for pZIKV panel based on PRNT results , however , the sDENV and ZIKVwprDENV panels were not distinguished [56] . For the WNV NS1 IgG MIA , the overall sensitivity was 86 . 1% probably due to sampling during the early convalescent-phase for this pWNV panel ( S1 Table ) , and the specificity was 78 . 4% , mainly due to the cross-reactivity from the sDENV and ZIKVwprDENV panels ( Table 2 ) . Using the rMFI ratio of ZIKV NS1 to DENV1 or DENV2 NS1 , we can distinguish ZIKVwprDENV and sDENV panels with a sensitivity of 83 . 3–94 . 4% and specificity of 90 . 9–100 . 0% . This was consistent with our previous reports of IgG ELISAs using the rOD ratio of ZIKV NS1 to DENV1 NS1 or mixed DENV1-4 NS1 to distinguish these two panels with a sensitivity of 91 . 7–94 . 1% and specificity of 87 . 0–95 . 0% [44 , 45] . It is worth noting since DENV3 and DNV4 NS1 proteins were not recognized by several samples from the sDENV and ZIKVwprDENV panels ( Fig 2A and 2D ) , they were not included in the analysis of the rMFI ratio . Comparing the results of individual NS1 IgG MIA in this study and those of NS1 IgG ELISA reported previously [45] , we found comparable detection rates between MIA and ELISA , and positive correlations between the rMFI and rOD values for both convalescent-phase and post-convalescent-phase samples of most panels tested including pDENV1 , sDENV and ZIKVwprDENV panels except pZIKV panel ( Figs 3 and 4 and S2 Fig ) . Of note , the IgG MIA detection rates for DENV1-4 for the post-convalescent-phase ZIKVwprDENV panel were much lower than those for the sDENV panel ( Fig 4E and S2E Fig ) , suggesting that prior DENV exposure of the ZIKVwprDENV panel may have been only to a single DENV serotype . For the convalescent-phase pZIKV panel , the higher detection rate of ZIKV NS1 IgG MIA ( 100% ) than that of ELISA ( 5% ) and the positive correlation between rOD and rMFI values suggest that MIA was more sensitive than ELISA ( Fig 3A–3C ) . Thus , we did not observe a trend of increased detection rates of NS1 IgG MIA from convalescent to post-convalescent phases for primary infection panels ( pZIKV , pDENV1 ) ( Fig 3B and 3F ) as previously reported for NS1 IgG ELISA and blockade of binding of NS1 ELISA [38 , 45] . Notably we incubated 16 ng antigen coated on each well with 50 μL of serum ( 1:400 ) in ELISA , whereas we incubated ~10 ng antigen ( in 1250 beads ) with serum ( final dilution 1:200 ) per well in MIA . The higher concentration of serum and more surface area of antigen coupled on beads may account for the higher sensitivity of the IgG MIA compared with IgG ELISA for the pZIKV convalescent-phase panel . Although neutralization tests are still considered a confirmatory assay , they are time-consuming and can be performed only in reference laboratories . Compared with PRNT and ELISA , the multiplex MIA requires less time ( 2 . 5 h vs . 7 h for ELISA and 5–6 days for PRNT ) and less sample volume ( 1 μL vs . 8 μL for ELISA and 144 μL for PRNT for 8 antigens or viruses ) . The newly developed multiplex NS1 IgG MIA could have wide-ranging applications , such as serodiagnosis , blood screening , serosurveillance of ZIKV , DENV and WNV infections , and retrospective study of ZIKV infection among pregnant women with CZS [57 , 58] . The current octaplex ( 6 NS1 antigens plus PBS and BSA controls ) IgG MIA serves as a “proof-of-concept” assay to demonstrate that NS1-based MIA can distinguish three flavivirus infections; incorporation of other antigens would increase the detection capacity for different clinical settings and studies . These together would further our understanding of the epidemiology , pathogenesis and complications of ZIKV in regions where multiple flaviviruses co-circulate [1–4] . There are several limitations of this study . First , due to limited samples of < 3 months PSO from patients with primary DENV infection ( S1 Table ) , the study focused on NS1 IgG MIA . Future studies on NS1-based IgM MIA are warranted . Second , despite the availability of two-time point samples for the pZIKV and ZIKVwprDENV panels , future studies involving more sequential samples are needed to validate these observations . Additionally , the sample size in each panel with well-documented infection is small . Third , although this multiplex assay can distinguish various panels of samples with three flavivirus infections , future tests that can distinguish other pathogenic flaviviruses such as JEV , YFV and tick-borne encephalitis virus ( TBEV ) remain to be exploited [59 , 60] . Moreover , samples with well-documented repeated flavivirus infections such as DENV with previous ZIKV infection and sequential DENV and WNV infections are lacking and remain to be investigated in the future . In light of the successful implementation of several flavivirus vaccines and vaccine trials in flavivirus-endemic regions , serological tests that can distinguish ZIKV infection from vaccinations with DENV , JEV , YFV and TBEV vaccines are warranted [59 , 60] . | Although there was a decrease of Zika virus ( ZIKV ) infection since late 2017 , the specter of congenital Zika syndrome and its re-emergence in flavivirus-endemic regions emphasize the need for sensitive and specific serological tests to distinguish ZIKV , dengue virus ( DENV ) and other flaviviruses . Compared with traditional tests based on envelope protein , several nonstructural protein 1 ( NS1 ) -based assays had improved specificity , however , none can discriminate three flaviviruses in a single assay . Moreover , secondary DENV infection and ZIKV infection with previous DENV infection , both common in endemic regions , cannot be distinguished . Herein we developed a high-throughput and multiplex IgG microsphere immunoassay using the NS1 proteins of four DENV serotypes , ZIKV and West Nile virus to test samples from laboratory-confirmed cases with different primary and secondary flavivirus infections . Combination of four DENV NS1 assays revealed a sensitivity of 94 . 3% and specificity of 97 . 2% . The ZIKV and WNV NS1 assays had a sensitivity/specificity of 100%/87 . 9% and 86 . 1%/78 . 4% , respectively . Based on the signal ratio of ZIKV NS1 to DENV1 NS1 , the assay can distinguish ZIKV infection with previous DENV infection and secondary DENV infection with a sensitivity of 88 . 9–90 . 0% and specificity of 91 . 7–100 . 0% . This has applications to serodiagnosis and serosurveillance in endemic regions . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"viruses",
"rna",
"viruses",
"antibodies",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"serology",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"immunoassays",
"biochemistry",
"west",
"nile",
"virus",
"flaviviruses",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"organisms",
"zika",
"virus"
]
| 2019 | A high-throughput and multiplex microsphere immunoassay based on non-structural protein 1 can discriminate three flavivirus infections |
The contactin-associated protein-like 2 ( CNTNAP2 ) gene is a member of the neurexin superfamily . CNTNAP2 was first implicated in the cortical dysplasia-focal epilepsy ( CDFE ) syndrome , a recessive disease characterized by intellectual disability , epilepsy , language impairments and autistic features . Associated SNPs and heterozygous deletions in CNTNAP2 were subsequently reported in autism , schizophrenia and other psychiatric or neurological disorders . We aimed to comprehensively examine evidence for the role of CNTNAP2 in susceptibility to psychiatric disorders , by the analysis of multiple classes of genetic variation in large genomic datasets . In this study we used: i ) summary statistics from the Psychiatric Genomics Consortium ( PGC ) GWAS for seven psychiatric disorders; ii ) examined all reported CNTNAP2 structural variants in patients and controls; iii ) performed cross-disorder analysis of functional or previously associated SNPs; and iv ) conducted burden tests for pathogenic rare variants using sequencing data ( 4 , 483 ASD and 6 , 135 schizophrenia cases , and 13 , 042 controls ) . The distribution of CNVs across CNTNAP2 in psychiatric cases from previous reports was no different from controls of the database of genomic variants . Gene-based association testing did not implicate common variants in autism , schizophrenia or other psychiatric phenotypes . The association of proposed functional SNPs rs7794745 and rs2710102 , reported to influence brain connectivity , was not replicated; nor did predicted functional SNPs yield significant results in meta-analysis across psychiatric disorders at either SNP-level or gene-level . Disrupting CNTNAP2 rare variant burden was not higher in autism or schizophrenia compared to controls . Finally , in a CNV mircroarray study of an extended bipolar disorder family with 5 affected relatives we previously identified a 131kb deletion in CNTNAP2 intron 1 , removing a FOXP2 transcription factor binding site . Quantitative-PCR validation and segregation analysis of this CNV revealed imperfect segregation with BD . This large comprehensive study indicates that CNTNAP2 may not be a robust risk gene for psychiatric phenotypes .
The contactin-associated protein-like 2 ( CNTNAP2 ) is located on chromosome 7q35-36 . 1 , and consists of 24 exons spanning 2 . 3Mb , making it one of the largest protein coding genes in the human genome . This gene encodes the CASPR2 protein , related to the neurexin superfamily , which localises with potassium channels at the juxtaparanodal regions of the Ravier nodes in myelinated axons , playing a crucial role in the clustering of potassium channels required for conduction of axon potentials [1] . CNTNAP2 is expressed in the spinal cord , prefrontal and frontal cortex , striatum , thalamus and amygdala; this pattern of expression is preserved throughout the development and adulthood [2 , 3] . Its function is related to neuronal migration , dendritic arborisation and synaptic transmission [4] . The crucial role of CNTNAP2 in the human brain became clear in 2006 when Strauss et al , reported homozygous mutations in Old Order Amish families segregating with a severe Mendelian condition , described as cortical dysplasia-focal epilepsy ( CDFE ) syndrome ( OMIM 610042 ) [5] . In 2009 , additional patients with recessive mutations in CNTNAP2 were reported , with clinical features resembling Pitt-Hopkins syndrome [6] . To date 33 patients , mostly from consanguineous families , have been reported with homozygous or compound deletions and truncating mutations in CNTNAP2 [5–9] , and are collectively described as having CASPR2 deficiency disorder [7] . The common clinical features in this phenotype include severe intellectual disability ( ID ) , seizures with age of onset at two years and concomitant speech impairments or language regression . The phenotype is often accompanied by dysmorphic features , autistic traits , psychomotor delay and focal cortical dysplasia . CNTNAP2 is also thought to contribute to diverse phenotypes in patients with interstitial or terminal deletions at 7q35 and 7q36 . Interstitial or terminal deletions encompassing CNTNAP2 and several other genes have been described in individuals with ID , seizures , craniofacial anomalies , including microcephaly , short stature and absence of language [10] . The severe language impairments observed in patients with homozygous mutations or karyotypic abnormalities involving CNTNAP2 suggested a possible functional interaction with FOXP2 , a gene for which heterozygous mutations lead to a monogenic form of language disorder [11] . Interestingly , Vernes et al . , found that the FOXP2 transcription factor has a binding site in intron 1 of CNTNAP2 , regulating its expression [12] . Considering that a large proportion of autistic patients show language impairments and most individuals with homozygous mutations in CNTNAP2 manifest autistic features , several studies investigated the potential involvement of CNTNAP2 in autism spectrum disorder ( ASD ) . In particular , two pioneering studies showed that single nucleotide polymorphism ( SNP ) markers rs2710102 and rs7794745 were associated with risk of ASD [13 , 14] . These studies were the first implicating CNTNAP2 in autism , and opened a chapter of additional analyses in ASD and other psychiatric phenotypes during the next decade . In subsequent studies , rs2710102 was implicated in early language acquisition in the general population [15] , and showed functional effects on brain activation in neuroimaging studies [16–19] . Furthermore , genotypes at rs7794745 were associated with reduced grey matter volume in the left superior occipital gyrus in two independent studies [20 , 21] , and alleles of this SNP were reported to affect voice-specific brain function [22] . Genetic associations with ASD for these , and several other SNPs in CNTNAP2 , have been reported in a number of studies [23–28] . Along with the first reports of SNPs associated with ASD , copy number variant ( CNV ) deletions have also been described in ID or ASD patients , which were proposed to be highly penetrant disease-causative mutations [13 , 29–38] . To better understand the role of CNTNAP2 in ASD pathophysiology , knockout mice were generated . Studies of these mice reported several neuronal defects when both copies of CNTNAP2 are mutated: abnormal neuronal migration , reduction of GABAergic interneurons , deficiency in excitatory neurotransmission , and the delay of myelination in the neocortex [2 , 39 , 40] . These intriguing findings prompted additional investigations of CNTNAP2 across other psychiatric disorders or language-related traits , with additional reports of SNPs being associated with schizophrenia ( SCZ ) , bipolar disorder ( BD ) , specific language impairment ( SLI ) and several other phenotypes or traits [12 , 15 , 41–50] . Consequently , other studies reported CNV deletions in CNTNAP2 in schizophrenia [51 , 52] , bipolar disorder [52–54] , and ADHD [55]; but also in neurological disorders , especially epilepsy [56–61] , and language-related phenotypes [62–65] . Interestingly , several of these structural variants were found in intron 1 of CNTNAP2 , encompassing the FOXP2 transcription factor binding site . Epilepsy is clinically frequent in psychiatric disorders , especially schizophrenia and bipolar disorder [66–69] , and is present in approximately 20% of autistic patients [70 , 71] . Similarly , cognitive deficits involving language-related domains are also comorbid traits in schizophrenia and bipolar disorder [67 , 72–74] , and are common clinical features in ASD [75] , with many ASD patients remaining non-verbal throughout life [75 , 76] . While CNTNAP2 is now considered a strong candidate gene for ASD and psychiatric disease more generally ( summarised in Table 1 ) , several of these early supportive studies were performed with limited sample sizes , or were individual case reports which lacked comparison with control individuals , hence providing circumstantial evidence as a psychiatric risk gene . We therefore aimed in this current study to perform systematic genetic analyses with large datasets to examine the evidence for a role of the CNTNAP2 gene in multiple psychiatric phenotypes–performing a comprehensive analysis of common and rare variants , CNVs and de novo mutations–using both publicly available datasets and in-house data .
During the last decade , several association studies have been performed to assess the role of common variants of CNTNAP2 in autism or speech-related phenotypes [12–15 , 23–28 , 46–48 , 50] , as well as several other psychiatric phenotypes [41–45 , 49] . In Table 2 , we summarise all markers found significantly associated in these previous studies , and report the corresponding P-value from the Psychiatric Genomics Consortium GWAS for seven major psychiatric disorders: ADHD , anorexia nervosa , ASD , bipolar disorder , MDD , OCD and schizophrenia . Nominal associations were found with ASD for the following markers: rs802524 ( P = 0 . 016 ) , rs802568 ( P = 0 . 008 ) , rs17170073 ( P = 0 . 008 ) , and rs2710102 ( P = 0 . 036; which is highly correlated with 4 SNPs: rs759178 , rs1922892 , rs2538991 , rs2538976 ) . Furthermore , nominal association was also observed with schizophrenia for rs1859547 ( P = 0 . 044 ) ; with ADHD for rs1718101 ( P = 0 . 038 ) ; with MDD for rs12670868 ( P = 0 . 047 ) , rs17236239 ( P = 0 . 006 ) , rs4431523 ( P = 0 . 001 ) ; and with anorexia nervosa for rs700273 ( P = 0 . 013 ) . The nominal association with ASD at rs1770073 and rs2710102 represents the only case in which the association in the original report replicates in the PGC dataset for the same phenotype . The two SNPs rs7794745 and rs2710102 , which were repeatedly reported as being associated in earlier studies with smaller sample size and proposed to be functional SNPs , were not strongly associated with any phenotype ( the most significant signal being P = 0 . 036 for rs2710102 in ASD ) . None of those associations survived corrections for multiple comparisons ( Table 2 ) . Next , we explored the contribution of common variants across CNTNAP2 by performing a gene-based association study in MAGMA using GWAS summary statistics from PGC data of seven psychiatric disorders in European populations ( Table 3 ) . Association plots for all SNPs included in analysis of each individual phenotype are shown in supporting information ( S1 Fig ) . The test included a dense coverage of SNPs across CNTNAP2: from 1 , 214 SNPs in MDD up to 12 , 264 SNPs in schizophrenia . The results suggest that common variants overall do not contribute to disease susceptibility of these phenotypes ( Table 3 ) . The most significant association observed was for MDD phase 1 analysis ( P = 0 . 029 ) , which is the dataset with the most modest coverage of markers . To explore whether any gene-based signal is not being detected due to a high signal-to-noise ratio ( i . e . inclusion of a large number of SNPs of no functional consequence ) , we selected 63 predicted functional SNPs in CNTNAP2 and performed a meta-analysis across psychiatric disorders ( for regional association plot , see S2 Fig ) . Nominal significance of association was observed for 11 predicted functional SNPs with P-values ranging from 0 . 01 and 0 . 05 , but none survive correction for multiple comparisons ( Table 4 ) . The only predicted functional SNP which was previously reported as being associated with ASD was rs34712024 [25] , but this variant was not associated with autism in the PGC dataset ( P = 0 . 67; Table 2 ) , nor other psychiatric phenotypes examined separately or together ( Tables 2–4 ) . MAGMA gene-based association analysis using this more restricted pool of common putative functional variants revealed significant association with ADHD after correction for multiple testing ( corrected P-value = 0 . 033 ) and a nominal association with schizophrenia which did not survive multiple testing correction ( S1 Table ) . However , this signal is reduced to trend level in the cross-disorder meta-analysis for this functional SNP-set ( P-value = 0 . 11; S1 Table ) . De novo variants in protein-coding genes which are predicted to be functionally damaging are considered to be highly pathogenic and have been extensively explored to implicate genes in psychiatric diseases , especially in ASD and schizophrenia [77] . We explored publicly available sequence data from previous projects in psychiatric disorders to assess the rate of coding de novo variants in CNTNAP2 using two databases ( NPdenovo , http://www . wzgenomics . cn/NPdenovo/; and denovo-db , http://denovo-db . gs . washington . edu/denovo-db/ ) . No truncating or missense variants were identified across CNTNAP2 in 15 , 539 families ( including 2 , 163 controls ) , and synonymous variants were reported in only two probands with developmental disorder ( Table 5 ) . Finally , we explored the potential impact of pathogenic ultra-rare variants ( URV ) in CNTNAP2 using available sequencing datasets of 4 , 483 patients with ASD and 6 , 135 patients with schizophrenia compared with 13 , 042 controls . We considered only those variants predicted to be pathogenic in both SIFT and Polyphen and which are ultra-rare ( MAF<0 . 0001 in Non-Finnish European population; S3 Table ) . No difference in the total number of URV was observed between controls and patients with ASD ( P = 0 . 11 ) , or schizophrenia ( P = 0 . 78 ) ( Table 6 ) . Several deletions and duplications have been described in neuropsychiatric phenotypes thus far . In Fig 1 , we present a comprehensive representation of all previously reported structural variants found in CNTNAP2 in psychiatric disorders such as ASD or ID [13 , 29–38] , schizophrenia or bipolar disorder [51–54 , 78] , ADHD [55] , neurologic disorders such as epilepsy , Tourette syndrome or Charcot-Marie-Tooth [56–61]; and finally language-related phenotypes such as speech delay , childhood apraxia of speech and dyslexia [62–65] . Interestingly , the reported structural variants frequently map in intron 1 , and extend to exon 4 in some cases . The distribution of those structural variants across different phenotypes does not vary with those found in control populations from the database of genomic variants ( http://dgv . tcag . ca/dgv/app/home ) ( Fig 1 ) , suggesting that structural variants in CNTNAP2 are not rare events associated exclusively to disease but are present with rare frequency in the general population . Unfortunately , as many reported CNVs come from individual case reports for which the number of subjects screened is not reported , direct frequency comparisons of this data are not meaningful . CNV microarray analysis was previously performed in two affected individuals from an extended family which included five relatives affected with bipolar I disorder [78] . A drop in signal intensity for 340 consecutive probes was compatible with a deletion of 131 kb in intron 1 of CNTNAP2 ( hg19/chr7:146203548–146334635; Fig 2A ) , encompassing the described binding site for the transcription factor FOXP2 ( hg19/chr7:146215016–146215040 ) [12] . The deletion was detected in one of the two affected individuals examined by CNV array . To infer deletion segregation amongst additional relatives , WES-derived genotypes were used to create haplotypes across chromosome 7q35 ( Fig 2B ) . CNV segregation ( by haplotype inference ) was uninformative due to: 1 ) incomplete genotype data ( unaffected descendants of deceased patient 8404 were not included in the WES study ) and 2 ) a likely recombination at 7q35 in the family . Thus experimental validation and CNV genotyping was performed in all individuals with DNA available to assess the presence of the CNTNAP2 intronic deletion and its disease association . Using quantitative PCR , the deletion was validated in proband subject 8401 , and was detected in one unaffected descendant of deceased patient 8404 ( Fig 2B and 2C ) , implying that: 1 ) affected subject 8404 would have carried the deletion , had DNA been available; and 2 ) the CNV is unlikely to be highly penetrant as it was observed in an unaffected adult relative . The structural variant was not detected in the remaining affected relatives and therefore did not segregate with disease status in this family ( Fig 2B ) .
During the last decade , the CNTNAP2 gene has received considerable attention in the psychiatric genetics field , with a number of studies examining gene dosage , and common or rare variants associations across multiple major psychiatric disorders , which together provided compelling evidence that CNTNAP2 may be a risk gene with pleiotropic effects in psychiatry . While homozygous mutations in this gene lead to a rare and severe condition described as CASPR2 deficiency disorder ( CDD ) [7] , characterized by profound intellectual disability , epilepsy , language impairment or regression [7 , 8] , heterozygous mutations or common variants have been suggested to be implicated in autism , whose clinical features overlap with some observed in CDD . CNTNAP2 is categorised in the SFARI database as syndromic gene and one of the highest-ranking “strong candidate” gene for ASD ( https://gene . sfari . org ) . Heterozygous deletions encompassing the CNTNAP2 gene were described not only in autism but a wide range of phenotypes , including psychiatric or neurologic disorders , and language-related deficiencies . These structural variants were generally described as causative or highly penetrant [13 , 29 , 31 , 55 , 57 , 59] . Examination of the distribution of all structural variants described thus far in psychiatric or neurologic patients showed comparable localisation to those found in the general population , suggesting that structural variants affecting CNTNAP2 may be less relevant in disease susceptibility than previously considered . We were not able to directly compare frequencies of observed structural variants in cases versus controls due to reporting bias in case reports and a lack of information on how many cases were screened to identify those subjects with reportable CNTNAP2 CNVs , which is a limitation of this study . In the ExAC database , CNTNAP2 had fewer CNV variants than expected ( 11 observed vs . 16 expected , z = 0 . 43; http://exac . broadinstitute . org ) , and its haploinsufficiency score of 0 . 59 is in the 8th decile of all genes [79] , suggesting that CNTNAP2 has a tendency to be intolerant to structural variants . A specific case-control CNV analysis is needed to examine CNV frequency differences , but would require a very large sample due to the rarity of CNVs at this locus . A close clinical psychiatric examination of the 66 parents with heterozygous deletions across CNTNAP2 of CDD provides information on the prevalence of psychiatric conditions in individuals carrying CNTNAP2 CNVs . All heterozygous family members carrying deletions or truncating mutations were described as phenotypically healthy , suggesting a lack of correlation between these deletions and any major psychiatric condition . Furthermore , parents who were carriers for heterozygous deletions in psychiatric/neurologic patients were described as unaffected at the time of reporting [13 , 29 , 31 , 37 , 54 , 62] , with two exception: one father of a proband with neonatal convulsion , and another father of an epileptic patient , were reported as affected [56 , 59] . Moreover , discordant segregation for deletions in CNTNAP2 was also observed in an ASD sib-pair [13] . Several psychiatric patients who were reported to carry heterozygous structural variants in CNTNAP2 were also described with translocations or other chromosomal abnormalities [29 , 30 , 33 , 34 , 56 , 58 , 62–65] , therefore it is possible that these aberrations may explain the phenotype independently from the observed CNVs in CNTNAP2 . We also describe a new CNV deletion which does not segregate with disease in an extended family with bipolar disorder . This CNV removes the FOXP2 transcription factor binding site in intron 1 of CNTNAP2 , and overlaps with structural variants described in a number of other psychiatric patients . This heterozygous deletion was identified in two individuals with bipolar I disorder from an extended family with five affected members , but was observed also in one unaffected relative ( who underwent diagnostic interview at age >40 and therefore was beyond the typical age of symptom onset ) . Hence , the deletion was not segregating with the disease and is unlikely to represent a highly penetrant risk variant in this family , although we cannot exclude a multiple hit model where the CNV deletion interacts with other etiologic risk variants at other loci to exert phenotypic effect . CNTNAP2-/- knock-out mice have been proposed as valid animal model for ASD considering the phenotypic similarities between ASD and the CASPR2 deficiency disorder [2] . CNTNAP2-/- knock-out mice showed abnormalities in the arborisation of dendrites , maturation of dendritic spines , defects in migration of cortical projection neurons , and reduction of GABAergic interneurons [2 , 4] . Controversially , ASD is not a core feature in the most recent patient series reported with CASPR2 deficiency disorder [7 , 8] . The association previously proposed around the relationship between heterozygous deletions in CNTNAP2 and ASD does not have a support from mouse models , as heterozygous mice did not show any behavioural or neuropathological abnormalities that were observed in homozygous knockouts [2] . Notwithstanding this , it is possible that the combination of heterozygous CNTNAP2 deletions in a genomic background of increased risk ( through inheritance of other common and rare risk variants at other loci ) may lead to psychiatric , behavioural or neuropathological abnormalities . Common variants in CNTNAP2 are another class of genetic variation associated with several psychiatric or language-related phenotypes . The most interesting finding from studies of this variant class converge on markers rs7794745 and rs2710102 , originally reported in ASD [13 , 14] , and replicated later in ASD or implicated in other phenotypes [12 , 15 , 23 , 24 , 46–48] . Neuroimaging studies have supported the notion that these common variants play a role in psychiatric disorders . SNP rs2710102 has been implicated in brain connectivity in healthy individuals [16 , 18 , 19] , and rs7794745 was implicated in audio-visual speech perception [80] , voice-specific brain function [22] , and was associated with reduced grey matter volume in left superior occipital gyrus [20 , 21] . These studies focused principally on language tasks in general population , given the reported suggestive implications of CNTNAP2 in language impairment traits of ASD or language-related phenotypes . However , the direct role of CNTNAP2 in language is still unclear; indeed the language regression observed in patients with CASPR2 deficiency disorder are concomitant with seizure onset and may represent a secondary phenotypic effect caused by seizures [7] . On the other hand , the first genetic association of rs7794745 and rs2710102 with ASD , as well as the other psychiatric diseases were based in studies with limited sample size , and recent studies failed to replicate associations between the two markers and ASD [81 , 82] . Individual alleles associated in the past with limited numbers of patients warrant replications in adequately powered samples to ascertain bona fide findings considering the small size effects of common variants [83] , such as that attempted here . Using the largest case-control cohorts currently publicly available ( PGC datasets ) , we did not find evidence for significant association of previously reported common variants , or a combined effect for common variants of CNTNAP2 in the susceptibility of psychiatric disorders , nor did we find predicted functional SNPs with a role across disorders . Finally , we examine evidence for rare variant contributions in CNTNAP2 . Rare variants in the promoter or coding region were reported to play a role in the pathophysiology of ASD [25 , 33] , although a recent study including a large number of cases and controls did not find association of rare variants of CNTNAP2 in ASD [84] . Here we report the largest sample investigated thus far in ASD and schizophrenia , which suggests that rare variants in CNTNAP2 do not play a major role in these two psychiatric disorders . Furthermore , examination of de novo variants in combined psychiatric sequencing projects of over 15 , 500 trios suggest that de novo variants in CNTNAP2 do not increase risk for psychiatric disorders . While functional studies show a relationship between certain deletions or rare variants of CNTNAP2 with neuronal phenotypes relevant to psychiatric illness [25 , 54 , 85] , we show that the genetic link between these variants and psychiatric phenotypes is tenuous . However , this does not dispel the evidence that the CNTNAP2 gene , or specific genetic variations within this gene , may have a real impact on neuronal functions or variability of brain connectivity in the general population . It is now possible to combine large datasets to ascertain the real impact of candidate genes described in the past in psychiatric disorders . Here we performed analyses using large publicly available datasets investigating a range of mutational mechanisms which impact variability of CNTNAP2 across several psychiatric disorders . In conclusion , our results converge to show a limited or likely neutral role of CNTNAP2 in the susceptibility of psychiatric disorders . However , the impact of this gene in language deficit per se is not directly examined in this study and warrants additional investigation .
We sought to replicate previously reported CNTNAP2 SNP associations in a range of psychiatric phenotypes or traits using GWAS summary-statistic data of the Psychiatric Genomics Consortium ( https://med . unc . edu/pgc/results-and-downloads ) . Firstly , we report the corresponding P-values of specific previously associated markers for case-control cohorts with autism spectrum disorder ( ASD ) , schizophrenia ( SCZ ) , bipolar disorder ( BD ) , attention-deficit hyperactivity-disorder ( ADHD ) , major depressive disorder ( MDD ) , anorexia nervosa ( AN ) , and obsessive compulsive disorder ( OCD ) . If a specific SNP marker was not reported in an individual GWAS dataset , we selected another marker in high linkage disequilibrium ( r2~1 , using genotype data from the CEU , TSI , GBR and IBS European populations in 1000genomes project; http://www . internationalgenome . org ) . Next , a gene-based association for common variants was calculated with MAGMA [86] , using variants within a 5 kb window upstream and downstream of CNTNAP2 . Selected datasets were of European descent , derived from GWAS summary statistics of the Psychiatric Genomics Consortium ( https://med . unc . edu/pgc/results-and-downloads ) : SCZ ( 33 , 640 cases and 43 , 456 controls ) , BD ( 20 , 352 cases and 31 , 358 controls ) , ASD ( 6 , 197 and 7 , 377 controls ) , ADHD ( 19 , 099 cases and 34 , 194 controls ) , MDD ( 9 , 240 cases and 9 , 519 controls ) , OCD ( 2 , 688 cases and 7 , 037 controls ) , and AN ( 3 , 495 cases and 10 , 982 controls ) [87–93] . Analyses were performed combining two different models for higher statistical power and sensitivity when the genetic architecture is unknown: the combined P-value model , which is more sensitive when only a small proportion of key SNPs in a gene show association; and the mean SNP association , which is more sensitive when allelic heterogeneity is greater and a larger number of SNPs show nominal association . Finally , we selected SNPs predicted to be functional within a 5kb window upstream/downstream of CNTNAP2 ( e . g . located in transcription factor binding sites , miRNA binding sites etc; https://snpinfo . niehs . nih . gov ) , and assessed a potential cross-disorder effect using GWAS summary statistics data of the PGC by performing a meta-analysis in PLINK [94] . The Cochran’s Q-statistic and I2 statistic were calculated to examine heterogeneity amongst studies . The null hypothesis was that all studies were measuring the same true effect , which would be rejected if heterogeneity exists across studies . For all functional SNPs , when heterogeneity between studies was I>50% ( P<0 . 05 ) , the pooled OR was estimated using a random-effects model . The impact of rare variants of CNTNAP2 was assessed using sequencing-level data from the following datasets: WES from the Sweden-Schizophrenia population-based Case-Control cohort ( 6 , 135 cases and 6 , 245 controls; dbGAP accession: phs000473 . v2 . p2 ) ; ARRA Autism Sequencing Collaboration ( 490 BCM cases , BCM 486 controls , and 1 , 288 unrelated ASD probands from consent code c1; dbGAP accession: phs000298 . v3 . p2 ) ; Medical Genome Reference Bank ( 2 , 845 healthy Australian adults; https://sgc . garvan . org . au/initiatives/mgrb ) ; individuals from a Caucasian Spanish population ( 719 controls [95 , 96] ) ; in-house ASD patients ( 30 cases; [97] ) ; and previous published dataset in ASD ( 2 , 704 cases and 2 , 747 controls [84] ) . The selection of potentially etiologic variants was performed based on their predicted pathogenicity ( missense damaging in both SIFT and polyphen 2 , canonical splice variants , stop mutation and indels ) and minor allele frequency ( MAF<0 . 0001 in non-Finnish European populations using the Genome Aggregation Database; http://gnomad . broadinstitute . org/ ) . A chi square statistic was used to compare separately the sample of schizophrenia patients ( 6 , 135 cases ) and the combined ASD datasets ( 4 , 512 cases ) with the combined control datasets ( 13 , 042 individuals ) . Two databases for de novo variants were used to identify de novo variants in CNTNAP2 [98 , 99] , which comprise data for the following samples: autism spectrum disorder ( 6 , 171 families ) , schizophrenia ( 1 , 164 families ) , epilepsy ( 647 families ) , intellectual disability ( 1 , 101 families ) , developmental disorders ( 4 , 293 families ) and controls ( 2 , 163 ) . The extended family presented here ( Fig 2B ) provides a molecular follow-up from a previously reported whole exome sequencing ( WES ) study of multiplex BD families , augmented with CNV microarray data [78] . This multigenerational pedigree , was collected through the Mood Disorders Unit and Black Dog Institute at the Prince of Wales Hospital , Sydney , and the School of Psychiatry ( University of New South Wales in Sydney ) [100–104] . Consenting family members were assessed using the Family Interview for Genetic Studies ( FIGS ) [105] , and the Diagnostic Interview for Genetic Studies ( DIGS ) [106] . The study was approved by the Human Research Ethics Committee of the University of New South Wales , and written informed consent was obtained from all participating individuals . Blood samples were collected for DNA extraction by standard laboratory methods . Three of the five relatives with bipolar disorder type I ( BD-I ) had DNA and WES-derived genotype data available , and six unaffected relatives with DNA and WES data were available for haplotype phasing and segregation analysis ( Fig 2B ) . Genome-wide CNV analysis was performed via CytoScan HD Array ( Affymetrix , Santa Clara , CA , USA ) in 2 distal affected relatives ( individuals 8410 and 8401; Fig 2B ) , using the Affymetrix Chromosome Analysis Suite ( ChAS ) software ( ThermoFisher , Waltham , MA , USA ) . Detailed information on CNV detection and filtering criteria have been previously described [78] . We identified a 131kb deletion in intron 1 of CNTNAP2 in individual 8401 . WES-derived genotypes were used for haplotype assessment to infer CNV segregation amongst relatives , as previously described [78] . Next , we experimentally validated the CNTNAP2 CNV via quantitative PCR ( qPCR ) in all available family members . Validation was performed in quadruplicate via a SYBR Green-based quantitative PCR ( qPCR ) method using two independent amplicon probes , each compared with two different reference amplicon probes in the FOXP2 and RNF20 genes ( S4 Table ) . Experimental details are available upon request . | Genetic mutations that disrupt both copies of the CNTNAP2 gene lead to severe disease , characterized by profound intellectual disability , epilepsy , language difficulties and autistic traits , leading to the hypothesis that this gene may also be involved in autism given some overlapping clinical features with this disease . Indeed , several large DNA deletions affecting one of the two copies of CNTNAP2 were found in some patients with autism , and later also in patients with schizophrenia , bipolar disorder , ADHD and epilepsy , suggesting that this gene was implicated in several psychiatric or neurologic diseases . Other studies considered genetic sequence variations that are common in the general population , and suggested that two such sequence variations in CNTNAP2 predispose to psychiatric diseases by influencing the functionality and connectivity of the brain . To better understand the involvement of CNTNAP2 in risk of mental illness , we performed several genetic analyses using a series of large publicly available or in-house datasets , comprising many thousands of patients and controls . Furthermore , we report the deletion of one copy of CNTNAP2 in two patients with bipolar disorder and one unaffected relative from an extended family where five relatives were affected with this condition . Despite the previous consideration of CNTNAP2 as a strong candidate gene for autism or schizophrenia , we show little evidence across multiple classes of DNA variation , that CNTNAP2 is likely to play a major role in risk of psychiatric diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"neuropsychiatric",
"disorders",
"medicine",
"and",
"health",
"sciences",
"adhd",
"anxiety",
"disorders",
"pervasive",
"developmental",
"disorders",
"obsessive-compulsive",
"disorder",
"bipolar",
"disorder",
"social",
"sciences",
"autism",
"developmental",
"psychology",
"neuroscience",
"cognitive",
"psychology",
"behavioral",
"disorders",
"autism",
"spectrum",
"disorder",
"mood",
"disorders",
"language",
"developmental",
"neuroscience",
"genome",
"complexity",
"genomics",
"neurodevelopmental",
"disorders",
"neuroses",
"schizophrenia",
"mental",
"health",
"and",
"psychiatry",
"psychology",
"neurology",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"cognitive",
"science",
"introns"
]
| 2018 | Comprehensive cross-disorder analyses of CNTNAP2 suggest it is unlikely to be a primary risk gene for psychiatric disorders |
Many complex systems have been found to exhibit critical transitions , or so-called tipping points , which are sudden changes to a qualitatively different system state . These changes can profoundly impact the functioning of a system ranging from controlled state switching to a catastrophic break-down; signals that predict critical transitions are therefore highly desirable . To this end , research efforts have focused on utilizing qualitative changes in markers related to a system’s tendency to recover more slowly from a perturbation the closer it gets to the transition—a phenomenon called critical slowing down . The recently studied scaling of critical slowing down offers a refined path to understand critical transitions: to identify the transition mechanism and improve transition prediction using scaling laws . Here , we outline and apply this strategy for the first time in a real-world system by studying the transition to spiking in neurons of the mammalian cortex . The dynamical system approach has identified two robust mechanisms for the transition from subthreshold activity to spiking , saddle-node and Hopf bifurcation . Although theory provides precise predictions on signatures of critical slowing down near the bifurcation to spiking , quantitative experimental evidence has been lacking . Using whole-cell patch-clamp recordings from pyramidal neurons and fast-spiking interneurons , we show that 1 ) the transition to spiking dynamically corresponds to a critical transition exhibiting slowing down , 2 ) the scaling laws suggest a saddle-node bifurcation governing slowing down , and 3 ) these precise scaling laws can be used to predict the bifurcation point from a limited window of observation . To our knowledge this is the first report of scaling laws of critical slowing down in an experiment . They present a missing link for a broad class of neuroscience modeling and suggest improved estimation of tipping points by incorporating scaling laws of critical slowing down as a strategy applicable to other complex systems .
Rapid transitions to a qualitatively different state can be observed in many complex systems . Their sometimes catastrophic outcomes in systems from diverse fields such as climate , ecology , medicine and economics have led to an increased interest in the underlying structure and dynamics of these transitions [1 , 2] . While the consequences of such shifts are often undesired , the proximity to a transition can also have various beneficial aspects such as to allow for rapid switching between different states and for small changes to have a large effect on the system state . In the brain , for example , this double-edged role is illustrated by the unwanted transition from normal to epileptic brain activity on one side [3 , 4] , and , on the other side , the role state transitions play in changing between mutually exclusive motor programs [5] or the generation of action potentials to efficiently convey information . Better insight into these transitions has come from a dynamical systems’ perspective . For individual neurons , this approach identified two robust mechanisms for the transition from subthreshold near-steady activity to repetitive spiking , saddle-node and Hopf bifurcation [6–9] . The type of threshold behavior predicted by these bifurcations has been able to account for various observations in biological neurons . For example , the smooth frequency vs . current ( f-I ) curve observed in pyramidal neurons stimulated with steady current is predicted by a saddle-node on invariant cycle bifurcation [8 , 10] . Conversely , a discontinuous f-I curve characterized by an abrupt onset of firing as current injection is ramped up has been discussed in the context of an underlying ( subcritical ) Hopf bifurcation [7 , 10 , 11] . Although the above mapping between bifurcation and continuity of the f-I curve is not perfect [7] and the experimental determination of the underlying bifurcation to spiking can be problematic [12–14] , these differences in f-I curves have led to a classification of neurons according to type 1 and type 2 behavior referring to continuous and discontinuous f-I curves , respectively . However , other crucial predictions following from theory of bifurcations have not been demonstrated experimentally . In particular , theory implies that system dynamics should recover more slowly from small perturbations upon approaching the bifurcation or tipping point , a phenomenon called critical slowing down [15] . Critical slowing down can be monitored by measuring the recovery rate of system variables after small perturbations but also manifests itself by an increase in its fluctuations , i . e . variance due to the longer relaxation times near the bifurcation , as well as higher autocorrelation values [16 , 17] . Although theory provides precise quantitative predictions on signatures of critical slowing down for different bifurcations , direct experimental evidence in neurons approaching their spiking threshold has been lacking . The confirmation of critical slowing down and its characteristic scaling in biological neurons therefore represents a missing link between experiment and theory [18] relevant for a large class of neuroscience modeling . Critical slowing down has furthermore attracted considerable attention in a wide range of systems outside of neuroscience . In many real-world settings , warning signals of impending critical transitions are highly desirable because it is often difficult to revert a system to the previous state once a critical transition has occurred [1 , 19] . While qualitative changes in markers related to slowing down have previously been used to probe the proximity to a tipping point in various systems [20–24] , a quantification of their scaling laws , to our knowledge , has never been attempted in an experimental setting . Consequently , the affirmation of scaling laws of critical slowing down in a real-world system could offer refined approaches to the prediction of tipping points by incorporating knowledge about these scaling relations . In the present work , we quantitatively study the scaling laws of critical slowing down for the transition from quiescence to spiking in cortical neurons recorded in the acute brain slice . We show that this transition equates a critical transition exhibiting slowing down where changes in variance and recovery rate are necessary consequences when the bifurcation point is approached . Using bifurcation theory we derive the precise scaling laws relevant in the context of neuronal spiking and compare them to the scaling of variance and recovery rate observed in biological neurons . Our analysis suggests the scaling of these markers of critical slowing down to be governed by a saddle-node bifurcation in both type 1 and type 2 neurons . Furthermore , incorporation of these scaling laws improves bifurcation point prediction from a limited window of observation . To our knowledge this work represents the first quantitative analysis of scaling laws governing critical slowing down in a real-world experimental system .
Using the whole-cell patch configuration and acute slices prepared from 2–4 week old rats , we recorded intracellularly the membrane potential of cortical pyramidal neurons and fast-spiking interneurons in response to current injections . We developed a stimulation protocol that allowed us to monitor markers of critical slowing down [1 , 2 , 16] while systematically increasing the injected current to drive the neuron towards its spiking threshold , i . e . tipping point ( Fig . 1 a , b ) . Specifically , while GABAergic and glutamatergic synaptic transmission was blocked by bath application of PTX ( 50μM ) and AP5/DNQX ( 50μM/10μM ) , respectively , we applied a slowly depolarizing step current to gradually drive the membrane potential towards the spiking threshold . In addition to the step current , we applied brief , subthreshold current pulses at regular intervals as small perturbations . When the neuron started to spike the current injection was stopped . We quantified the neuron’s recovery to each perturbation by fitting an exponential decay to the return of the membrane potential trajectory within a few hundred milliseconds to derive a recovery rate λ ( Fig . 1 b ) . The unperturbed one—second segments before the current pulses were used to calculate variance and autocorrelation from subthreshold voltage . Starting at the resting membrane potential , recovery rates exhibited a gradual decline which became more pronounced towards the spiking threshold . Concomitantly , autocorrelation and variance showed a marked increase towards the membrane potential value at which spiking started ( Fig . 1 c ) . The decrease in recovery rate together with the increase in variance and autocorrelation amount to conclusive evidence for critical slowing down in subthreshold neuronal activity prior to spiking . The existence of critical slowing down is a direct consequence of a bifurcation underlying the transition from quiescence to spiking in neurons [6 , 7 , 25 , 26] . Although there have been experimental reports of changes in subthreshold activity dependent on the level of depolarization in cortical neurons , such as the width and decay of exitatory postsynaptic potentials ( EPSPs ) [12 , 27 , 28] or the amount of subthreshold voltage noise [29] , these observations have not been put into context with critical slowing down at a bifurcation . Even more so , there has been no quantification of these phenomena , which is of particular relevance since bifurcation theory makes precise predictions for the scaling of these markers of critical slowing down . In the following , we will outline in more detail , why it is reasonable to look for critical slowing down in the statistics of subthreshold membrane potential fluctuations near a neuron’s transition from quiescence to spiking . Specifically , we will first present the quantitative scaling laws predicted by theory and second relate them to the scaling observed in experimental measurements from biological neurons . Quiescence or spiking can be regarded as two different states a neuron can be in . The mathematical study of such transitions is called bifurcation theory . From a dynamical systems’ point of view , a neuron’s transition from quiescence to spiking therefore corresponds to a bifurcation in neuron dynamics . It is because of this proximity to a bifurcation that neurons are excitable , i . e . , have the ability to exhibit a qualitative change in their dynamics . Neurons can be driven from quiescence toward their spiking threshold by slowly increasing current injections . The observation of a very drastic end of quiescence ( or steady-state ) upon increase of the injected current beyond a certain threshold suggests that the most likely dynamical transitions are either a saddle-node or a Hopf bifurcation . A saddle-node bifurcation ( i . e . a saddle-node bifurcation or a saddle-node bifurcation on an invariant circle bifurcation ) is characterized by a single eigenvalue of the linearized subsystem passing through the imaginary axis . A Hopf bifurcation ( i . e . a subcritical Andronov-Hopf bifurcation ) has a complex conjugate pair of eigenvalues passing through the imaginary axis [7 , 30–32] . These insights yield the important conclusion that , no matter how one decides to model neurons , one has to analyze the statistics near saddle-node and Hopf bifurcations to obtain the scaling laws near the transition to spiking . Besides being near a local bifurcation , moderate noise levels in the system are a necessary condition to observe critical slowing down , as dynamics could instantly jump to a different attractor when noise levels are too high . Here , we first review the scaling laws governing subthreshold dynamics for both saddle-node and Hopf bifurcation . These scaling laws for recovery rate and variance are known analytically [33] . We will derive the scaling for recovery rate for saddle-node and Hopf bifurcation relevant for the transition to neuronal spiking from their bifurcation normal forms and numerically illustrate the scaling relations for both bifurcations in a model system that captures our experimental approach . We next investigated whether the predicted scaling from theory can be observed in experiment . We thereby focused on neurons for which type 1 and type 2 behaviors have been reported . We hypothesized that the knowledge of the bifurcation that underlies a critical transition could offer a refined approach to predict the bifurcation point from a limited window of observation by using the precise scaling laws governing slowing down . This approach is motivated by many real-world systems exhibiting rare but often catastrophic transitions to a different state . Any insights to better anticipate and predict those transitions would therefore be highly desirable [2] . In particular , it is likely that the control parameter driving the system towards the tipping point , in our case the injected current , may not be directly accessible , but that instead one might be able to monitor some other observable of the system , such as the membrane potential in our experiment . Accordingly , we attempted to predict the voltage V c p at which spiking occurs in pyramidal neurons using only measurements of recovery rate as a function of the neuron’s membrane potential . Omitting the five last data points before spiking ( Fig . 6 a , blue markers ) , we fit λ to the membrane potential by λ = a ( V c p − V ) θ for the remainder of measurements ( Fig . 6 a , red markers ) . This fit allowed the determination of V c p as a fit parameter if θ is known ( Fig . 6 a , red vertical line ) . We compared this predicted value to the measured critical membrane potential V c m defined as the average voltage over the one—second interval prior to spiking ( Fig . 6 a , blue vertical line ) analogous to the definition of Ic . The differences between predicted V c p and measured V c m to the last value taken into account for fitting ( Fig . 6 a , green vertical line ) , i . e . ΔVp and ΔVm , exhibited a significant correlation when data were fit with the saddle-node exponent θ = 0 . 5 ( Fig . 6 b ) . Conversely , there was no significant correlation when prediction was attempted with the exponent θ = 1 . 0 for a Hopf bifurcation ( Fig . 6 c ) . This demonstrates that the knowledge of the underlying bifurcation and its scaling relations for slowing down can be used to estimate the bifurcation point from observation of data .
Numerous experimental studies have demonstrated nonlinear dynamical behavior at the transition to spiking in excitable cells ranging from chaotic attractors and frequency doubling of cardiac pacemaker cells [41] to intermittent bursting in cultured cortical neurons during slow driving [42] . In neurons , the reduction of the spiking mechanism to bifurcations has greatly enhanced the understanding of neuron functioning and is captured by many mathematical neuron models [7] . Although critical slowing down is expected to occur upon approaching the bifurcation point , its actual existence in real neurons had not been rigorously demonstrated . Specifically , the quantification of its scaling laws had been a missing link to theory . The distinction between type 1 and 2 excitability has proven useful to describe the coding properties of neurons [43 , 44] despite the fact that neuronal properties may change on slow time scales for example due to adaptation or bursting [14] , cholinergic modulation [13] , or changes in the density and distribution of ion channels [45] . A saddle-node bifurcation related to type 1 excitability has been indirectly derived to control spiking in pyramidal neurons from their graded f-I curves [7 , 8 , 10 , 11] , their non-monotonic I-V curves , histograms of ISIs , and infinitesimal phase resetting curves [38 , 46–48] . The scaling of variance and recovery rate experimentally observed here in pyramidal neurons is well in line with the precise scaling laws predicted by theory for a saddle-node bifurcation . In FS neurons , whose dynamics at threshold has been described to exhibit type 2 behavior , we also observed scaling with exponents ±0 . 5 . Consequently , these exponents suggest a saddle-node bifurcation instead of a Hopf bifurcation dominating critical slowing down when the spiking threshold is approached from resting membrane potential . Although type 2 excitability is often brought in context with an underyling subcritical Hopf bifurcation , it has been emphasized that this mapping is certainly not clear-cut [7] . A saddle-node ( off a stable limit cycle ) bifurcation , for example , can result in both type 1 or type 2 excitability and could therefore explain the scaling with exponents ±0 . 5 observed here . An interesting alternative to a simple saddle-node bifurcation is a folded node which could also underlie the transition to spiking in FS neurons since it generates exponents ±0 . 5 and can also account for subhreshold oscillations [49] similar to the ones observed here and in [10] . Another possibility in line with exponents ±0 . 5 would be a singular Hopf , i . e . , a mix of fold and Hopf bifurcations whose subthreshold dynamics , however , is governed by the saddle-node [50] . Finally , we should not exclude the possibility that our experimental analysis could provide wrong exponents and that the transition in FS neurons is still governed by a Hopf bifurcation . The fact that the afterhyperpolarization in these neurons is deeper than the fixed point at the previous current level has previously been discussed as one possible indication for a Hopf bifurcation since a similar bistability can be observed in some reduced neuron models , for example [51] . However , given the robust measurement of exponents ±0 . 5 here and the compatibility of bifurcations exhibiting these exponents with other observations such as subthreshold oscillations as well as the missing definite proof for a Hopf bifurcation in these neurons , it appears more likely that it is rather one of the bifurcations discussed above governing critical slowing down in FS interneurons . In particular when one considers the various factors that can modify a neuron’s bifurcation structure [13 , 14 , 45] what type of bifurcation actually governs the transition to spiking in a neuron under investigation can only be determined experimentally . The robust observation of scaling laws for slowing down as demonstrated here , is therefore likely to provide informative insights into the composition of an underlying bifurcation structure and can be a useful additional tool in studying the excitability in neurons besides other markers such as the I-V curve , histograms of ISIs and infinitesimal phase resetting curves , for example . Apart from neurons , the different exponents characterizing saddle-node and Hopf bifurcation open the possibility to infer the underlying bifurcation based on subthreshold scaling laws in other systems , too . The observation of systematic changes in recovery rate and variance as a result of critical slowing down provides a framework to understand previous findings where these metrics were found to be changing depending on the proximity to the spiking threshold . In particular , the width and decay of exitatory postsynaptic potentials ( EPSPs ) have been observed to be dependent on the level of depolarization in neurons [12 , 27 , 28] and have been interpreted in the context of changing inward/outward current balances as the membrane potential approaches the spike threshold . In the framework of critical transitions , an incoming EPSP can be understood as a small perturbation to the membrane potential analogously to the brief current pulses in our protocol and will therefore exhibit the same changes in its recovery to baseline . Thus , our results not only are qualitatively in line with the observed broadening of the EPSP shapes observed when approaching the spiking threshold but also provide the distinct quantitative scaling laws by which these changes manifest . Similarly , a positive correlation of the subthreshold voltage noise level to holding potential in pyramidal neurons has previously been observed [29] . This observation links directly to the increase in variance reported here . The scaling laws of variance provide a quantitative framework to describe these previously unexplained observations in the context of critical slowing down . The decrease in recovery rate as a result of critical slowing down upon approaching the spiking threshold is likely to have implications on information processing in neurons . It can be expected that changes in the width of EPSPs , analogous to changes in the recovery from small current injections in our protocol , will have an effect on the way by which inputs from other neurons are integrated . The systematic widening of postsynaptic potentials close to spike threshold should progressively facilitate the temporal integration of small inputs to a neuron the closer it gets to the spiking threshold . In this regard , the systematic changes in the form of scaling laws observed in biological neurons can be useful to constrain more realistic computational neuron models . For example , most leaky integrate-and-fire neuron models , by omitting the dynamical modeling of action potential generation , do not take the effects of critical slowing down into account , unless specifically incorporating changes in inward/outward current balance near threshold [52] . One can argue that it might be beneficial for a neuron to balance its excitability in a way that its membrane potential is close to firing threshold allowing for rapid switching between quiescence and spiking at minimal energetic cost . To maintain such a high-conductance state [53] , it is conceivable that individual neurons self-organize their excitability [54] and subthreshold statistics such as variance or the length and decay of a transient response such as an EPSP , for example , could consequently be utilized to maintain a neuron close to the spiking threshold . The identical scaling of these statistics in both type 1 and type 2 neurons suggested by our analysis could therefore indicate a universal mechanism by which this tuning towards the bifurcation occurs . Our finding that both pyramidal neurons and fast spiking interneurons are guided by the same scaling law close to spike onset might have important implications for the balance of fast excitation/inhibition ( E/I ) in neuronal networks . A precise E/I-balance has been shown experimentally to be maintained in vivo and in vitro as the network undergoes different levels of excitation [55 , 56] . Modeling work has demonstrated the E/I-balance to establish a decorrelated network state [57–59] . The I-F curves between pyramidal neurons and fast-spiking interneurons display rather different firing dynamics in response to current pulses . Our work , however , demonstrates that both neuronal populations exhibit similar subthreshold scaling close to spike onset which suggests a symmetrical dynamical regulation of the E/I-balance , which might simplify its maintenance . Beyond single neurons , shifts to different dynamical regimes also occur on a larger spatial scale in neuronal systems . Such transitions of cortical network dynamics can be quite subtle and occur , for example , under physiologic conditions in the course of wake and sleep [60] , or are exemplified by the rapid transitions to pathologic seizure states in epilepsy [3 , 4] . A tipping point at the network level has also recently been described as ‘coherence potential’ in the ongoing avalanche dynamics of awake monkeys and in vitro [61] . It will be interesting to explore whether these network transitions exhibit similar scaling laws to those reported here for individual cells and whether they could consequently lead to a better understanding and perhaps even prediction of their occurences . From a more general perspective , our work outlines and applies a strategy of identifying a bifurcation by the scaling relations for markers of slowing down and how to consequently incorporate this knowledge to improve prediction of the transition point . The often irreversible changes that can occur in a large variety of complex systems make signals that warn of these transitions highly desirable [1 , 19] . Although in the specific case of neuron firing one might think of alternative approaches to anticipate the onset of spiking such as simple thresholding or assuming an integrate and fire model with a certain amount of noise , these methods may likely not be applicable to other real-world systems . Recently , a particularly promising approach to predict these kind of critical transitions in a large variety of complex systems has been based on variables related to critical slowing down as these can often be readily monitored independently of a system’s specificities . So far , a large body of research work has attended to qualitative changes in markers of slowing down to anticipate tipping points [20–24] . Our work constitutes , to our knowledge , the first experimental system in which the quantitative scaling laws governing slowing down have been reported . We suggest that the refined prediction based on scaling laws demonstrated here could also be applicable to other complex systems . While a direct measurement of recovery rates may not always be feasible , indirect measures such as variance can also be used to infer the underlying transition mechanism . Once an underlying bifurcation has been identified , in principle , the precise scaling laws can be used to predict the tipping point as demonstrated in the current study . Although prediction performance is naturally impeded by stochastic perturbations which can trigger critical transitions even before the bifurcation point is reached [1 , 62 , 63] , we demonstrate that given sufficient data and moderate noise levels , reasonable quantitative predictions become possible . In this respect , our results can be regarded as a proof of concept that an estimation of the proximity to the tipping point based on quantitative scaling of critical slowing down is possible and provide a step forward in estimating the fragility in complex systems .
Procedures were in accordance with National Institutes of Health guidelines . Animal procedures were approved by the National Institute of Mental Health Animal Care and Use Committee . The brains of Sprague Dawley rats ( P14-P28 ) were removed and cut into acute coronal slices of medial prefrontal or somatomotor cortex at 350μm thickness ( VT1000S , Leica ) in ice-cold artificial cerebral spinal fluid ( ACSF; 124mM NaCl , 1 . 2mM CaCl2 , 1mM MgSO4 , 3 . 5mM KCl , 26 . 2mM NaHCO3 , 0 . 3mM NaH2 PO4 , and 10mM D−Glucose ) bubbled with carbogen ( 95% O2 , 5% CO2 ) . All recordings were performed under perfusion flow rate of 3–4ml/min while continuously monitoring and maintaining temperature at 35±0 . 5°C . The ACSF’s osmolarity was 290±10mOsm . NMDA- and AMPA-mediated synaptic transmission was blocked with bath-application of 50μM AP5 and 10μM DNQX , respectively , and GABAa-mediated transmission with 50μM PTX . Patch pipettes were pulled from borosilicate glass using a P-97 micropipette puller ( Sutter Instrument , CA , USA ) , and had a resistance of 4–9MΩ . The intracellular patch solution contained 132mM K−Gluconate , 6mM KCl , 8mM NaCl , 10mM HEPES , 2mM Mg−ATP , 0 . 39mM Na−GTP , pH adjusted to 7 . 2–7 . 4 with KOH . Putative pyramidal or fast-spiking neurons were visualized using an infrared CCD camera ( Hamatsu ) on a BX50WI ( Olympus ) upright water immersion microscope . Somatic gigaseals ( > 2–4GΩ ) were made to visually identified cells within superficial layers . After break-through , intracellular membrane potentials were recorded in current-clamp mode ( Axopatch 200B , Axon Instruments ) , pre-amplified and low-pass filtered at 10 kHz ( Cyberamp 380 , Axon Instruments ) , and digitized at 25 kHz for voltage and 2 . 5 kHz for current traces using a CED 1401 ( Cambridge Electronic Design ) . We applied a step current that increased by 3 pA every 4 . 01 s to slowly drive neurons towards the tipping point at which they would start spiking . On top of this slowly increasing current we induced small perturbations to the membrane potential by injecting current pulses of 50 pA for 5 ms at 1800 ms time and 3805 ms on each step ( Fig . 1 a ) . The recovery after small perturbations allowed to measure recovery rates , the unperturbed segments before current pulses to estimate variance and autocorrelation from subthreshold voltage . Data were collected continuously with Spike2 ( CED ) and analyzed off-line . The recovery rate after each perturbation by 5 ms current injection of 50 pA was determined by fitting the 4800 sample ( corresponding to 192 ms at 25000 Hz sampling rate ) long segment of subthreshold voltage following the pulse current injection . Prior to fitting , the mean voltage of the segment was subtracted . The recovery of the voltage V ( t ) after each perturbation was fit by an exponential decay V ( t ) = a ∙ e − λ t + b ( 13 ) a , b , λ∈ℝ using the Python ( Python Software Foundation , version 2 . 6 ) function scipy . optimize . curve fit . For each perturbation , the recovery rate λ was then recorded together with the mean voltage over the 1 second interval prior the perturbation and the mean injected current I during the 1 second interval prior the perturbation for further analysis . The distance to the bifurcation point , ΔI , is then given by ΔI = Ic−I where Ic is defined as the average current injected during the one—second interval before the first spike . Note , that ΔI is directly related to y in our model systems . Variance was calculated from subthreshold voltage segments prior to each current pulse . For the results in the main part of the manuscript , segments of one—second duration were taken . Similarly , autocorrelation was calculated from subthreshold voltage segments of one—second duration prior to each current pulse . After subtraction of the mean we derived the autocorrelation function ACF ( s ) of a signal Fi ( t ) with length N , mean μ and variance v by A C F * ( s ) = ∑ t = 1 N − s ( F i ( t ) − μ ) ( F i ( t + s ) − μ ) v , s = 1 , … , N / 2 ( 14 ) and normalization by the first value A C F ( s ) = A C F * ( s ) A C F * ( 1 ) . For the analysis in the experimental part of the manuscript , we used the value of the autocorrelation function at lag 50 ms . We determined the power-law exponents governing the scaling for recovery rate , variance and autocorrelation by a linear fit in log-space . Specifically , logarithmic values of recovery rates λ , variance v and lag-50ms autocorrelation ( ACF50ms ) were fit individually for each trial as a function of the corresponding logarithmic values of ΔI by logx = A ∙ log Δ I + B ( 15 ) using the Python function scipy . optimize . curve fit . Here , x are the values of recovery rate , variance and autocorrelation , respectively , and A corresponds to the related exponent ( i . e . θ , τ or κ ) obtained in the fit . For the determination of exponents , we required fits to have R2 ≥ 0 . 1 . To determine a robust estimate of the exponent the fit values were calculated for different minimal values ΔImin , i . e . ΔI values smaller than ΔImin were discarded in the fit . The exponents given in the main text are the mean values over the different ΔImin . | Neurons efficiently convey information by being able to switch rapidly between two different states: quiescence and spiking . Such sudden shifts to a qualitatively different state are observed in many complex systems; the often dramatic consequences of these tipping points for diverse fields such as economics , ecology , and the brain have spurred interest to better understand their transition mechanisms and predict their sudden occurrences . By studying the transition from neuronal quiescence to spiking , we show that the quantitative scaling laws for critical slowing down , i . e . , a system’s tendency to recover more slowly from perturbations upon approaching its transition point , inform about the underlying bifurcation mechanism and can be used to improve the prediction of a system’s tipping point . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Critical Slowing Down Governs the Transition to Neuron Spiking |
RNAi is a ubiquitous pathway that serves central functions throughout eukaryotes , including maintenance of genome stability and repression of transposon expression and movement . However , a number of organisms have lost their RNAi pathways , including the model yeast Saccharomyces cerevisiae , the maize pathogen Ustilago maydis , the human pathogen Cryptococcus deuterogattii , and some human parasite pathogens , suggesting there may be adaptive benefits associated with both retention and loss of RNAi . By comparing the RNAi-deficient genome of the Pacific Northwest Outbreak C . deuterogattii strain R265 with the RNAi-proficient genomes of the Cryptococcus pathogenic species complex , we identified a set of conserved genes that were lost in R265 and all other C . deuterogattii isolates examined . Genetic and molecular analyses reveal several of these lost genes play roles in RNAi pathways . Four novel components were examined further . Znf3 ( a zinc finger protein ) and Qip1 ( a homolog of N . crassa Qip ) were found to be essential for RNAi , while Cpr2 ( a constitutive pheromone receptor ) and Fzc28 ( a transcription factor ) are involved in sex-induced but not mitosis-induced silencing . Our results demonstrate that the mitotic and sex-induced RNAi pathways rely on the same core components , but sex-induced silencing may be a more specific , highly induced variant that involves additional specialized or regulatory components . Our studies further illustrate how gene network polymorphisms involving known components of key cellular pathways can inform identification of novel elements and suggest that RNAi loss may have been a core event in the speciation of C . deuterogattii and possibly contributed to its pathogenic trajectory .
Genome reduction is a common adaptation among bacterial pathogens and commensals , and has been hypothesized to occur for a number of reasons , including increased specificity to a host or environmental range , or to increase virulence more directly through loss of an antivirulence gene or gene cluster . The former case can be explained primarily through loss of genes that play only accessory roles . These genes can become dispensable as an organism becomes obligately associated with a host , which then acts as an alternative source for these gene products , such as amino acids or metabolic intermediates [1–3] . In some cases , network polymorphisms can result from loss of one of the components , which then enables additional inactivating mutations to occur in other components of the crippled or disabled pathway , such as loss of the Gal80 repressor in Saccharomyces kudriavzevii [4] . Genes also can be lost as a result of an “antivirulence” function , as is seen in Shigella and E . coli , where the presence of the lysine decarboxylase cadA interferes with the synthesis of enterotoxins through production of cadaverine [5] . This model , termed the black hole hypothesis , suggests that gene losses can be the result of active interference with pathogenesis , likely as the result of gain of a new incompatible function . In either model , understanding the gene network polymorphism can elucidate the biology and evolution of the pathogen , facets that are particularly relevant for new and emerging pathogens . Cryptococcus deuterogattii , previously C . gattii molecular type VGII [6] , is an emerging human fungal pathogen in the Pacific Northwest ( PNW ) of the United States and southwest Canada [7–9] . While the sibling species C . neoformans predominantly infects immunocompromised individuals , many of the C . deuterogattii infected patients in the Pacific Northwest outbreak were otherwise healthy . Both species cause severe pulmonary and central nervous system infections , and are fatal if untreated . Surprisingly , whole genome sequencing revealed that the C . deuterogattii strain R265 is missing both of the Argonaute genes , essential components of the RNAi-induced silencing complex ( RISC ) [10 , 11] . Further examination revealed that in addition to the loss of both Argonaute genes , one of the two Dicers and the only RNA-dependent RNA polymerase have also undergone pseudogenization through large sequence losses similar to those of the Argonaute genes [12] . The loss of critical canonical components of the RNAi pathway raises a number of questions about the origins and biology of the C . deuterogattii species as well as the function of RNAi within the Cryptococcus pathogenic species complex as a whole . RNA interference ( RNAi ) is a highly conserved mechanism among eukaryotes that facilitates homology-dependent gene silencing . This transcriptional regulatory strategy was initially observed in Caenorhabditis elegans where exogenously introduced double-stranded RNA ( dsRNA ) triggers silencing of the transcript complementary to the dsRNA sequence [13] . Since its discovery in C . elegans , numerous species of plants , animals , fungi , and protists have been found to employ similar strategies to either protect their genomes from foreign DNA or to orchestrate gene expression and diverse cellular , developmental , and physiological processes [14–17] . Repetitive sequences are often found in mobile genetic elements and previous studies found an association between RNAi and transposable elements , which are ubiquitous in eukaryotic organisms . Transposon activation and movement impairs genome stability and increases the mutational burden of the host . Therefore , eukaryotes employ different strategies to inhibit and limit transposon expansion . Arabidopsis thaliana , Drosophila melanogaster , Saccharomyces castellii , Neurospora crassa , and C . elegans all utilize RNAi strategies to control and inhibit transposon expression [18–22] . C . neoformans also employs an RNAi-related pathway to inhibit transposable elements . In previous studies , Wang et al . showed that the insertion of a tandem multicopy transgene triggered a homology-dependent gene silencing mechanism during sexual development and termed this process sex-induced silencing ( SIS ) [10] . This process was identified specifically with a SXI2a-URA5 transgene array inserted into the ura5 locus , resulting in the presence of three functional copies of URA5 and one nonfunctional copy . During mating , progeny that inherit the array silence the URA5 gene in an RNAi-dependent manner approximately 50% of the time . In addition , Wang et al . later found that transgene silencing can also occur during vegetative growth , named mitotic-induced silencing ( MIS ) , but at a relatively lower frequency in mitotic ( ~0 . 2% ) compared to meiotic progeny ( ~50% ) [23] . Further analysis showed that SIS and MIS require the RNAi components Rdp1 ( RNA-dependent RNA polymerase ) , Ago1 ( Argonaute ) , and Dcr1/2 ( dicer-like proteins ) [10 , 23] . SIS and MIS function to inhibit transposon movement and thus serve as a genome defense mechanism during meiosis and mitosis . The initial observation of transposon silencing during sexual development was made in the highly virulent C . neoformans lineage . Later studies found that transgene-related SIS also occurs in C . deneoformans and that the RNAi components are required for transposon silencing during both bisexual and unisexual development [24] . The lack of the critical Argonaute , Dicer , and RdRp components of the RNAi pathway in C . deuterogattii suggests that the loss of RNAi may represent a gene network polymorphism . In fact , the RNAi pathway is intermittently conserved and lost across eukaryotes [12 , 25–27] . In Leishmania and trypanosomes , RNAi losses were previously taken advantage of in order to identify additional , previously unknown components of the RNAi pathway via comparative genomics [16] . To test the hypothesis that the RNAi pathway represents a gene network polymorphism , we surveyed the genomes of the R265 ( C . deuterogattii ) , WM276 ( C . gattii ) , H99 ( C . neoformans ) , and JEC21 and B-3501A ( C . deneoformans ) strains and found 14 genes missing from C . deuterogattii , including the canonical components of the RNAi pathway RDP1 , AGO1 , and DCR1 . Here we focus on four of these lost components: ZNF3 , previously identified as a regulator of hyphal development during unisexual and bisexual reproduction [28]; CPR2 , a G-protein coupled receptor ( GPCR ) previously studied for its role as an accessory constitutively active pheromone receptor [29]; QIP1 , independently identified as an RNAi component via a mass spectrometry approach [30]; and FZC28 , a putative transcription factor with no obvious phenotypes in a systematic genome-wide transcription factor deletion study [31] . Here we demonstrate that the loss of the RNAi components represents a bona fide system polymorphism , with several previously unknown RNAi components lost in C . deuterogattii . In addition , we show that mutants of these missing genes in C . neoformans fall into two classes: mutants that lose both vegetative silencing and sex-induced silencing , and mutants that are affected only in the frequency of sex-induced silencing . This suggests that sex-induced silencing may be a specialized , highly induced variant of the vegetative transgene-induced silencing pathway , rather than a separate pathway . Taken together , our results show that a substantial loss of genes contributing to two related RNAi pathways has occurred in C . deuterogattii . By using comparative genomics , these gene losses reveal key insights that aid in elucidating the functions of these RNAi-based genome conservation pathways .
The C . deuterogattii lineage ( previously VGII C . gattii ) is responsible for the recent , ongoing outbreak on Vancouver Island and its expansion into the Pacific Northwest of the United States . Initial analysis of the R265 C . deuterogattii reference genome revealed that both the key canonical RNAi components AGO1 and AGO2 are missing , indicating that the VGII lineage of C . deuterogattii may lack a functional RNAi pathway [10 , 11] . Upon further examination , we discovered that two of the other canonical components , DCR1 and RDP1 , had both suffered truncations removing key functional domains and are therefore pseudogenes . Of the known RNAi canonical components , only DCR2 remains intact in C . deuterogattii ( Fig 1A ) [11 , 32–35] . We hypothesized that this loss of multiple RNAi components may represent a gene network polymorphism where all of the components of a pathway are intact in one species , but have been selectively lost in another closely related species . We further hypothesized that a whole genome comparison of C . deuterogattii with other related Cryptococcus species would reveal novel components of the RNAi pathway lost in C . deuterogattii but otherwise maintained throughout the pathogenic species complex . We compared the publicly available reference genomes of JEC21 ( C . deneoformans ) [36] , B-3501A ( C . deneoformans ) [36] , H99 ( C . neoformans ) [37] , and WM276 ( C . gattii ) [11] with R265 ( C . deuterogattii ) [11] to identify otherwise conserved genes that were missing or truncated in the C . deuterogattii lineage . We found seven conserved genes that were not annotated in R265 and seven others that were dramatically shortened ( over 50% different in length ) as a result of extensive deletions of genomic sequence ( Table 1 ) . All 14 genes were lost across the entire VGII group , based on 53 publicly available whole genome sequences from C . deuterogattii [32] . These genome sequences did reveal some diversity in these regions . Estimation of Tajima’s D in windows across the genome and within the regions left by the deletion events showed a highly negative value for the genome as a whole ( mean of -1 . 122 ) , and a slightly more positive ( mean of -0 . 796 ) , but not statistically significant value ( p = 0 . 0901 ) for the deletion windows ( S1 Fig ) . We did not identify any transposable elements or repeats that may have mediated the deletion events . One of the seven missing genes was the previously identified canonical RNAi component AGO1 . In each case , localized deletions of sequence occurred , encompassing entire ORFs , start codons , and/or functional domains of the candidate genes ( Fig 1B–1H and S2 Fig ) . Our screen identified two potential transcription factors , FZC27 and FZC48 , and three genes , including GWC1 , GWO1 , and QIP1 , which have been previously identified as participating in the degradation of unspliced mRNA through RNAi [30] . Two of the 14 missing or truncated genes , CPR2 and ZNF3 , were previously shown to play roles in unisexual and bisexual reproduction , but were not described as having a role in RNAi [17 , 35] . We chose to focus on four genes as candidates to interrogate for a role in the SIS and MIS RNAi pathways: CPR2 , FZC28 , ZNF3 , and QIP1 . CPR2 encodes a seven transmembrane domain GPCR closely related phylogenetically to the Ste3 family of pheromone receptors , but it is constitutively active and independent of pheromone ligand binding [29] . Cpr2 signals via the same G proteins as the pheromone receptor Ste3 , and overexpression of CPR2 can rescue the sterility defect of ste3Δ mutants , although it may bias cells towards unisexual reproduction [29] . FZC28 is a transcription factor about which very little is known . It was identified and mutated as part of a genome-wide transcription factor deletion library , and experiments in that study identified no obvious phenotypes [31] . In previous studies we found that ZNF3 is required for hyphal development during unisexual and bisexual reproduction in C . deneoformans [28] . Deletion of the gene blocks hyphal development and impairs pheromone expression during mating . However , it does not play a direct role in the pheromone-signaling cascade . Surprisingly , microarray expression analysis revealed that deletion of Znf3 increased transposon and transposon-related gene expression during bisexual reproduction [28] . Znf3 is also somewhat rapidly diverging in amino acid sequence . While it is found in the Cryptococcus pathogenic species complex and the neighboring sensu stricto ( including C . amylolentus ) and sensu lato groups ( including C . heveanensis ) , the sequence is not well conserved , and it shares only weak homology over a 211 amino acid stretch ( 23% identity and 38% positive ) with the reciprocal best BLAST hit ortholog in Tremella mesenterica . The encoded protein in Cryptococcus neoformans contains three zinc finger domains , two predicted nuclear localization signals ( NLS ) , and a conserved coiled coil region , often involved in protein-protein interactions , as well as a putative ribonuclease conserved domain indicating that it may be involved in cleavage of RNA . QIP1 is named for N . crassa QIP , which functions during quelling and MSUD by binding to RISC and stimulating cleavage of the passenger strand of the duplex siRNA [38] . Moreover , a previous study directly implicated Qip1 in the transcriptional squelching of transposons and the degradation of mRNAs that have poorly spliced non-canonical introns [30] . Dumesic et al . localized Qip1 in the nucleus and showed that it physically interacts with Rdp1 as part of the Spliceosome-Coupled and Nuclear RNAi ( SCANR ) complex [30] . Analysis of N . crassa Qip revealed a conserved 3’-5’ exonuclease domain belonging to the DEDDh superfamily of nucleases , showing high similarity to the E . coli DNA polymerase III ε subunit [33] . Although , the C . neoformans Qip1 protein does not contain any detected conserved functional domains , it exhibits weak similarity to the helical domain of Class IIa histone deacetylases , which may suggest a role different than that of N . crassa Qip . In previous studies , Wang et al . found that a tandem multicopy insertion of a SXI2a-URA5 transgene triggered silencing of the URA5 gene during bisexual reproduction and vegetative growth in C . neoformans [10 , 23] . When F1 progeny were isolated from a cross between WT MATα URA5 ( H99α ) and MATa SXI2a-URA5 ( JF289 ) , ~25% were found to be uracil-auxotrophic despite the fact that all of them had intact copies of the URA5 allele . Further analysis revealed that ~50% of the progeny that inherited the SXI2a-URA5 transgene were uracil auxotrophic . Recent studies showed that the transgene induced silencing mechanism is activated efficiently during bisexual and unisexual reproduction ( SIS ) and less efficiently during vegetative growth ( MIS ) [23 , 24] . Deletion of RNA-dependent RNA-polymerase Rdp1 abolished transgene induced silencing during SIS and MIS in both C . neoformans and C . deneoformans . To investigate the role of the missing genes from R265 in silencing we generated deletion mutants in the JF289a isolate bearing the SXI2a-URA5 transgene ( derived from strain KN99a ) , and the congenic WT H99α strain . Two independent deletion mutants for each gene were isolated and analyzed . To determine the silencing efficiency of the mutants during sexual reproduction , unilateral ( one parent is mutant ) and bilateral ( both parents are mutants ) crosses were performed on MS media . We dissected random F1 spore progeny from each cross and these were tested for growth in the absence of uracil and genotyped for the presence of the SXI2a-URA5 transgene ( Fig 2A and S3 Fig ) . In unilateral matings with a deletion allele only present in one of the two parents , two meiotic progeny were ura- for qip1Δ ( out of 14 inheriting the array , ~14% ) , none were ura- for znf3Δ ( out of 18 inheriting the array , 0% ) , and three were ura- for cpr2Δ ( out of 22 inheriting the array , ~13 . 6% ) indicating significantly reduced silencing efficiency compared to WT ( Fig 2A and S4 Fig ) . These results suggest that all three components play a role in RNAi during sexual development . In contrast , the silencing efficiency of the SXI2a-URA5 transgene in the fzc28Δ , and fzc47Δ unilateral mutant matings was similar to WT ( ~50% ) ( Fig 2C ) . All of the ura- progeny carry an intact copy of the SXI2a-URA5 transgene , as verified by PCR . Previous studies showed that bilateral matings of all three canonical RNAi component mutants ( ago1Δ , dcr1Δ , rdp1Δ ) yielded ~20 fold fewer spores , with rdp1Δ mutants also demonstrating disorganized and atypical basidia , but with no effect on the sporulation efficiency of the spores that were produced [10] . Similarly , although deletion of ZNF3 severely impaired mating in C . deneoformans [28] , hyphal development during bisexual reproduction was similar to WT in C . neoformans znf3Δ mutants , albeit somewhat delayed . In contrast , in bilateral qip1Δ x qip1Δ mutant crosses we found that spore production was severely impaired and the few spores that were isolated failed to germinate , indicating that Qip1 is required for completion of the sexual cycle and may play a role in meiosis ( S4B Fig ) . On the other hand , deletion of RDP1 or ZNF3 did not affect sporulation efficiency . Deletion of ZNF3 in both parents completely abolished silencing , as none of the progeny that inherited the transgene were ura- ( S4A Fig ) . These results indicate that Znf3 is required for silencing during mating and deletion of the gene causes a severe SIS silencing defect , similar to rdp1Δ . Silencing of the URA5 gene was also impaired in fzc28Δ and cpr2Δ bilateral matings ( Fig 2C ) . However , fzc47Δ mutation in both parents did not impair silencing of the URA5 transgene and it was similar to WT , despite a modest increase in silencing rate in a unilateral cross ( Fig 2C and S4 Fig ) . We then examined the silencing frequency of the SXI2a-URA5 transgene in the mutant strains by measuring spontaneous 5-FOA resistance following mitotic growth in rich media . The strains bearing the qip1Δ and znf3Δ deletions failed to yield any colonies on 5-FOA media , indicating that these two genes are required for transgene-induced mitotic silencing ( Fig 3 ) . In contrast , deletion of two transcription factors , FZC28 and FZC47 , obtained from a recently reported systematic transcription factor deletion collection and crossed into the JF289 background [31] , and the GPCR CPR2 , did not alter the mitotic silencing frequency of the SXI2a-URA5 transgene compared to WT . In conclusion , we found that Znf3 and Qip1 are required for silencing during both MIS and SIS and deletion of the genes generates a phenotype similar to mutation of RDP1 , whose gene product is essential for RNAi function in C . neoformans . These results suggest that Znf3 and Qip1 are novel regulators or components of the RNAi pathway . In addition we found that a new transcription factor Fzc28 and the GPCR Cpr2 influence transgene-induced silencing specifically during sexual development , possibly coupling the sexual cycle with the RNAi pathway but likely not acting mechanistically during silencing itself . In a previous study we found that deletion of Znf3 in C . deneoformans activates transposon expression [28] and here we have shown that it is required for MIS and SIS . Recent studies revealed that transposable element expression increases during sexual reproduction and the components of the RNAi pathway maintain genome integrity through an efficient transposon silencing mechanism [10] . Deletion of RDP1 results in centromeric and telomeric retrotransposon overexpression during sexual development in C . neoformans [10] . We examined the transcript abundance of two transposons , Tcn1 and Tcn2 , in znf3Δ mutant crosses and found that abundance was dramatically increased , similar to rdp1Δ and ago1Δ mutant crosses ( Fig 4A ) . Deletion of QIP1 also yielded elevated levels of transposon transcript abundance , indicating that Qip1 also plays a major role in transposon quenching during sexual development ( Fig 4A ) . To further investigate the role of Znf3 in transposon silencing on a genome-wide scale , we performed a comparative transcriptome analysis of znf3Δ x znf3Δ and rdp1Δ x rdp1Δ crosses during sexual development and vegetative growth . Bilateral crosses of znf3Δ x znf3Δ and rdp1Δ x rdp1Δ mutants were incubated on solid V8 medium ( pH = 5 ) for 24 hours , as well as H99α x JF289a wild type crosses . RNA was isolated from the mating cultures , transcribed to cDNA , and hybridized to a C . neoformans genome microarray . Genome-wide expression analysis revealed that among the transcripts with altered expression level , the majority were increased in the znf3Δ mutant cross relative to WT during sexual development , indicating that Znf3 has a repressive role during sexual development . The few transcripts whose abundance was decreased in the znf3Δ and rdp1Δ crosses are involved in hypoxia , oxidation , ion channels , sugar transport , and possibly sporulation . During znf3Δ sexual development more than 80 independent microarray tags exhibited a twofold increase in abundance compared with the WT . Further analysis revealed that the majority of these tags correspond to sequences from hypothetical proteins or align to intergenic regions of the C . neoformans H99 genome . Alignment to a retrotransposon library [39] showed that almost all of the intergenic probes that were increased in znf3Δ mutants correspond to retrotransposon sequences found in multiple sites in the genome ( S3 Table ) . We found that these retrotransposons have long terminal repeats ( LTR ) and reside in the centromeric and telomeric regions of the chromosomes . In addition , most of the upregulated hypothetical proteins in znf3Δ x znf3Δ crosses were found to be RNA and DNA helicases , RNA-dependent DNA polymerases , and other transposon-related proteins ( S3 Table ) . During vegetative growth fewer transcripts were upregulated in znf3Δ mutants; however , the transcripts that exhibited differential abundance were also involved in transposon expression or activation . As was observed previously , the Tcn1 , Tcn2 , and Tcn3 elements were increased in znf3Δ × znf3Δ crosses , while their abundance was diminished during znf3Δ vegetative growth but remained significantly higher than the WT . We compared the transcriptional profile to the rdp1Δ x rdp1Δ mutant cross profile , and the whole genome transcript profiles between the two mutants were highly similar ( Fig 4B ) . The highly correlated transcript profiles of upregulated genes suggests that Znf3 and Rdp1 have similar functions and may mediate retrotransposon silencing through the same RNAi pathway . Interestingly , in spite of the loss of RNAi components in C . deuterogattii , transposon copy number does not appear to have dramatically increased in the genome ( S5 Fig ) . The vast majority of transposable elements are present at substantially lower copy number in C . deuterogattii compared to C . gattii . However , several classes of transposable elements are present in approximately equal amounts ( TCN3 and TCN6 ) or at even higher levels ( TCN4 and LTR13 ) in C . deuterogattii ( R265 ) than in C . gattii ( WM276 ) , based on a BLAST search using a C . neoformans library [39] . In previous studies we found that , although Znf3 regulates sexual development , ZNF3 expression remains stable during vegetative growth and mating in C . deneoformans [28] . In addition , mRNA levels for the RNAi components are relatively similar between mitotic growth and mating based on northern blot analysis; however , their protein abundance was significantly higher during sexual development suggesting that the RNAi components are translationally induced or stabilized during the sexual cycle [10] . Based on this evidence we hypothesized that ZNF3 and QIP1 expression might also remain the same between the two conditions in C . neoformans . RNA was isolated during mitotic growth and mating from WT and bilateral mutant crosses and the abundance of their transcripts was analyzed using quantitative RT-PCR . Unlike the canonical RNAi components , we found that both ZNF3 and QIP1 expression was significantly higher during mating compared to WT ( Fig 5A ) . This was a surprising result given that the expression of the highly conserved ZNF3 gene in C . deneoformans remains the same and similar to WT during both conditions [28] . Moreover , Znf3 and Qip1 have similar roles with the RNAi components in SIS and MIS whose expression remains stable . This indicates that Znf3 and Qip1 expression may have a unique mode of regulation distinct from Rdp1 and Ago1 . We next assessed whether the RNA abundance during sexual development is correlated with the protein level between the two conditions . The C-termini of Znf3 and Qip1 were fused with mCherry at the endogenous genomic loci . MIS and SIS assays were conducted to test if the chimeric proteins retain their functional roles in silencing . Znf3 tagged with mCherry was completely defective in SIS and MIS , indicating that the mCherry tag interferes with function . On the other hand , Qip1 tagged with mCherry exhibited wild type levels of silencing during vegetative growth and sexual development . The protein levels were examined during both conditions and we found that , although the Qip1 protein was present during both vegetative growth and mating , it was significantly more abundant during sexual development , similar to the difference observed in RNA abundance ( Fig 5B ) . These results indicate that Qip1 , and possibly also Znf3 , have a unique mode of regulation that is possibly distinct from that of other RNAi components . The MIS and SIS silencing phenotypes of znf3Δ and qip1Δ mutants are very similar to rdp1Δ mutants . Previous studies have suggested that an unknown RNA-binding factor may govern translational regulation of the transcripts of the RNAi components to result in elevated protein levels specifically during sexual development [10] . We found that Znf3 bears both zinc fingers and an RNase domain and transcription of the gene is sexually induced . Considering that Znf3 has a similar phenotype to Rdp1 , it could be involved in the translational regulation of the RNAi components , and the severe loss of silencing phenotype of znf3Δ mutants might be attributable to an absence of these factors . It is unlikely that Znf3 regulates the transcription of the RNAi components based on microarray expression analysis . Nevertheless , we performed quantitative RT-PCR in the absence of each of the RNAi components during sexual development . Surprisingly , we observed a modest increase in the expression of the RNAi components during sexual development compared to vegetative growth ( Fig 5C ) . In previous studies , northern blot analysis was employed to investigate the expression of these genes during vegetative growth and mating , and the modest 2- to 4-fold increase we observed using RT-PCR was possibly below the level of detection by northern blot . However , deletion of ZNF3 and QIP1 did not alter the expression of RDP1 or AGO1 , suggesting that Znf3 and Qip1 do not act as transcriptional regulators of the canonical RNAi components or mediate the modest increase in expression we observed in mating conditions We also investigated the expression of ZNF3 and QIP1 in the absence of the canonical RNAi components during sexual development . Deletion of RDP1 did not affect the ZNF3 transcript levels during sexual development , indicating that RDP1 does not control the expression of this gene ( Fig 5C ) . The expression of Znf3 was modestly but significantly increased in the ago1Δ mutants , which is the catalytic subunit of the RISC complex . Interestingly , expression of QIP1 during sexual development decreased to vegetative levels in the absence of RDP1 . To explore a possible role of Znf3 in the translational regulation of the RNAi components , we deleted ZNF3 and investigated the protein levels of Ago1 and Rdp1 fused with mCherry under the control of the endogenous promoter during sexual development . We detected a strong protein signal for both Ago1-mCherry and Rdp1-mCherry during sexual development with or without ZNF3 ( Fig 5D ) . Western blot analysis revealed that deletion of ZNF3 resulted in a modest decrease in the protein abundance of Rdp1 under mating conditions ( Fig 5E ) . It is possible that this decrease may not have been detectable via direct microscopy of cells expressing the Rdp1-mCherry fusion protein ( Fig 5D ) . These results indicate that although Znf3 , is not involved in transcriptional regulation of the canonical RNAi components , it could be involved in either translational regulation or in modulating protein stability via a role as a scaffolding protein . RNAi silencing is a multifunctional pathway and different steps occur at different sites within the cell . The presence of tandem repeated genes or retrotransposons in the genome induces the transcription of aberrant ssRNA in the nucleus through an unknown mechanism and Rdp1 generates dsRNA from these sequences and evokes the RNAi pathway . The dsRNA travels to P-bodies , where processing and RNA silencing occurs . Dcr1/2 and Ago1 , which localize to P-bodies , generate siRNAs that target mRNAs with complementary sequences for degradation [10] . These findings suggest that additional components of the pathway will localize either to the nucleus or to P-bodies . Znf3 has two NLS signals , therefore we initially hypothesized that Znf3 might localize to the nucleus where it could act as a transcription factor , or bind and degrade dsRNAs generated by Rdp1 . To investigate the localization of Znf3 , and because endogenous C-terminal tagging had failed to produce functional protein , the N-terminus of the protein was fused to mCherry , and expressed from the constitutively active GPD1 promoter . The H99α and JF289a strains were transformed with the mCherry-Znf3 plasmid and evaluated by direct fluorescence microscopy . Surprisingly , we observed multiple bright foci in the cells indicating that the protein was present in more than one cellular compartment during sexual development ( Fig 6A ) . To determine this cellular localization , we utilized two established marker components , one for P-bodies and the other for the nucleus . Dcp1 , found in P-bodies , is responsible for decapping mRNAs during exonucleolytic degradation , while Nop1 is a component of the small subunit processome ( a ribosome assembly intermediate ) complex of the nucleolus [10 , 40] . GFP-Dcp1 and GFP-Nop1 were expressed from plasmids that were ectopically introduced into the genomes of strains expressing the mCherry-Znf3 protein and localization was observed during vegetative growth and sexual development . Surprisingly , we found that Znf3 localizes only in the P-bodies during both vegetative growth and sexual reproduction , despite the putative NLS signals ( Fig 6A ) . These results suggest that Znf3 may participate directly in the RNAi silencing process and it may represent a novel element of the RNAi pathway . Previous studies found that Qip1 localizes in the nucleus and that it physically interacts with Rdp1 and Ago1 during vegetative growth [30] . Although Ago1 is primarily localized in P-bodies during mating , where RNA silencing occurs , it has been also reported in the nucleus under vegetative growth conditions [30] . To further investigate the localization of Qip1 , the protein was fused at the C-terminus with mCherry and expressed from the endogenous QIP1 promoter . The fluorescent signal was evaluated via microscopy during vegetative growth and sexual development . Co-localization of Qip1-mCherry with GFP-Dcp1 or GFP-Nop1 revealed surprising results . During sexual development , where the RNAi pathway is highly induced , Qip1 was localized exclusively in P-bodies , potentially reflecting a role in RNA degradation ( Fig 6B ) . During vegetative growth we observed Qip1 in association with either the P-bodies or the nucleus ( Fig 6B ) . These results suggest that Qip1 may interact with Rdp1 in the nucleus during vegetative growth , possibly as a component of the SCANR complex to participate in the processing of the stalled splicing intermediate [30] . During mating Qip1 migrates to the P-bodies where it may subserve its conserved role in the RISC complex . We observed that both cpr2Δ and fzc28Δ mutants had defects in SIS but not in MIS . This suggests that these two pathways may differ in more than just their efficiency . As a result , we sought to test whether the role of Cpr2 in SIS was linked to its role in mating or independent of this function . We tested this by analyzing mutants lacking Ste3α , a pheromone receptor that shares the same G proteins as Cpr2 . Ste3α mutants fail to mate , so a deletion was instead constructed in an a/α diploid ( Fig 7A ) . FACS was used to verify that two independent ste3αΔ/a deletions remained stably diploid ( Fig 7B ) . The ste3αΔ/a diploids were then sporulated and the progeny were dissected and tested for silencing of the URA5 transgene . Both independent mutants demonstrated a defect in SIS , with a silencing frequency of only ~21–22% compared to 50% silencing in WT crosses . To test whether this effect was mediated by the shared downstream MAP kinase cascade ( Fig 7D ) , we utilized a ste3αΔ mutant complemented with an overexpressed CPR2 gene under the control of the GPD1 promoter to test whether ectopic overexpression of Cpr2 could compensate for the ste3αΔ mutant defect in SIS . This strain was mated with the JF289a SXI2a-URA5 transgene array strain and spores were dissected , germinated , and phenotyped ( Fig 7E and 7F ) . Overexpression of Cpr2 restored the SIS efficiency of the ste3αΔ/a mutant to ~ 67–68% . This suggests that both Cpr2 and Ste3 act coordinately to induce the RNAi pathway during the sexual cycle , and may not act as essential RNAi components themselves . Similarly , the transcription factor Fzc28 is a candidate to be the downstream effector of the Ste3/Cpr2 pathway , as it has no MIS defect but an absolute defect in SIS , suggesting it is essential for that arm of the pathway .
In this study we found that four novel proteins are required for silencing of the SXI2a-URA5 transgene during sexual development ( SIS ) and/or vegetative growth ( MIS ) , and that they fall into two distinct classes: proteins essential for both RNAi pathways , and proteins influencing just the sex-induced arm of the pathway . Deletion analysis reveals that Znf3 and Qip1 are required for MIS and SIS RNAi silencing , similar to the canonical RNAi component Rdp1 , while Cpr2 and the novel transcription factor Fzc28 influence only in SIS . In further support of this model we found that rdp1Δ and znf3Δ mutants have very similar transcript profiles during mating characterized by increased abundance of messages from retrotransposons and other transposon-related genes . However , unlike RNAi components , whose expression remains largely stable during mitotic growth and mating , ZNF3 and QIP1 are transcriptionally induced during the sexual cycle . Although Znf3 has two NLS tags , we found that it localizes in P-bodies , where Dcr1/2 and Ago1 are also localized . Qip1 also localizes in P-bodies during sexual development; however , in some cases it migrates to the nucleus during vegetative growth . This may indicate that these two proteins act mechanistically in the silencing pathway , rather than as regulators of the canonical components . RNA silencing is a highly conserved mechanism of transcriptional regulation . Since its discovery in C . elegans it has been identified in numerous species throughout the eukaryotic kingdom and it is hypothesized to be an ancestral feature of the last common eukaryotic ancestor [13 , 14 , 41] . An RNAi-related phenomenon was initially identified in plants and fungi , and later multiple species have been found to undergo RNA silencing mechanisms , including the fungi Neurospora crassa , Mucor circinelloides , and Schizosaccharomyzes pombe [20 , 42–44] . However , RNAi has been independently lost in some species , such as Saccharomyces cerevisiae and Ustilago maydis , which are missing all of the components of the RNAi pathway [18 , 25] . Nevertheless , the closely related species of S . castellii and C . albicans retain some of the RNAi components and substitute for the absence of others by employing noncanonical factors to produce dsRNA and shRNAs that map to transposable elements [45] . As a result , it is possible to learn more about the intact RNAi pathway of a species by comparing it to a related species that has lost some or all of its RNAi components [16 , 26] . This study shows that examining individual cases of RNAi loss with available sequenced genomes for closely related organisms is likely to be fruitful . The Ustilago clade is another basidiomycete example to which this approach could be applied [27] . In this study we identified a number of novel RNAi components , three of which , Cpr2 , Znf3 , and Fzc28 , had no functional domains or similarity to a known RNAi component that would have suggested they might be involved in an RNAi pathway . Notably , there are two known canonical RNAi components that we did not identify using this approach: Ago2 and Dcr2 . In the first case , this is because Ago2 is not conserved across all of the Cryptococcus RNAi-proficient genomes , as it is missing from the H99 Cryptococcus neoformans reference genome . In the second case , Dcr2 is retained in C . deuterogattii , which may suggest either that Dcr2 already has an additional non-RNAi role in Cryptococcus , that it has acquired a second role during the loss of RNAi and speciation of C . deuterogattii , or that the entire RNAi pathway has not been lost . As a precedent for the first two hypotheses , in C . albicans , a noncanonical Dicer plays a role in snRNA processing [46] . Also notable is that our screen identified components that play a role in only the sex-induced pathway but not in mitotic silencing . Loss of these components suggests one of two hypotheses regarding the RNAi loss: either loss of RNAi began with the inactivation of SIS , and without SIS the evolutionary pressure to maintain the MIS pathway was no longer strong enough to prevent loss of core components , or alternatively , loss began with the core mechanistic components involved in the vegetative arm , and the pressure to maintain the specialized regulatory machinery of SIS vanished , allowing loss of both CPR2 and FCZ28 . The latter case seems potentially more likely , as the loss of canonical RNAi components can allow relatively robust transposon movement even without undergoing the sexual cycle [30 , 47] . In addition , the loss of these SIS-specific components also provides an opportunity to elucidate the signaling processes connecting mating to induction of the RNAi pathway . It is also interesting that the Cryptococcus species that has lost RNAi , C . deuterogattii , is the species causing an ongoing outbreak in the Pacific Northwest . The loss of RNA silencing is possibly associated with higher virulence in this strain , but because C . neoformans or C . deneoformans strains missing RNAi elements are not altered in virulence in a murine host [10] , the loss of RNAi in the C . deuterogattii lineage may have instead had a longer-term impact on virulence trajectory . Indeed , loss of RNAi liberates transposons in Cryptococcus [10 , 30 , 47] , which could provide adaptive benefits through the generation of increased genetic and phenotypic diversity . We showed herein that Znf3 and Qip1 influence transgene induced silencing during mitotic growth and sexual reproduction . Deletion of these genes severely impaired silencing efficiency , even in unilateral crosses where only one parent was mutant . This phenotype is similar to rdp1Δ mutations , which abolish silencing during unisexual mating and impact silencing efficiency in unilateral bisexual crosses [10] . Given that Rdp1 is a major component of RNAi silencing , and that it is responsible for the initial steps generating dsRNAs , we propose that Znf3 also plays an important role in the pathway . Although the RNAi components Dcr1/2 and Ago1 are required for SIS , their deletion in a unilateral cross only lowers the silencing efficiency , indicating that their role may be redundant or largely complemented when one wild type nucleus is still present . Therefore , it is possible that Znf3 interacts with these proteins and may participate in the formation of the RISC complex . Znf3 is a large protein ( ~1515 aa ) , and so it could act as a scaffold to bring components of the RNAi pathway together in complex with the dsRNA substrate . Localization of Znf3 in P-bodies , where Dcr1/2 and Ago1 process the dsRNAs , may further support this hypothesis . Further , we provided evidence that Znf3 may play a role in protein translation or protein stability , which could be through a role as a scaffold in the P-bodies , or even as an RNA-binding protein through its zinc finger motifs . Dumesic et al . showed that Qip1 localizes in both the cytoplasm and the nucleus during growth in rich media , and that it physically interacts with both Rdp1 and Ago1 [30] . We further confirmed that Qip1 localizes in the nucleus during vegetative growth; however , we showed that during sexual development Qip1 expression is highly induced and it is localized in P-bodies during sexual development . Possibly , Qip1 enhances the function of Rdp1 in the nucleus during vegetative growth , by participating in the generation of aberrant dsRNAs from repetitive loci . During sexual development , where transposon movement is highly induced , Qip1 may resume its conserved role in RNA processing in P-bodies , where it may interact with Ago1 and participate in the cleavage of the passenger strand . Interestingly , Qip1 appears to play a role in meiosis , as bilateral mutant crosses failed to yield recombinant progeny . Lee et al . showed that Qip is also essential for meiotic silencing and meiosis in N . crassa [48] . However , the two proteins are significantly different and C . neoformans Qip1 does not contain a canonical exonuclease domain , unlike Qip . This may suggest that either Qip1 plays a different mechanistic role in C . neoformans than Qip in N . crassa , despite a similar phenotypic outcome , or that Qip1 may contain an unrecognized exonuclease domain . Sex-induced silencing is an efficient mechanism that protects the genome against mobile elements . Previous studies showed that ~5% of the Cryptococcus genome consists of transposons that cluster together in blocks and reside in both telomeric and centromeric regions on the chromosomes [36] . Transposon activation and movement may drive genome instability and phenotypic variation . Wang et al . found that transposons are transcriptionally induced specifically during sexual development , but they are silenced post-transcriptionally by the RNAi pathway [10] . These results suggest that transposons are derepressed during mating , which could increase the mutational burden of the progeny unless counteracted by the SIS RNAi pathway . During unisexual reproduction this mechanism could generate de novo genotypic and phenotypic variation in clonal populations and enable rapid adaptation to new environments . Thus , loss of the RNAi components may confer a beneficial advantage in clonal mitotic or sexual populations .
The strains and plasmids used in this study are listed in S1 Table . The strains were maintained in glycerol stocks at -80°C and grown on rich YPD media at 30°C ( Yeast extract Peptone Dextrose ) . Strains with selectable markers were grown on YPD containing nourseothricin ( NAT ) and/or G418 ( NEO ) . Uracil auxotrophic isolates were tested on both SD medium lacking uracil and synthetic medium containing 5-FOA ( 1 g/l ) . Mating assays were performed on 5% V8 juice agar medium ( pH = 5 for C . neoformans or pH = 7 for C . deneoformans ) or on MS ( Murashige and Skoog ) medium minus sucrose ( Sigma-Aldrich ) . The mating cultures were incubated in the dark at room temperature for 1 week . To visualize and isolate spores , strains of interest were co-cultured on solid V8 medium for 2 weeks at room temperature in the dark without parafilm . Basidiospores from the edges of the colonies were randomly isolated using a microdissection microscope equipped with a 25-μm microneedle ( Cora Styles Needles ‘N Blocks , Dissection Needle Kit ) as previously described [49] . Following germination the colonies were tested on YPD , YPD + NAT or NEO , SD-ura , and 5-FOA media . Genomic DNA was isolated using a CTAB protocol as previously described [50] . The presence of the SXI2a-URA5 transgene in the progeny was assessed by PCR using the primer pair JOHE16835/JOHE16836 . Example gel images can be found in S3 Fig for the unilateral crosses depicted in Fig 2 . The gene of interest was disrupted using a standard overlap PCR approach described previously [17] . Briefly , the 5’ and 3’ flanking regions of the ZNF3 , QIP1 , and CPR2 genes were amplified from H99α genomic DNA , and the selectable markers NAT and NEO were amplified from plasmids pAI3 and pJAF1 , respectively . The flanking sequences and the selectable markers were used to generate a full-length deletion cassette in an overlap PCR reaction with the flanking primers . The deletion cassettes were introduced into the H99α and JF289a strains by biolistic transformation [51] . Gene replacement via homologous recombination was confirmed by PCR and Southern hybridization . The primers used to generate the deletion mutants are listed in S2 Table . All deletions were constructed from at least two independent cultures , inoculated from different single isolated colonies of the parent strain , and independent mutants were isolated from different transformations . For the fzc47Δ and fzc28Δ mutants , two independent deletions were available in the KN99 background from a recent deletion collection [31] . They were crossed with JF289a and spores were dissected to isolate segregants that inherited the deletion , the transgene array , and were MATa . These segregants were named strains SEC5 , SEC6 , SEC7 , and SEC8 . To determine the cellular localization of the Znf3 protein , the mCherry protein was fused to the C-terminus of the protein under the control of the endogenous promoter using a standard overlap PCR approach . Briefly , 1 kb of sequence upstream of the start codon and 1 kb downstream of the stop codon were amplified from the wild type strain H99 genomic DNA using primers listed in S2 Table . The mCherry sequence fused with the NEO marker was amplified from plasmid pLKB25 . The flanking sequences and the fluorescence marker were combined as a template for an overlap PCR reaction . The overlap PCR products were introduced into the H99α and JF289a strains by biolistic transformation . Transformants were analyzed by PCR and Southern hybridization . To construct a plasmid encoding mCherry-Znf3 , the mCherry protein was fused to the N-terminus of Znf3 under the control of the constitutively active GPD1 promoter . The genomic sequence of ZNF3 was amplified from H99α genomic DNA using primer pair JOHE37890/JOHE37891 and cloned into plasmid pLKB49 [52] digested with XbaI and PacI . Plasmid pMF81 was introduced into the H99α and JF289a strains by biolistic transformation and the transformants were screened by PCR and direct fluorescence microscopy . Stable transformants expressing mCherry-ZNF3 and QIP1-mCherry were also transformed via ectopic insertion with the pXW11 ( GFP-DCP1 ) and pSL04 ( GFP-NOP1 ) plasmids to visualize P-bodies and the nucleus , respectively . To visualize mCherry-Znf3 and Qip1-mCherry together with GFP-Dcp1 and GFP-Nop1 , the strains of interest were grown on YPD medium to determine localization during vegetative growth or mixed with the opposite mating-type strain on V8 medium for 24 hours to visualize the proteins during sexual reproduction . Briefly , cells were grown overnight in liquid YPD and washed with PBS . Cells were then counted and mixed in equal proportions of MATa and MATα and spotted on V8 pH5 medium . These plates were incubated in the dark at room temperature for 24 hours . Cells were then scraped from the plates and mixed into sterile water and placed on prepared slides covered with 1 . 5% water agar . Imaging was performed with a Zeiss Axio Imager widefield fluorescence microscope at the Light Microscopy Core Facility at Duke University . Analysis was performed using the Metamorph Premier software package . For vegetative growth the cells were grown in 5 ml liquid YPD overnight at 30°C . The following day the cells were harvested , washed , frozen in liquid nitrogen , and lyophilized . Samples were kept at -80°C until analysis . To isolate RNA from mating assays , the desired α and a strains were grown in YPD liquid , washed with sterile water , mixed in equal amounts in eppendorf tubes , and a 1 ml suspension was spotted on V8 agar pH = 5 and incubated in the dark at room temperature for 24 hrs . The next day the mating cultures were harvested , washed with sterile water , frozen , lyophilized , and stored at -80°C . Total RNA was extracted using the RiboPure-Yeast Kit ( Ambion ) following the manufacturer’s instructions ( Life Technologies #AM1926 ) . Denaturing agarose gel electrophoresis and NanoDrop were used to assess quality and concentration of the RNA samples . The RNA was amplified using the Ambion® MessageAmp™ Premier RNA Amplification kit following the manufacturer’s instructions . cDNA was synthesized using AffinityScript reverse transcriptase ( Stratagene ) , Cy3/Cy5 labeled , and hybridized to a C . neoformans microarray slide ( Cryptococcus Community Microarray Consortium , Washington University , St . Louis , MO ) . Labeling and hybridization were conducted in the DNA Microarray Core Facility at Duke University . The slides were washed , scanned with a GenePix 4000B scanner ( Axon Instruments ) , and processed with GenePix Pro ( version 4 . 0 ) . All microarrays were conducted in triplicate . GeneSpring software was used for statistical analysis employing Lowess normalization , reliable gene filtering , and ANOVA analysis ( significance p<0 . 05 ) . Published reference genomes from H99 , B3501A , JEC21 , and WM276 were compared with R265 using FungiDB ( http://fungidb . org/fungidb/ ) [11 , 36 , 37 , 53] . Genes were selected that were present in all four of the non-C . deuterogattii reference genomes but absent or greater than 50% different in length in the R265 genome . Positive hits were manually examined using synteny maps produced by FungiDB in order to confirm that orthologs had been correctly identified . The majority of the initial hits were false positives attributable to sequencing gaps or incorrect gene annotation . For the remaining hits , sequences from all five reference genomes were manually aligned using Clustal [54] to verify that sequence deletions had occurred . In addition , de novo assemblies of 53 VGII genomes from previously published data were used to validate that deletions of all of the components were conserved in the VGII lineage and not restricted to the R265 genome [32 , 55] . Estimation of Tajima’s D was performed using a custom Perl script and the Bio::PopGen:Statistics package of BioPerl [56] . Briefly , genomes were aligned and SNPs were called as previously described [32] . To avoid sampling bias , a representative genome was chosen from each clonal expansion in the sequencing dataset for a total of 17 individual lineages out of the original 53 genomes . SNPs were sampled over a range using VCFTools [57] , alternate references were constructed using GATK [58] , and regions were aligned using ClustalW [54] . These alignments were imported by Bio::AlignIO and sampled using Bio::PopGen:Statistics [59] within BioPerl . To sample the genome as a whole , 2 kb windows every 10 kb throughout the genome were chosen . Locations with missing data for an individual were discarded , resulting in a total of 1159 data points . | Genome instability and mutations provoked by transposon movement are counteracted by novel defense mechanisms in organisms as diverse as fungi , plants , and mammals . In the human fungal pathogen Cryptococcus neoformans , an RNAi silencing pathway operates to defend the genome against mobile elements and transgene repeats . RNAi silencing pathways are conserved in the Cryptococcus pathogenic species complex and are mediated by canonical RNAi components . Surprisingly , several of these components are missing from all analyzed C . deuterogattii VGII strains , the molecular type responsible for the North American Pacific Northwest outbreak . To identify novel components of the RNAi pathways , we surveyed the reference genomes of C . deuterogattii , C . gattii , C . neoformans , and C . deneoformans . We identified 14 otherwise conserved genes missing in R265 , including the RDP1 , AGO1 , and DCR1 canonical RNAi components , and focused on four potentially novel RNAi components: ZNF3 , QIP1 , CPR2 , and FZC28 . We found that Znf3 and Qip1 are both required for mitotic- and sex-induced silencing , while Cpr2 and Fzc28 contribute to sex-induced but not mitosis-induced silencing . Our studies reveal elements of RNAi pathways that operate to defend the genome during sexual development and vegetative growth and illustrate the power of network polymorphisms to illuminate novel components of biological pathways . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
]
| [
"biotechnology",
"cryptococcus",
"neoformans",
"medicine",
"and",
"health",
"sciences",
"rna",
"interference",
"cryptococcus",
"pathology",
"and",
"laboratory",
"medicine",
"fungal",
"genetics",
"gene",
"regulation",
"pathogens",
"regulatory",
"proteins",
"microbiology",
"dna-binding",
"proteins",
"developmental",
"biology",
"fungi",
"plant",
"science",
"genetic",
"elements",
"transcription",
"factors",
"plant",
"genomics",
"epigenetics",
"morphogenesis",
"fungal",
"pathogens",
"mycology",
"genetic",
"interference",
"proteins",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"plant",
"genetics",
"comparative",
"genomics",
"biochemistry",
"rna",
"fungal",
"genomics",
"nucleic",
"acids",
"sexual",
"differentiation",
"genetics",
"transposable",
"elements",
"biology",
"and",
"life",
"sciences",
"genomics",
"mobile",
"genetic",
"elements",
"plant",
"biotechnology",
"computational",
"biology",
"organisms"
]
| 2016 | Gene Network Polymorphism Illuminates Loss and Retention of Novel RNAi Silencing Components in the Cryptococcus Pathogenic Species Complex |
The basal nucleus of the amygdala ( BA ) is involved in the formation of context-dependent conditioned fear and extinction memories . To understand the underlying neural mechanisms we developed a large-scale neuron network model of the BA , composed of excitatory and inhibitory leaky-integrate-and-fire neurons . Excitatory BA neurons received conditioned stimulus ( CS ) -related input from the adjacent lateral nucleus ( LA ) and contextual input from the hippocampus or medial prefrontal cortex ( mPFC ) . We implemented a plasticity mechanism according to which CS and contextual synapses were potentiated if CS and contextual inputs temporally coincided on the afferents of the excitatory neurons . Our simulations revealed a differential recruitment of two distinct subpopulations of BA neurons during conditioning and extinction , mimicking the activation of experimentally observed cell populations . We propose that these two subgroups encode contextual specificity of fear and extinction memories , respectively . Mutual competition between them , mediated by feedback inhibition and driven by contextual inputs , regulates the activity in the central amygdala ( CEA ) thereby controlling amygdala output and fear behavior . The model makes multiple testable predictions that may advance our understanding of fear and extinction memories .
In classical fear conditioning an animal learns to associate an initially neutral stimulus ( the conditioned stimulus , CS ) with an aversive stimulus ( the unconditioned stimulus , US ) after paired exposure to the CS and the US . Subsequent repeated non-reinforced presentations of the CS alone result in a decline of the conditioned response , a process called fear extinction [1] . Fear extinction is a highly context-dependent process: the conditioned fear response returns when the animal is exposed to an extinguished CS outside the extinction context [2] , [3] . Studies over the last decades have identified the amygdaloid complex as a key brain structure involved in both fear conditioning and extinction [4]–[6] . In the lateral nucleus of the amygdala ( LA ) , signals carrying information about the CS and the US converge onto the same neurons where they become associated through activity-dependent plasticity mechanisms [7]–[9] . The LA can directly or indirectly influence activity in the central nucleus ( CEA ) [10] , the major output nucleus of the amygdala that can trigger fear responses via its projections to the hypothalamus and to the brainstem [11] . The basal nucleus of the amygdala ( BA ) has been suggested to play an important role in contextual fear conditioning [12] , [13] , cued fear conditioning [14] , fear extinction [15]–[17] and context-dependent fear renewal [17] . Recently , two distinct fear and extinction specific neuronal sub-populations in the BA have been identified [17] . The balance of activity between fear and extinction neurons was correlated with states of high and low fear , respectively . Moreover , pharmacological inactivation of the BA blocked the acquisition of fear extinction and context-dependent fear renewal , suggesting that BA fear and extinction neurons may underlie the induction of behavioral changes and contribute to the formation of fear and extinction memories . These findings raise the question of what the potential mechanisms underlying the differential activation of these two neuronal sub-populations are . Here , we used a modeling approach based on in vivo physiological data to address this specific question and to draw more general conclusions on potential neural mechanisms involved in fear and extinction memories in the BA . In vivo stimulation of identified fear and extinction neurons revealed that the two neuronal populations receive differential functional input from the hippocampus and from the medial prefrontal cortex ( mPFC ) [17] . This finding could reflect anatomical specificity of inputs and/or selective functional plasticity of non-specific inputs . Independently of these two possibilities , in our model , we assume that anatomically and/or functionally distinct inputs from the hippocampus or the mPFC modulate the activity of BA fear and extinction neurons in a context-specific manner . That is , sub-populations of BA neurons are innervated by hippocampus/mPFC efferents that represent the current context . In addition , all BA neurons receive inputs from US/CS responsive LA neurons during conditioning and extinction . Those sub-populations of BA neurons that receive simultaneous LA and context-specific inputs become responsive during conditioning or extinction and , thus , emulate the “fear” and “extinction” neurons reported by Herry et al . [17] . Activation of BA neurons per se , however , is not sufficient to cause or prevent a behavioral response , but the selective activation of BA neurons conveys important information about the context-CS relation to the CEA . Although we do not model here the CEA , we stipulate that context-dependent BA activity provides an instructive signal to CEA neurons . In the CEA , it is likely that conditioning [18] and possibly extinction learning-induced changes act upon this signal in order to activate or suppress a fear response . If more experimental data , sufficient to constrain the possible parameter space , become available , then our present model of the BA could be extended to study the impact of context-dependent BA activity on learning-induced changes in the CEA as well . We test the plausibility of context-dependent activation of BA neurons in two different approaches: first , in an abstract firing rate model; second , in a more realistic spiking neuron network ( SNN ) model of the BA . Based on the results of our model we provide plausible explanations for several experimental observations in fear and extinction learning and make specific , experimentally testable predictions .
The description of the evolution of the firing rates of BA neurons during fear conditioning and extinction reported by [17] provide certain simple , yet important , indications on the underlying dynamics in the BA network: To test the feasibility of the above observations and their inferences in explaining the emergence of fear and extinction neurons in BA , we first studied the dynamics of a mean-field ( or firing rate ) model of the BA . Subsequently , we constructed a spiking neuron network ( SNN ) model to examine our hypotheses and their implications under more realistic conditions . The mean-field model of BA consisted of two neuron populations , A and B , described by Wilson-Cowan type rate dynamics [19] ( Fig . 2A ) . Both populations were identical in their properties ( Eqs . 1–2 ) and received both CS input and non-specific background input . There is ample experimental evidence that in different contexts , different sets of hippocampal neurons ( e . g . in CA1 ) are active [20]–[22] . Thus , to mimic context-specific inputs - either directly from hippocampus or indirectly via the mPFC or other brain structures such as entorhinal cortex - we provided population A with additional input reflecting , and likewise , population B with additional input reflecting . Populations A and B were mutually interconnected with inhibitory synapses . The system of differential equations describing the activity of the populations A and B is as follows: ( 1 ) ( 2 ) where . The evolution of the connection strengths is given by ( 3 ) ( 4 ) Here , represents the connection strength from population ( or external input ) Y to population X , is the time constant governing the dynamics of population X , kX is the maximum firing rate of population X , and rX captures the refractoriness of neurons in X . The transfer function S is a sigmoid function , integrating all inputs to population X in a non-linear fashion and producing a bounded output rate . The parameters p and θ of the sigmoid function determine the steepness and the position of its maximum slope , respectively . The term η ( t ) , with zero mean , reflects the stochastic input to the two populations , mimicking the background activity in the BA . Equations 3 and 4 describe the dynamics of the connection strengths of the CS afferents onto populations A and B respectively . These weights were increased in an additive way whenever the respective CS and CTX inputs were present simultaneously and remained constant otherwise . The parameters aA and aB specify the learning rates ( see also Eqs . 6–8 ) . We simulated fear conditioning and extinction by applying CS input to both populations in the form of short pulses of 50 ms duration each , based on the experimental design used in [17] . Contextual input was provided continuously . Note that we did not make any explicit distinction between the unconditioned stimulus ( US ) and conditioned stimulus ( CS ) . Instead , we assumed that during conditioning , neurons in the LA initially responded to the US and eventually to the CS , while continuing to respond to the CS during extinction [23] . The output of these LA neurons was then fed downstream to the BA . In addition , US or CS inputs from the thalamus or the primary sensory cortex may directly target BA neurons [24] . In our model , we represented those inputs , independently of their origin , as CS-US in the conditioning context and CS in the extinction context . For the description of the SNN we adopted the good model description practice proposed by [25] , which provides guidelines for a standardized way of describing complex neural networks . We share the authors' belief that such model description facilitates reproducibility and direct comparisons between models . Within this framework , we organized the description in different subsections , complemented by additional information on the model parameters . This collected information is presented in an easily accessible , tabular form in the Supplementary Materials ( Table S1 ) . Our choice to use leaky-integrate-and-fire ( LIF ) neurons was motivated by four major arguments: ( i ) multiple combinations of sub-cellular parameters can result in the same network state [26]; ( ii ) even simple neuron models such as LIF with minor modifications are sufficient to reproduce complex in vivo spike patterns [27]; ( iii ) realistically-sized large scale networks of LIF neurons can now be simulated with the currently available simulation technology [28]; this is hardly possible for similarly large networks built of detailed compartmental models and , finally , ( iv ) the extent to which sub-cellular properties of individual neurons influence the global network dynamics is presently not clear . Most importantly , however , here we are interested in understanding the key network level properties of the BA which play a critical role in the formation of fear and extinction memories . For this purpose , the LIF neurons , although they are reduced models of a biological cell , provide an adequate level of biophysical realism , sufficient to identify these key network properties . We modeled the BA as a random recurrent network , consisting of excitatory ( EXC ) and inhibitory ( INH ) neurons [24] , [29] . A total number of 4000 neurons corresponds roughly to 10% of all neurons in the rat BA [30] . The schematic diagram of the network is shown in Fig . 2B . Each connection from a pre- to a post-synaptic neuron had an assigned probability , the value of which depended on the types of pre- and postsynaptic neurons involved ( EXC and INH , respectively ) : , , , and . Thus , each EXC neuron received on average excitatory and inhibitory connections . Likewise , each INH neuron received excitatory and inhibitory connections . Neurons were allowed to form recurrent connections to themselves . For the simulations shown in the last figure , we systematically varied the connection probability of the recurrent inhibition from 0 . 1 to 1 . 0 . EXC and INH neurons received inputs encoding information on the CS . Similarly to the rate model , these inputs represented initial responses of LA neurons to combined CS and US presentations , later only to the CS . They might also reflect more peripheral , thalamic or cortical responses to CS-US . A fraction of BA EXC neurons ( 20% , randomly chosen ) received inputs representing CS and . Similarly , another 20% of BA EXC neurons received inputs representing CS and . Thus , similar to the rate model , we assumed that BA EXC neurons receive contextual information directly from the HPC ( or entorhinal cortex ) and/or via the mPFC . Crucially , CS-US and contextual inputs converged onto the same neurons [8] . Furthermore , EXC and INH neurons received unspecific background inputs ( BKG ) , representing activity originating in other areas , either within or outside the amygdaloid complex . The BKG inputs accounted for the baseline spiking activity of EXC and INH neurons at <1 Hz and 10–15 Hz , respectively [24] . The exact temporal and spatial patterns of the spiking inputs to the BA are not known . Here , we used independent Poisson spike generators with different firing rates to produce the specific inputs . Contextual and BKG inputs provided a tonic drive to BA neurons . By contrast , the CS input had a short duration of 50 ms , based on the experimental design used in [17] . All external inputs formed excitatory synapses onto their target neurons . Neurons were modeled as leaky-integrate-and-fire ( LIF ) neurons . The subthreshold dynamics of each LIF neuron were governed by the following equation ( 5 ) A spike was generated whenever the membrane potential crossed a predefined static threshold θ in upgoing direction . The potential was then reset to a value Ek and clamped for tref ms before the synaptic integration started again ( Table S1F ) . Neurons made either excitatory or inhibitory connections onto their postsynaptic targets via conductance-based synapses [31]–[33] . The synapses of all connections were non-modifiable , except those providing CS and contextual input to EXC neurons . These latter , plastic synapses were modified according to the following phenomenological rule: ( 6 ) ( 7 ) ( 8 ) Note that three variables were used: the synaptic weight w and the auxiliary variables c and h . Each time a presynaptic neuron fired , the value of c increased by a fixed amount . Afterward , this value relaxed towards zero . Thus , variable c acted as a synaptic tag , encoding the recent activity in the synapse receiving CS input . Likewise , variable h encoded information about recent activity in neighboring synapses receiving contextual input . At the offset of each CS presentation , the variables c and h were probed in the synapses of all EXC neurons and the strength of each synapse was modified accordingly . The synaptic strengths before and after the update are denoted by w− and w+ , respectively . If CS and contextual inputs at the same neuron coincided within a temporal window of ∼100 ms , then both synapses were strengthened [34] . By contrast , if only one of the inputs was present , both synapses were weakened ( Eq . 6 ) . This decrease of synaptic strength was based on studies reporting that synapses in LA , which had been strengthened during fear conditioning , depotentiated after extinction training [35] , [36] . We assumed a similar mechanism to hold for the BA . This type of bidirectional plasticity rule implemented in our model is similar to the BCM rule [37] , the “calcium-control hypothesis” [38]–[40] and the ABS rule [41] , [42] . Common in all these rules is the specification that the level of postsynaptic Ca2+ determines the direction of plasticity ( for review see [43] ) . A large increase in Ca2+ causes LTP , whereas a moderate increase results in LTD . Low levels of Ca2+ do not cause any modification at all . We essentially incorporated this bidirectional induction of plasticity in our rule using fixed thresholds ( Fig . 3C ) , rather than sliding ones , as is the case e . g . in the BCM rule . The parameters a1 and a2 denote the learning rates for potentiation and depotentiation of the synapses , respectively . Ca2+ influx depends on NMDA receptor activation and sufficient postsynaptic depolarization . The latter can be caused by coincident presynaptic input or by a backpropagating action potential ( BAP ) . However , in our model , a BAP was not required . That is , we assumed that if the total presynaptic firing rates were high enough , they could cause sufficient depolarization to unblock NMDA receptors . This assumption is supported by experimental evidence showing that a BAP is neither necessary nor sufficient for synaptic plasticity [44] , [45] . Note that this plasticity rule is also compatible with changes induced purely in the presynaptic terminal . In fact , experimental evidence suggests that presynaptic induction , completely independent of postsynaptic activity , occurs in the LA [46] . Thus , the plasticity rule implemented in our model incorporates both changes that are dependent on post-synaptic depolarization , but not postsynaptic spiking , and changes that are presynaptic and entirely independent of post-synaptic depolarization or spiking . Because in our model the presynaptic spiking was caused by CS and contextual inputs , their total activity encoded in the variables c and h , respectively , determined the direction of plasticity . Thus , both c and h functioned as eligibility traces for synaptic modification [34] , [47] . They could be interpreted as describing any relatively slow process associated with the effects of Ca2+ , e . g . autophosphorylation of CaMK-II [39] , [48] . The terms and in the update rule were introduced to provide upper and lower bounds to the synaptic weights , such that they did not increase or decrease indefinitely . They also controlled the step-size with which synapses were modified: the closer a weight was to wmax ( wmin ) the smaller were its increments ( decrements ) . The parameter m represented the action of neuromodulators released during fear conditioning and extinction . It is known that many neuromodulators target the BA [5] , possibly affecting synaptic plasticity in a complex way . Among the possible candidates are norepinephrine ( NE ) [49]–[52] , dopamine ( DA ) [53] , [54] and opioids [5] . Here , however , lacking more detailed experimental data , we cannot be more specific about which exact neuromodulators are involved and how they interact . Fortunately , this lack of knowledge does not pose a problem for the plasticity rule we propose , because it is general enough to accommodate any combination of neuromodulators that may turn out to be involved in BA fear processing . The dynamics of the mean-field model were simulated in MATLAB . The SNN simulations were written in python ( http://www . python . org ) , using the PyNN interface [55 , http://neuralensemble . org/trac/PyNN] to the NEST simulation environment [56 , http://www . nest-initiative . org] .
Fig . 4 shows the response of the mean-field model , i . e . the firing rate model , of BA during fear conditioning and extinction . To simulate fear conditioning in , we stimulated the population A five times with CS , US and inputs ( Eqs . 1 , 2 ) . This resulted in a progressive strengthening of CS synapses onto population A ( ) ( Fig . 4C ) , accompanied by a corresponding increase in the response of population A ( Fig . 4A ) . To simulate fear extinction training in , we stimulated population A with CS input and population B with CS and input six times to mimic a different context . Now , in , the synaptic strength of the CS input synapses ( ) onto population B progressively got stronger , whereas remained unchanged ( Fig . 4D ) . The slow increase in the response of population B resulted in a small decrease in the response of population A , due to the recurrent inhibition . When the strength of became larger than ( Fig . 4D ) , the activity of population B dominated and , hence , the response of population A was suppressed ( Fig . 4B ) . The differential activation of two neuronal sub-populations in two different contexts can be interpreted as fear ( population A ) and extinction ( population B ) neurons as observed in [17] . This is purely a functional characterization of the two sub-populations , which are identical otherwise . That is , we used exactly the same parameters for both sub-populations and the differential activation results solely from differences in contextual inputs they receive . Thus , the two populations were not different in terms of their intrinsic properties . Of course , cases where the two subpopulations do have different properties can be easily accommodated in the model resulting in an enhancement of the differential activation . To be consistent with [17] , we used the terms fear and extinction neurons to refer to those subpopulations that are active in and respectively . Note that we did not include any component that imitates behavioral output , i . e . freezing . Instead , we assume , in agreement with experimental findings [17] , that high activity of fear neurons directly corresponds to a high level of freezing whereas high activity of extinction neurons and low activity of fear neurons corresponds to low levels of freezing . Although a simple firing rate model was able to account for the dynamic emergence of fear and extinction neurons , such mean-field models have only limited explanatory and predictive power . For instance , they assume uncorrelated activity in the underlying neuronal populations and , thus , cannot be used to predict any correlations in firing rate or spike timing that may emerge in the network . In addition , these models cannot be used to predict the spike patterns of individual neurons . Thus , to understand the dynamics of the BA network beyond average firing rates only , we simulated a biologically realistic large-scale network composed of spiking neurons . Again , fear conditioning and extinction were simulated by applying five CS-US presentations in and six CS presentations in respectively . In the two different environments tonic contextual input was provided to EXC neurons ( cf . Models ) . The results of the simulation are presented in Fig . 5 . Initially , all EXC neurons spiked at very low firing rates . Presentations of the CS-US led to a steady increase in the firing rates of one sub-population ( fear neurons ) within the EXC population , which peaked at the end of conditioning ( Figs . 5A , E amber dots ) . The increase in activity of fear neurons was a direct consequence of the potentiation of CS and contextual inputs onto fear neurons ( Figs . 5 G , I; amber triangles ) . In , the fear neurons still responded with high firing rates upon the first CS presentation , even though they did not receive contextual inputs ( Figs . 5A , F ) . With further CS presentations , however , synapses became potentiated ( Eq . 6 , Figs . 5H , J; cyan dots ) , causing a steady increase in the firing rate of the second sub-population of neurons ( extinction neurons ) ( Figs . 5A , F; cyan dots ) . The increased recurrent inhibition in the network then caused a decrease in the activity of the fear neurons ( Figs . 5A–C , F ) . At the end of extinction , the population rate of the extinction neurons peaked , whereas the firing rate of the fear neurons had returned to the initial , pre-conditioning values . The reduction of fear neurons activity was further facilitated by small depotentiation of CS and contextual input synapses onto the fear neurons ( Eq . 6 , Figs . 4H , J; amber triangles ) . Note that depotentiation of CS synapses onto extinction neurons also occurred during conditioning ( Fig . 5G ) as described by the learning rule . By contrast , CTX synapses were not decreased during conditioning , because their initial values were close to the lower bound ( w− ) ( Fig . 5I ) . During conditioning and extinction the baseline firing rates increased as well ( Fig . 5A ) . This increase was induced by the strengthening of the contextual inputs ( Figs . 5H , I ) , providing an explanation for contextual conditioning . However , because only a small percentage of neurons exhibited this increase in firing rates , this could make it difficult to measure it experimentally . This fact reveals a key advantage of network models which allow for simultaneously sampling a large number of neurons . Based on this , predictions can be inferred which otherwise would not have been possible . Note that , again , the assignment of BA EXC neurons in fear and extinction sub-populations is purely a functional one . That is , neurons were characterized post-experiment as fear or extinction cells depending on whether they responded to the CS after conditioning or after extinction training respectively . In particular , they were not predetermined in terms of their intrinsic properties and the two sub-populations resulted solely from the differences in the contextual inputs they received . Also , it is important to emphasize that whereas the population rates of fear and extinction neurons increased gradually during conditioning and extinction training respectively , this was not the case for individual neurons . Instead , they changed their state quite abruptly from non-responding to responding ( Fig . 6A ) . The further the training advanced , the more neurons started to respond . Hence , the gradual increase in population rates ( Figs . 5E , F ) reflects the growing recruitment of responding neurons , rather than a gradual increase of single neuron activity itself ( Fig . 6 ) . The responsive neurons fired maximally two spikes per CS presentation . The baseline firing rates for the inhibitory population were normally distributed with a mean of 10 Hz , whereas the CS-evoked rates shifted their distribution towards a mean of 20 Hz . This is consistent with the neuronal firing patterns in vivo reported by [17] . Although we performed our main simulations using separated contextual inputs to distinct neuronal subpopulations within the BA ( cf . Models ) , this is not a necessary requirement of the model . In fact , performing simulations with varying amounts of contextual input overlap showed that fear and extinction neurons still existed as distinct populations , even when contextual inputs had an overlap of around 50% ( Fig . S1 ) . In addition , the simulations revealed the existence of a third sub-populations of neurons . These were the neurons receiving inputs in both contexts and , thus , were active during both fear conditioning and fear extinction ( so called persistent neurons ) . Note that , similar to the case of fear and extinction neurons , the characterization of cells as “persistent” is functional and denotes the fact that these neurons were responding to the CS during both conditioning and extinction . Moreover , these neurons had much stronger CS and CTX synapses , which resulted in higher firing rates . This observation of the model is consistent with the experimental data [17] , suggesting that conditioning and extinction are not affected by overlapping inputs , unless the overlap is high ( >50% ) . Following extinction training in , presentations of the CS in the original fear conditioning context ( ) resulted in context-dependent renewal ( ABA renewal ) of conditioned fear responses [2] . This renewal phenomenon points at two important aspects of possible neural mechanisms underlying fear extinction: ( i ) extinction is mainly new learning and only partly unlearning of previously acquired fear memories ( [57]; see also Discussion ) , ( ii ) extinction learning is context-dependent . We simulated ABA renewal by changing context at the end of extinction ( Fig . 7 ) . This resulted in a sudden switch of activity between fear and extinction neuronal subpopulations . That is , although the activity of extinction neurons was high after extinction learning , the contextual switch caused the activation of fear neurons and a significant drop in the extinction neurons activity . These results are in complete accordance with the experimental findings reported by [17] . It is important to note that this rapid activity switch is purely a network phenomenon and not an effect of synaptic plasticity , as the change is much too fast for the plasticity mechanisms to act . We illustrate this point by depicting the average membrane potentials of 100 randomly selected fear and extinction neurons ( Figs . 5D , 7D; amber and cyan traces respectively ) . It is evident that in either context there was a clear difference between the membrane potentials of the two cell populations , stemming from the fact that one of the populations continuously received a higher excitatory drive due to the additional contextual input . Switching contexts led to a corresponding instantaneous switch in the assignment of the contextual input and , hence , in opposite shifts in the average membrane potentials of the two sub-populations , which was immediately reflected in corresponding shifts in the firing rates . We also modeled the case where the renewal context was different from both the conditioning and the extinction context ( ABC renewal ) . The results of the simulations revealed that if after extinction training the CS was presented in a third , different , fear neurons became rapidly active again and suppressed extinction neurons ( Fig . S2 middle ) . However , our model also indicated that the absolute response of fear neurons - and thus the magnitude of the fear response- would be weaker than in the ABA case . The reason is that in CTX synapses had not been strengthened during the conditioning procedure . This provides an account for the experimentally observed ABC renewal [58] , [59] explaining why ABC renewal may occur in the first place and also why the effect may be weaker compared to ABA renewal . Moreover , our simulations also suggested that massive extinction ( extinction over-training ) in can abolish ABC renewal , because depotentiation of CS and afferents onto BA neurons yield less excitatory input to these neurons . Extinction over-training can also impair ABA renewal , although to a lesser extent ( Fig . S2 right ) . The reason that ABA renewal is more robust and ABC renewal more vulnerable to massive extinction stems from the fact that in the latter case not only CS synapses onto fear neurons are weakened , but also potentiated CTX synapses are entirely missing . These findings are in agreement with and provide a possible explanation for the experimentally observed effects of massive extinction [60] . Although we did not focus on extinction of contextual fear , it is important to note that our model also accounts for this specific conditioning phenomenon . Indeed , the plasticity rule dictates that in the absence of the CS synaptic weights will decay . That is , CTX synapses , which had been strengthened during conditioning in and encode contextual fear , will depotentiate in the same context if the CS is not present . This will yield decreased fear neuron activity and , thus , extinction of contextual fear . Note that within the framework of our model , this form of extinction is truly unlearning and not masking of contextual fear memories . The experimentally reported connection probabilities from excitatory to inhibitory neurons as well as among inhibitory neurons in the BA are around 0 . 5 [61] . This is a much higher value than the ones we used in our initial simulations ( Figs . 5–7 , Table S1E ) . To test the effects of such higher connectivity , we performed additional simulations adopting the experimentally reported values for the connection probabilities . The qualitative behavior of the model did not change ( data not shown ) . However , a new aspect in the network dynamics emerged . High frequency oscillations - typically in the gamma range ( 30–80Hz ) - occurred throughout the simulation in both excitatory and inhibitory populations . These oscillations were present already in the ongoing activity patterns and CS-US presentation enhanced them even further ( Fig . 8A ) . They resulted from the high shared connectivity and , hence , large amount of shared inputs that caused correlated spiking in the neurons . The oscillation frequency was determined by synaptic time constants and delays in the network . Gamma oscillations in networks of excitatory and inhibitory neurons have been reported in many experiments [62]–[67] and discussed in multiple theoretical studies [68]–[75] . Moreover , several studies have reported gamma oscillations in the amygdala under certain conditions , e . g . in anesthetized animals [76] , in slow wave-sleep [77] , in the presence of reward predicting stimuli [78] and in paradigms involving consolidation of emotional memories [79] . Therefore , there is at least partial experimental and theoretical support for the gamma range oscillations observed here in high connectivity BA network simulations . Yet , in networks with high mutual connectivity between excitatory and inhibitory neurons and within inhibitory neuron populations such as in the BA , oscillations should be a prevailing feature and should , therefore , be readily identifiable in vivo recordings under all conditions and not only in the special cases mentioned above . It is , thus , possible that certain mechanisms operate in the BA that could dampen gamma oscillations ( but see Discussion ) . We , therefore , used our network model to investigate this issue in further simulations by exploring the parameter space of the network properties that could quench oscillations . Two mechanisms proved to be effective in reducing the power of gamma oscillations . The first one was the introduction of heterogeneity in the inhibitory population [80] , [81] . This approach was motivated by experimental data showing that interneurons in the BA exhibit a large diversity in terms of their morphological and electrophysiological properties [24] , similar to interneurons in the cortex [82] and hippocampus [83] . In the latter case , the diversity was expressed in a wide range of values for synaptic rise times , reversal potentials , response latencies etc . In a preliminary study [84] , we introduced heterogeneity in one of the neuronal properties in our model , the spiking threshold , by drawing values from a bimodal distribution with peaks at −35 mV and −28 mV . This corresponds to the experimentally measured threshold values of two subclasses of parvalbumin-expressing interneurons in the BA: the fast-spiking ( FS ) and the delayed-firing ( DF ) interneurons [24] , [61] . In such heterogeneous networks , oscillations were indeed reduced , but not totally eliminated [84] . A second , more effective way to reduce the network oscillations was to decrease the synaptic delays between inhibitory neurons ( Fig . 8B ) . First , we studied the oscillations dynamics for different connection probabilities in a network of homogeneous neurons , with synaptic delays drawn from a uniform distribution ( 1–2 ms ) . In such networks , increasing connectivity ( >0 . 2 ) enhanced the oscillations and synchrony to their maximum ( Fig . 8B , green solid lines ) . Only for very weak synapses ( 1 nS ) , that is , when the network was mainly driven by external inputs , increasing the connectivity did not add to the oscillations ( Fig . 8B , gray solid line ) . Increasing the width of the synaptic delay distribution did not reduce the synchrony and oscillations in high-connectivity networks ( data not shown , Vlachos et al . in prep . ) . However , choosing short delays from a narrow uniform distribution ( 0 . 2–1 ms ) considerably reduced the oscillations , up to connection probabilities of 0 . 4 ( Fig . 8B , green and gray dashed lines ) . Thus , in a recurrent network , smaller delays have a powerful effect in reducing synchrony and oscillations . This finding is in agreement with a previous numerical study [69] and also with more recent analytical approaches [74] , [75] , [85] . At first sight , synaptic delays less than 1 ms might appear unrealistically small . However , delays as short as 0 . 5 ms have been reported among inhibitory neurons in the hippocampus [66] . Moreover , the delays between inhibitory neurons in the BA have been reported to be around 0 . 7 ms [61] , or even smaller ( Lüthi , unpublished data ) . Therefore these short delays , might indeed account for the lack of gamma band oscillations observed under baseline conditions in experimental recordings . Because inhibition plays a critical role in our model , we tested the effects of partial inactivation of inhibitory neurons . For this , we performed two additional sets of simulations , in which , during acquisition of extinction , we deactivated 50% and 90% of INH neurons , respectively . The results are shown in Fig . 8C . As expected , with reduced inhibition the activity of both fear and extinction neurons increased . The increase of activity of the latter population was more pronounced , due to the fact that it received additional excitatory drive from contextual inputs in . This suggests that blockage of inhibitory activity should lead to enhanced , context-specific extinction . This is consistent with the finding that GABA blockage enhances extinction of contextual freezing [86] . However , there is a potential caveat here . Activity of both fear and extinction neurons is increased upon blockage of inhibition and it is not clear how downstream structures , specifically CEA neurons , would respond to this . If the relative difference between fear and extinction neuron activity matters , then extinction should be facilitated by impaired inhibition . If , by contrast , the ratio between fear and extinction neuron activity is more relevant , then extinction might be impaired . Note that these two possibilities apply to both blocking of inhibition during acquisition of extinction training and blocking of inhibition during expression of fear extinction . Because contextual input is one of the key aspects in our model we tested how removal of these inputs would affect the behavior of the network . The simulations yielded two different results depending on the exact time point of removal of contextual inputs . When contextual inputs were removed after fear conditioning , fear neurons remained active during extinction training and no extinction neurons emerged ( Fig . S3; left ) . This result is a direct consequence of our synaptic learning rule , because strengthening of synapses requires temporal overlap of CS and CTX inputs . Note that although fear neurons remained active , their firing rates were reduced , because they now lacked contextual input . Thus , our model suggests that lesions of hippocampal or prefrontal areas after fear conditioning may result in impaired extinction . This conclusion is supported by experimental evidence [87] . By contrast , when contextual inputs were removed after fear extinction , activity of neither fear nor extinction neurons was sufficiently strong to suppress the other neuron group ( Fig . S3; right ) . That is , because the decisive contextual input was lacking and , thus , both groups were simultaneously active , although to a lesser degree than in case either group was active alone . The behavioral consequences of these results are beyond the scope of our model , because here we did not model any downstream structures such as the central amygdala that presumably further process output from fear and extinction neurons . Thus , at present , we can only speculate that lesions of hippocampal or prefrontal areas after extinction training may result in impaired renewal , because fear neuron activity will be both decreased and also counteracted by simultaneous extinction neuron activity . In fact , experimental evidence supports this conclusion [88] . However , it is important to point out a subtle difference between our model and certain lesion experiments . In our model , removing CTX input means that the BA network does not receive any contextual input at all . By contrast , in some experiments in which the hippocampus had been lesioned or inactivated , contextual information may still have been accessible , because the context in which the CS was presented was still decisive for the behavioral outcome [89] , [90] . Our model enables us to make a number of specific predictions that can be tested experimentally:
One core feature of our model is that contextual inputs are gated to the BA . In this framework , the precise origin of these inputs does not matter; as long as the BA neurons receive differential inputs in two different contexts , the model behavior remains unaltered . However , there are strong indications from anatomical [29] , [93] , [94] and physiological [93] studies that the HPC is a major source of contextual information to the BA . In addition , a previous report showed context-dependent modulation of neuronal activity in the LA [95] . By designing our model to have contextual input directly influencing the activity of excitatory neurons in the BA , we have essentially postulated a similar mechanism for this subnucleus . This assumption is further supported by the finding that fear neurons show orthodromic responses to HPC stimulation [17] . A second source of contextual input may be the mPFC . There is anatomical evidence that the mPFC projects to the BA [29] . Moreover , [17] reported that mPFC stimulation induces orthodromic responses in identified extinction neurons . Here , we suggest that part of the information conveyed by these projections might be contextual . This assumption is based on evidence reporting extinction-related induction of LTP on hippocampus-mPFC afferents [96] . In our model both fear and extinction neurons receive context-specific information either directly from hippocampus or indirectly via the mPFC . This may also explain the ambiguous results that the hippocampus may or may not interact with the mPFC during extinction learning [97] . The context-specific modulation of activity in the BA presented here provides a general framework that can explain experimental findings on the involvement of the hippocampus in the acquisition , encoding , and context-dependent retrieval of both conditioning [13] , [98] , [99] and extinction memories [3] , [87] . Future refinements of the model , in combination with new experimental data are necessary for a better understanding of the detailed interactions between hippocampus , mPFC and amygdala . We showed that high connectivity between excitatory and inhibitory and within inhibitory neuron populations results in robust oscillations in the gamma range , characterized by high activity correlation among neurons . The main cause of these oscillations was the high degree of shared inputs among neurons as a result of the dense connectivity . We suggested two different , biologically plausible ways to reduce these oscillations: by either introducing heterogeneity in neuron properties and/or by reducing synaptic delays to sub-millisecond time scales . Yet another way would be to have synapses exhibit a certain transmission failure rate [100] , [101] , resulting in activity dependent reduction of the effective connectivity . However , we do not wish to imply that gamma oscillations do not exist in the BA . In fact , as noted earlier , gamma oscillations have been reported in the amygdala under various conditions [76]–[79] . Here , we want to emphasize the point that in networks with high connectivity , gamma range oscillations are a salient feature of the network dynamics . Therefore , they should be visible even in the ongoing activity , unless suppressing mechanisms , such as those elaborated here , are in effect . Several suggestions for a specific role of gamma oscillations have been made in the past . For instance , it has been proposed that in the cortex or the hippocampus oscillations might contribute to temporal encoding [102] , sensory binding [103] , attentional selection [104] and memory formation or retrieval [105] , [106] . It is currently unclear whether these hypotheses also apply to the amygdala . Oscillations in lower frequency ranges ( delta and theta ) have also been reported . For example , increased theta oscillations - that synchronized with hippocampal theta activity - were shown to be related to conditioned freezing [107] , [108] , whereas delta oscillations have been implicated in gating aversive stimuli [109] . Gamma oscillations , on the other hand , have been suggested to facilitate interactions between the amygdala and connected structures [78] , [110] . Here , because we modeled only the BA , we cannot give any informed predictions about how gamma oscillations may affect those various interactions . Moreover , in our current model , we have used plasticity only in the input connections and those are not affected by oscillatory activity in the recurrent network . However , before addressing the effects of gamma oscillation on the dynamics of the BA network , it is of key importance to resolve experimentally whether gamma oscillations are indeed present in BA activity and , if so , under which conditions . A well-known behavioral phenomenon is conditioned inhibition , referring to the ability of a second CS ( CS− ) to suppress the conditioned response , after it has been paired several times with the first CS ( CS+ ) in the absence of a US [57] , [111] . It is possible that the CS− , referred to as conditioned inhibitor , employs similar mechanisms to those described in our model to suppress the conditioned response . That is , neural subpopulations in the BA encoding the CS− might , similar to extinction neurons , use local inhibitory circuits to suppress fear neuron activity . Future work is needed to explore further this interesting line of reasoning . Our model accounts for experimental paradigms that use a different extinction context from the conditioning one , but not for those in which fear conditioning and extinction occur in the same context . For instance , if conditioning and extinction both occur in , then only those neurons that receive inputs in this context will be active . Thus , downstream structures will not be able to differentiate between fear conditioning and extinction training solely from spiking activity in the BA . It is evident that performing conditioning and extinction in the same context per se increases ambiguity about the meaning of the context . Thus , it is likely that circuits within the BA alone are not sufficient to solve this computational problem . Both , a detailed description of neural activity during this type of extinction and a more detailed analysis of interactions between the BA and downstream structures are required to address this behavioral phenomenon . Although a wealth of experimental studies exist on the amygdala and its role in fear conditioning and extinction , computational or theoretical approaches to study amygdala function are largely lacking . Most of the previous theoretical studies involve symbolic models [112] , [113] , mainly based on the Rescorla-Wagner rule [114] . These models have their merit in describing behavioral findings such as generalization , blocking etc . However , since these models treat the amygdala as a “black-box” , it is not within their scope to account for neuroanatomical or electrophysiological data , therefore providing little insight into the underlying neuronal mechanisms involved . Despite these apparent differences , it is still possible to draw some parallels to symbolic models . For instance , in our model , potentiation of synapses occurs only if CS and CTX inputs temporally overlap . This is similar to the SOP model , where US and CS have to coincide for strengthening of associations to take place [115] , [116] . Connectionist or parallel-distributed ( PDP ) models of fear related processes go one step further than symbolic models by introducing networks composed of multiple , mutually connected computational units . One such model was successful in capturing certain features observed in fear conditioning studies [117] . Its main limitation , however , is the fact that it does not take into account the different substructures within the amygdala , nor do the computational units used in the model map to any biophysically realistic counterparts . Fortunately , the computational power presently available allows us to improve these models and to overcome many of their limitations . The model presented here is to our knowledge the first large-scale spiking neuron network model that investigates the mechanisms of fear conditioning and extinction within the amygdala using biologically realistic neurons in adequate detail . The model closest to this is a compartmental model introduced by [118] to investigate the function of the LA in fear conditioning and extinction . However , [118] used a small network composed of only eight two-compartment neurons and focused on role of the kinetics of multiple ionic currents in fear conditioning and extinction . By contrast , we modeled the BA using a large network of 4000 LIF neurons , which enabled us to identify the network level interactions involved in the formation of fear and extinction memories . The present model provides a plausible explanation for the neural mechanisms underlying fear conditioning and extinction within the BA . We did not address the question of how the neural activity within the BA impacts on downstream structures , such as CEA or mPFC . We neither attempted to model the interactions between hippocampus and mPFC in conditioning and extinction , which would require additional experimental data to constrain the possible models . Given these restrictions , we provided a plausible mechanism of how contextual inputs may affect the activity of distinct neuronal subpopulations in the BA , enabling them to control downstream structures such as the CEA . We proposed that context-related aspects of fear and extinction memories are partially stored in the BA and that they provide a context-dependent instruction for the triggering or blocking of the fear-response . In addition , we showed how extinction training may mask previously acquired fear memories and , thus , provided an account for renewal . Finally , our model , next to yielding several interesting predictions discussed above , raises the important question of how downstream structures such as the CEA or mPFC discriminate the activity of the distinct neuronal subpopulations within the BA . Is this problem solved purely on an anatomical level , e . g . by differential projections of the BA subpopulations to specific target neurons ? Or do specific features in the activity of the BA subpopulations , e . g . the statistical structure of pairwise or higher-order correlations , also play a role , providing downstream networks with a mechanism to distinguish between them ? These questions need to be addressed in future work combining experimental and theoretical approaches . | The amygdaloid complex is one of the key brain structures involved in fear-related processes . A typical way to study neural correlates of fear expression ( e . g . freezing response ) in the amygdala is to perform a fear conditioning paradigm , which yields a conditioned fear response . This response can be reversed by another procedure called fear extinction . Thanks to the experimental approaches to date we have some understanding about the putative roles of specific subnuclei within the amygdala in the formation of these fear and extinction memories . Here , we complement the experimental studies by providing a computational model that addresses the question of how fear and extinction memories are encoded in the amygdala , and specifically , in the basal nucleus ( BA ) . We propose a specific neural mechanism to explain how the BA may integrate information about a salient , conditioned stimulus and the environment , thereby enabling it to switch the state of the animal from low to high fear and vice versa . We also provide possible explanations for various other behavioral findings , such as the recovery of fear after it had been extinguished ( renewal ) . Finally , we make specific , experimentally testable predictions that need to be addressed in future work . | [
"Abstract",
"Introduction",
"Models",
"Results",
"Discussion"
]
| [
"neuroscience/behavioral",
"neuroscience",
"neuroscience/animal",
"cognition",
"neuroscience/theoretical",
"neuroscience"
]
| 2011 | Context-Dependent Encoding of Fear and Extinction Memories in a Large-Scale Network Model of the Basal Amygdala |
Trachoma programs rely on guidelines made in large part using expert opinion of what will happen with and without intervention . Large community-randomized trials offer an opportunity to actually compare forecasting methods in a masked fashion . The Program for the Rapid Elimination of Trachoma trials estimated longitudinal prevalence of ocular chlamydial infection from 24 communities treated annually with mass azithromycin . Given antibiotic coverage and biannual assessments from baseline through 30 months , forecasts of the prevalence of infection in each of the 24 communities at 36 months were made by three methods: the sum of 15 experts’ opinion , statistical regression of the square-root-transformed prevalence , and a stochastic hidden Markov model of infection transmission ( Susceptible-Infectious-Susceptible , or SIS model ) . All forecasters were masked to the 36-month results and to the other forecasts . Forecasts of the 24 communities were scored by the likelihood of the observed results and compared using Wilcoxon’s signed-rank statistic . Regression and SIS hidden Markov models had significantly better likelihood than community expert opinion ( p = 0 . 004 and p = 0 . 01 , respectively ) . All forecasts scored better when perturbed to decrease Fisher’s information . Each individual expert’s forecast was poorer than the sum of experts . Regression and SIS models performed significantly better than expert opinion , although all forecasts were overly confident . Further model refinements may score better , although would need to be tested and compared in new masked studies . Construction of guidelines that rely on forecasting future prevalence could consider use of mathematical and statistical models .
The World Health Organization ( WHO ) , the International Trachoma Initiative , Ministries of Health , and their partners aim to control blinding trachoma by 2020 , implementing surgical campaigns , antibiotic distributions , hygiene initiatives , and environmental improvements [1] . Trachoma control is a massive undertaking: 50 million doses of antibiotics are now distributed annually , in 30 countries [2] . The Global Trachoma Mapping Project alone will complete population-based surveys in more than 1400 districts worldwide by the end of 2015 [3 , 4] . Surveys and treatment histories are now available for the vast majority of trachoma-endemic districts worldwide [5] . However , we do not know where WHO goals will likely be and not be achieved . Decisions on when to start and stop treatments are still based on guidelines dependent in large part on expert opinion [6] . Mathematical models have provided insight into the transmission of infectious diseases including trachoma [7–15] . However , they have rarely been used to make falsifiable predictions . As a candidate for prediction , trachoma may have some advantages over other infectious diseases . Trachoma has no nonhuman reservoirs , no long-lasting latent stage , and as yet no clinically important drug resistance , simplifying modeling greatly compared to diseases such as cholera , onchocerciasis , and tuberculosis [1] . With many infectious diseases such as SARS and Ebola [16 , 17] , epidemics occur sporadically in time and place; forecasting can be made in a predictable time frame with post-treatment trachoma . Community-randomized trials have provided longitudinal assessment of multiple communities after mass antibiotic distributions . In a sense , after each mass treatment has brought infection to a low level , infection returns in a synchronized manner in a number of communities , offering results somewhat analogous to a repeated experiment . Accurate forecasts could inform stakeholders of realistic goals , define trouble spots to focus resources , and suggest areas headed towards control even in the absence of intervention . Prediction has scientific value as well . The ability to predict the prevalence of an infectious disease is a test of our understanding of the epidemiology . Here , we use recent clinical trial data to forecast the prevalence of ocular chlamydial infection in children in 24 endemic communities in Niger . We compare model forecasts to expert opinions , and to a statistical regression that uses no special knowledge of the infectious process .
Forty-eight communities were followed as part of the Niger arm of the Partnership for the Rapid Elimination of Trachoma ( PRET ) study . Communities were randomized to either mass antibiotics of the entire community , or antibiotics targeted just to children 12 years and younger . The 24 communities included in this study received annual antibiotic treatment of all ages . Communities were assessed at baseline and then biannually for 3 years . All individuals were offered antibiotic treatment annually , within two weeks of the assessment: children under 6 months , those allergic to macrolides , and pregnant women were offered topical tetracycline , and all others were offered a single dose of oral azithromycin ( 20 mg/kg for children and 1 gram for adults ) . A random sample of 100 children 0–5 years old were selected from each community . If a community had less than 100 0–5 year-old children , then all were offered assessment . Each participating child had their upper right tarsal conjunctiva swabbed , and processed for PCR as previously described [18] . This study of de-identified data received ethical approval from the Committee on Human Research of the University of California San Francisco and was carried out in accordance with the Declaration of Helsinki . All adult subjects provided informed consent , and a parent or guardian of any child participant provided informed consent on their behalf . The informed consent given was oral: ( a ) we chose verbal consent because of the low literacy rates in the study area , ( b ) the IRB ( 10 . 00812 ) approved the use of oral consent , and ( c ) oral consent was documented on the registration form for each study participant prior to examination in the field . The WHO NTD-STAG Monitoring and Evaluation Working Group had a sub-group meeting to discuss trachoma surveillance on September 11–12 , 2014 in Atlanta , GA , USA at the Task Force for Global Health , co-sponsored by WHO and the NTD Support Center . Fifteen trachoma experts were asked to forecast the 36 month prevalence of infection in the 24 communities of the PRET-Niger study described above , and were provided , for each community , the biannual prevalence estimates from 0–30 months , the antibiotic coverage at 0 , 12 , and 24 months , the estimated population of 0–5 years olds at baseline , and the number of children sampled at baseline ( Table 1 ) . For each of the 24 communities , the experts were asked to provide their median estimate at 36 months , as well as the lower and upper bounds of their centralized 95% credible interval ( the 2 . 5th , 50th , and 97 . 5th percentile of their belief ) . The community expert opinion was constructed by estimating each of the 15 individual’s distribution for each village ( see Scoring below ) and then taking the arithmetic average assuming equal weights , and was used as the primary survey forecast , although each individual’s forecast was also scored separately . The number of trachoma publications by each of the 15 experts was assessed by a PubMED ( National Library of Medicine ) search on December 1 , 2014 ( expert name as author AND “trachoma” as keyword ) . Linear mixed effects regression was used to model the prevalence at 12 months and 24 months based on observations at 6 months and at 18 months , respectively . A random intercept was used for each village . To improve normality and homoskedasticity , the square root transform was applied to the prevalence fractions . The fitted model was then used to predict the prevalence at time 36 based on observations at 30 months . Standard errors were obtained using clustered bootstrap . All calculations were conducted using R ( R Foundation for Statistical Computing , Vienna , Austria , v . 3 . 1 for Macintosh , package lme4 ) . While the primary regression was of square-root transformed regression with a community-level random effect , we also included a linear regression model without a community-level random effect . We constructed a stochastic transmission model of transmission of Chlamydia trachomatis infection over time . For village j ( j = 1 , … 24 ) , we assumed a population of size Nj , taken from the number of children aged 0–5 years found in the census at the time of treatment k ( k = 1 , 2 , 3 corresponding to baseline , 12 and 24 months ) . We assumed a classical SIS ( susceptible-infectious-susceptible ) model structure , assuming that the force of infection is proportional to the prevalence of infection in the population of children aged 0–5 years with proportionality constant β , and a constant per-capita recovery rate γ [19] . Between periods of treatment , we assumed that the probability pi , j ( k ) ( t ) that there are i infectives in village j at time t after treatment time point k obeys the following equations [20 , 21]: dp0 , j ( k ) dt=γp1 , j ( k ) dpi , j ( k ) dt=β ( i−1 ) ( Nj−i+1 ) Njpi−1 , j ( k ) +γ ( i+1 ) pi+1 , j ( k ) −βi ( Nj−i ) Njpi , j ( k ) −γipi , j ( k ) , for1≤i≤Nj−1 ( 1 ) dpNj , j ( k ) dt=βNj−1NjpNj−1 , j ( k ) −γNjpNj , j ( k ) To model treatment , we assumed that each child aged 0–5 years in village j has probability cj ( k ) of receiving treatment with the antibiotic efficacy ek for treatment period k . We modeled each treatment according to pi , j ( k ) ( t=0 ) =∑i′=iNjpi′ , j ( k , pre ) ( i′i ) ( 1−cj ( k ) ) i ( cj ( k ) ek ) i′−i , where i′ is the number of infected individuals of children aged 0–5 years eligible for treatment , pi′ , j ( k , pre ) is the probability of i′ infected individuals of children aged 0–5 years before treatment time point k , and i is the number of infected individuals of children aged 0–5 years after treatment . Let Sj ( l ) and Mj ( l ) be the observed number of PCR-positive individuals of children aged 0–5 years and the sample size at each observation time point l ( l = 0 , 1 , 2 , 3 , 4 , and 5 corresponding to baseline , 6 , 12 , 18 , 24 and 30 months , respectively ) for village j , and Sj be the possible number ( ranging from 0 to Mj ( l ) ) of positive individuals of children aged 0–5 years detected in the sample at observation time point l . From village j with population ( children aged 0–5 years ) size Nj of which the number Yj of infectives equals i , the probability P ( Sj = s|Yj = i ) that s positives are observed from a sample of size Mj is given by the hypergeometric distribution: ( is ) ( Nj−iMj ( l ) −s ) / ( NjMj ( l ) ) . We assumed a standard beta-binomial prior ( the binomial distribution in which the probability of success at each trial follows the beta distribution ) P ( Yj=y ) = ( Njy ) B ( y+μ , Nj−y+ρ ) B ( μ , ρ ) ( where the shape parameters μ and ρ for each treatment were computed from the observed distribution of infection of 24 villages at baseline , 12 and 24 months , B ( z1 , z2 ) is the beta function ) [22] . The pre-treatment prevalence distribution was then computed for each village by applying Bayes’ theorem: pi , j ( k , pre ) =P ( Yj=i|Sj=s ) =P ( Sj=s|Yj=i ) P ( Yj=i ) ∑i=0NjP ( Sj=s|Yj=i ) P ( Yj=i ) . ( 2 ) For each village j , the initial condition is determined from Eq ( 2 ) , and the system numerically integrated for six or twelve months according to Eq ( 1 ) . Specifically , for each village j , the pre-treatment distributions of kth treatment is pi , j ( k , pre ) =P ( Yj=i|Sj=Sj ( 2k−2 ) ) . Given the number i of infected individuals of children aged 0–5 years , we computed the probability of the observed data of treatment k in village j according to P ( Sj=s ) =∑i=sNjpi , j ( k ) ( τ ) ( is ) ( Nj−iMj−s ) / ( NjMj ) ( where Mj here denotes the sample size at one of the observation time points in the period k , and τ ( 6 or 12 months ) is the interval between treatment time point and observation time point ) . We assumed independent villages , so that the total loglikelihood at time τ months after each treatment k may be computed by summing over all 24 villages ∑j=124log ( ∑i=0Njpi , j ( k ) ( τ ) ( is ) ( Nj−iMj−s ) / ( NjMj ) ) . The transmission coefficient and antibiotic efficacy in the model were optimized by using the Metropolis algorithm with the total likelihood of three treatment periods to fit the model to the observed numbers of PCR-positive individuals of children aged 0–5 years in each village at 6 , 12 , 18 , 24 and 30 months [23] . Forecasting the distribution of the observed number of PCR-positive individuals of children aged 0–5 years in a village at 36 months , conditionally on the observed numbers of PCR-positive individuals of children aged 0–5 years at baseline , 6 , 12 , 18 , 24 and 30 months from the same village , was done by using a hidden Markov model according to the equation of forecast distribution [24] . The primary modeling forecast was pre-specified as the SIS process model with a random effect , although the SIS model without a random effect was included as a sensitivity analysis . In addition , the forecast of each model as a distribution over 101 discrete units was included as a comparison to the distribution estimated by minimizing the Fisher’s information ( which allows a symmetric credible interval to approach a normal distribution , as well as the flexibility of asymmetric credible intervals to represent skewed distributions ) . Sensitivity analyses included changing the fixed mean infection duration assumed in the model to be 6 months , to 3 months or to 12 months . To ensure a fair comparison , all forecasts were scored from the proposed median and 95% CrI . Given the denominator of the sample for each village at 36 months , the discrete distribution which minimized the Fisher’s information while constrained to that expert’s median and 95% CrI was estimated ( Mathematica 10 . 0 ) . As a sensitivity analysis , the SIS model forecasts were also presented as a distribution from 0 to 100% , with the score compared to the score derived from the median and 95% CrI . The modeler , statistician , and each of the 15 experts surveyed were all masked to the 36-month results , as well as to the forecasts made by others . Different forecasts were pairwise compared using Wilcoxon’s signed-rank test ( Mathematica 10 . 0 ) , using the Holm–Bonferroni multiple comparison correction , assuming 3 tests . As a sensitivity analyses , we assessed whether the likelihood of the observed data would be greater ( or lesser ) had each forecast been more ( or less ) certain . Specifically , we perturbed each forecast by taking the density at each possible prevalence to the 1+ϵ power , normalizing , and determining the likelihood of the observed data . Note that this maintains the support of a forecast , maintains the ordering of the outcomes , and increases the Fisher’s information proportionally by ϵ ( or decreases information proportionally for ϵ<0 ) .
At the baseline census , communities had a mean of 146 children ( 95% CI 137 to 155 ) aged 0 to 5 years . The mean antibiotic coverage of children was 92 . 3% at baseline , 89 . 0% at 12 months , and 89 . 8% at 24 months . At baseline , the estimated prevalence of infection in the 24 communities ranged from 2% to 58% with a mean prevalence of 21 . 1% ( 95% CI 19 . 8% to 22 . 5% ) [18] . The community prevalence of infection at each biannual visit is displayed in Table 1 . The observed prevalence of infection at 36 month which was to be forecasted ranged from ranged from 0% to 22 . 5% with a mean prevalence of 5 . 8% ( 95% CI 5 . 2% to 6 . 4% ) . The 15 experts provided forecasts for each of the 24 communities , with the mean taken as the community forecast ( Fig 1 ) . Fig 2 shows the forecast distributions for the community of experts , regression , and the SIS model , and Table 2 ranks the likelihood of the observed 36-month prevalence for each ( S1 Fig and S1 Table in Supporting Information show the difference between observed and forecast prevalence ) . The estimated parameters of the SIS model with random effect are shown in Table 3 . The SIS model and the square root-transformed regression had significantly better likelihood than the experts ( p = 0 . 004 and p = 0 . 01 , respectively ) , and than the linear regression ( p = 0 . 01 and p = 0 . 02 , respectively ) . All forecasts were positively biased , on average estimating a greater prevalence than was observed . All forecasts had a lower ( worse ) likelihood if their Fisher’s information was marginally increased . No individual expert forecast was better than the community forecast ( the mean of the 15 experts ) . A priori , the SIS model assumed a mean duration of infection of 6 months , obtaining a loglikelihood of the observed 36 month data of -41 . 03 . Had we assumed the mean duration of infection was 3 months or 12 months , the loglikelihood would have been -41 . 47 or -39 . 91 , respectively . If we had assumed the 6 month duration of infection , but did not use a community-level random effect , the likelihood score would have been -41 . 57 . To fairly compare the different methods , the distribution of each forecast was estimated by minimizing the Fisher’s information given the estimated median and 95% CrI . For the SIS model , we also expressed each full distribution , obtaining a loglikelihood score of -40 . 90 , or nearly the same as the -41 . 03 obtained from minimizing the Fisher’s information . The mean number of trachoma citations on PubMED by the experts was 42 ( range 0 to 133 ) . The likelihood score and number of publications was actually inversely correlated ( Spearman’s correlation -0 . 33 , p = 0 . 24 ) , thus we were unable to demonstrate that this measure of expertise was associated with better forecasting . We performed logistic regression , assuming the individual PCR results most likely to have obtained the observed pooled results , but this performed no better than linear regression of the square-root transformed regression .
An SIS hidden Markov model and a regression model both produced forecasts with significantly higher likelihood of the observed data than a community of experts . The SIS model , which attempted to utilize an understanding of the infectious process and mass treatment , performed significantly better than linear regression , but only slightly ( and not significantly ) better than regression of the square root-transformed prevalence . In general , more uncertainty resulted in better scoring forecasts . For every forecast a mathematical perturbation which reduced the Fisher’s information resulted in a higher likelihood of the observed data . The inclusion of a community-level random effect in the SIS hidden Markov model improved forecasting , perhaps by increasing uncertainty . The composite survey contained less information than any individual survey , and did better than any single individual forecast . The benefit of adding uncertainty could suggest that forecasts are inherently over-confident , or that additional variance components of the data were not considered by any of the methods . Even though the SIS hidden Markov model and regression model had significantly higher likelihood than the community experts , the forecasted distributions of prevalence ( as shown in Fig 2 ) by all models were very similar and did not show which model was significantly better than other models . With more available data , models could improve forecasting . The SIS hidden Markov model did not include infection from outside the population of children aged 0–5 years in each community . Our previous model [13] used a simple constant exogenous infection rate to represent infection from older children or adults to children aged 0–5 years within the same community , and did not find significant differences between the estimated transmission coefficients with and without the exogenous infection rate for different durations of infection . Of course , such models could be further refined to reflect age structured transmission dynamics . In this setting , the other age groups ( older children and adults ) were being treated as well , and other studies have shown consistently higher prevalence in small children than in other age groups ( e . g . [25] ) . The prevalence of infection in different communities is clearly correlated visit-to-visit , with visits 6-months apart having a higher correlation than visits further apart . However , there may be a fundamental limit to the predictability at the community level , simply due to the vagaries of who infects whom and when they do so . Mathematical models and cross-sectional empirical studies have suggested that as disease is disappearing , the prevalence of infection should form an exponential distribution ( or its discrete analog , a geometric distribution ) , whether the disappearance is due to mass antibiotics , environmental improvements , or a secular trend [26 , 27] . This exponential distribution has a much heavier tail than , for example , the normal distribution , so outliers are to be expected even when all communities are assumed to have identical transmission characteristics . Six-months is a relatively short period in trachoma control—programs typically reassess endemicity every 1–5 years . If predictability decreases as time increases between visits , than we would expect that apparent hotspots at one visit may not be the most affected areas at a subsequent visit . This has been termed chasing ghosts by trachoma programs ( personal communication , PME ) . Forecasts , whether made by experts , statistics , or mathematical transmission models , are rarely done in a falsifiable manner . Here , all participants were presented with identical information and masked to the results and to the other forecasters . Forecasts described the distribution of all possible outcomes , not a prediction of the single most favorable , and were scored in a pre-specified manner . The availability of results from 24 communities allowed a statistical comparison between forecasts , reducing the chance that the overall score would be dependent on a single fortunate guess . Current WHO guidelines for starting mass drug administration are based on the district prevalence of the clinical signs of disease rather than infection , and future studies could assess forecasting at that level . In this study , we forecasted the community-level prevalence of ocular chlamydial infection . WHO guidelines currently include sub-district level intervention , at least for hypo-endemic districts with 5–10% prevalence of clinical activity in children . Individual community-level forecasting may become important for surveillance after mass antibiotic administrations have been discontinued . Programs currently make decisions based on recommendations offered by the WHO [1] . Guidelines have relied heavily on extrapolation of existing evidence and expert opinion , since not all scenarios have been , or likely will ever be , tested in community-randomized trials . Forecasting at the individual community level has not been particularly successful . While forecasting at the district level may be more feasible than forecasting at the individual community level , statistical and transmission model forecasts should be evaluated . If proven more effective , as they were in this setting , then it may be reasonable for programmatic decisions to be based on statistical or modeling forecasts rather than just expert opinion . | Clinicaltrials . gov NCT00792922 | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| []
| 2015 | Short-term Forecasting of the Prevalence of Trachoma: Expert Opinion, Statistical Regression, versus Transmission Models |
The lack of effective and well-tolerated therapies against antibiotic-resistant bacteria is a global public health problem leading to prolonged treatment and increased mortality . To improve the efficacy of existing antibiotic compounds , we introduce a new method for strategically inducing antibiotic hypersensitivity in pathogenic bacteria . Following the systematic verification that the AcrAB-TolC efflux system is one of the major determinants of the intrinsic antibiotic resistance levels in Escherichia coli , we have developed a short antisense oligomer designed to inhibit the expression of acrA and increase antibiotic susceptibility in E . coli . By employing this strategy , we can inhibit E . coli growth using 2- to 40-fold lower antibiotic doses , depending on the antibiotic compound utilized . The sensitizing effect of the antisense oligomer is highly specific to the targeted gene’s sequence , which is conserved in several bacterial genera , and the oligomer does not have any detectable toxicity against human cells . Finally , we demonstrate that antisense oligomers improve the efficacy of antibiotic combinations , allowing the combined use of even antagonistic antibiotic pairs that are typically not favored due to their reduced activities .
Antibiotic resistance is an important public health problem that emerged shortly after the discovery of antibiotics [1 , 2] . Pathogenic bacteria are either intrinsically resistant to some antibiotics or they acquire resistance via spontaneous mutations or horizontal gene transfer . These resistance mechanisms include deactivation or modification of antibiotics , pumping out antibiotics via efflux pumps , protection of antibiotic targets , and mutations in the target enzymes that decrease antibiotic affinity [3] . Even though the majority of these resistance mechanisms are well characterized at the molecular level , there has been limited success at avoiding the evolution of resistance in the clinic . There is a growing need for entirely new tools and strategies in order to stop or slow the evolution of antibiotic resistance in clinical settings [4] . Recent advances in biology , particularly whole genome sequencing technologies and gene-editing tools , have enabled us to identify resistance-conferring genetic changes and perform genetic manipulations that can reverse evolved antibiotic resistance [5–7] . By using novel gene-editing tools such as CRISPR-CAS9 or engineered bacteriophages , it is now possible to edit bacterial genomes to modulate antibiotic sensitivity of bacteria and also design sequence-specific antimicrobials [8–10] . However , gene-editing tools are currently difficult to implement given the practical and ethical problems with mutating bacterial genomes within an infected human patient . Instead , we designed antisense oligomers , which target the mRNA of bacterial resistance genes , preventing translation in a sequence-specific manner [11] . Briefly , phosphorodiamidate morpholino oligomers ( PMOs ) are synthetic nucleotide oligomers made from six-membered morpholine rings joined together by phosphorodiamidate linkages . Each morpholine ring has a natural nucleobase attached ( see [12] for PMO structure ) , and the oligomers are designed to bind complementary sequences in targeted mRNAs . PMOs are thought to exert their effects through translation inhibition as a result of steric hindrance when targeting bacterial mRNA within the close proximity of ribosome binding sequences [13] . Cell-penetrating peptides are conjugated to the phosphorodiamidate morpholino oligomers ( PPMOs ) , which enhance uptake of the oligomer into the bacterial cell [14] . Unlike short RNA molecules , which are also being considered as therapeutic agents , the synthetic PMO backbone renders them resistant to being hydrolyzed by nucleases [12 , 15] . Previous reports have shown PPMOs to be bactericidal in vitro and in vivo in a number of gram-negative pathogens when targeting essential genes [11 , 16] . Here , we demonstrate that PPMOs inhibit several resistance-conferring genes , improving efficacy of several distinct antibiotic classes .
Active excretion of antibiotic molecules via efflux proteins or reducing the uptake of drug molecules by mutating or down-regulating membrane proteins ( i . e . , porins ) are two of the common strategies that multidrug-resistant bacteria utilize in order to render antibiotics ineffective [3] . We , and others , have previously shown that several genes that encode membrane proteins in multidrug-resistant E . coli strains either accumulated point mutations or had changes in their regulation [5 , 17–20] . Thus , we hypothesized that deletion of such genes has the potential to increase antibiotic efficacy ( Fig 1A ) . To test this idea , we selected five genes ( acrB , emrB , marB , ompF , cmr ) that encode membrane proteins in E . coli and deleted them with all 32 possible combinations in order to find the best target genes and quantify epistatic interactions between these gene deletions ( S1 Fig ) [3 , 21–24] . We then measured the minimum inhibitory concentrations ( MICs ) of these mutants against 27 different antibiotics ( Fig 1B and 1C and S2 Fig ) . Deletion of acrB , alone or in combination with other genes , significantly increased the susceptibility of E . coli to several antibiotics , up to ~100-fold ( Fig 1B and 1C and S2 Fig ) . However , deletions of the other genes ( emrB , marB , ompF , cmr ) did not significantly change antibiotic susceptibility ( Fig 1B and 1C and S2 Fig ) . This suggests that these genes might be involved in acquired resistance when they are mutated or their regulation is altered , but they are not involved in intrinsic antibiotic resistance of E . coli against the 27 compounds we tested . Also , based on these measurements , there were no epistatic interactions between these gene deletions . The AcrAB-TolC efflux pump complex is among the best-characterized efflux pumps in E . coli and is composed of AcrB , the inner membrane antiporter , AcrA , the periplasmic adaptor protein , and TolC , the outer membrane channel ( Fig 2A ) [20 , 22 , 25–27] . Deleting acrB led to increased susceptibility ( Fig 1B and 1C ) , so we deleted the two other genes ( acrA and tolC ) that together form the AcrAB-TolC efflux pump complex ( Fig 2A ) in E . coli to identify their contribution to the intrinsic antibiotic resistance of E . coli . Indeed , deletion of any of these three genes increased antibiotic sensitivity of E . coli , and loss of intrinsic antibiotic resistance due to gene deletions was reversed by plasmid complementation ( S3 Fig ) . We designed three PPMOs to target the acrA , acrB , and tolC genes , respectively ( Fig 2A–2C ) . PPMOs were designed as 11-mers targeting gene regions near the translation start site with high-sequence specificity and low homology around other translation start sites in the E . coli genome ( Fig 2C ) . We first tested the efficacy of these PPMOs by quantifying inhibitory effects of several antibiotic compounds against E . coli in the presence of PPMOs ( Fig 2E , S4 Fig , S2 Table ) . All three PPMOs designed to target acrA , acrB , and tolC ( hereafter called acrA-PPMO , acrB-PPMO , and tolC-PPMO ) induced antibiotic sensitivity to multiple antibiotics , while a control PPMO ( control-PPMO ) , which has a base sequence with low homology to E . coli translation start sites , had no effect on antibiotic sensitivity ( Fig 2E and S4 Fig ) . Fig 2E demonstrates an example of this sensitizing effect with clindamycin , a protein synthesis inhibitor that is not commonly used against E . coli infections because of its high MIC . Enhancing the efficacy of clindamycin against E . coli is a significant finding that could make this drug potentially effective against gram-negative bacteria . Strikingly , use of acrA-PPMO ( Fig 2E , blue line ) showed a ~16-fold increase in clindamycin sensitivity comparable to the ~32-fold increase from deletion of the acrA gene ( Fig 2E , cyan line ) . In almost every PPMO and antibiotic combination , the effects of acrB-PPMO and tolC-PPMO were less potent than the effect of acrA-PPMO ( S4 Fig ) . Hence , we used acrA-PPMO for the rest of our experiments . The sensitization effect of acrA-PPMO varied between a 2- and 40-fold reduction in MIC , depending on the antibiotic compound ( S4 Fig , S2 Table ) . It was surprising to find that , for certain antibiotic compounds , acrA-PPMO treatment did not adequately recapitulate the effect seen with the acrA deletion mutant ( e . g . , compare chloramphenicol and oxacillin , S4 Fig ) . In order to further compare the phenotypic effects of the acrA deletion mutant to acrA-PPMO silencing , we tested the sensitization effect of acrA-PPMO with ten antibiotic compounds ( Fig 3B ) . Five of these compounds were selected because they were more potent against E . coli strains with the acrA gene deletion , and the remaining five antibiotic compounds were selected since their efficacies were not expected to change based on the acrA gene deletion data ( S2 Fig ) . Fig 3A presents two example antibiotic dose-response curves demonstrating the effect of acrA-PPMO when used together with cefotaxime or meropenem . Susceptibility of E . coli to cefotaxime increased when acrA was deleted or silenced ( Fig 3A , left ) , whereas susceptibility to meropenem remained the same as the wild-type when acrA was deleted or silenced ( Fig 3A , right ) . This pattern was consistent with the previous susceptibility data in all antibiotics tested ( Fig 3B ) . In other words , the phenotypic effect of silencing the acrA gene with acrA-PPMO was indistinguishable from the phenotypic effect of acrA deletion ( r = 0 . 94 , p < 0 . 001 , Pearson correlation test ) . The nucleotide sequence near the translational start site is conserved between several bacterial genera for acrA ( Fig 2C ) . We therefore hypothesized that acrA-PPMO would sensitize organisms with high sequence homology and would have no effect on organisms with low sequence homology . To demonstrate this , we compared the efficacy of acrA-PPMO against E . coli , Klebsiella pneumoniae , Salmonella enterica , Acinetobacter baumannii , Pseudomonas aeruginosa , and Burkholderia cenocepacia , which share between 36% and 100% acrA sequence homology to the E . coli target sequence ( Fig 2C ) . In E . coli , time-kill assays after 18 h of exposure to subinhibitory concentrations ( 1/4 MIC ) of piperacillin-tazobactam and 10 μM acrA-PPMO resulted in a three order of magnitude reduction in colony forming units ( CFUs ) from the starting inoculum compared to a three order of magnitude increase at the same concentration of antibiotic alone ( Fig 3C ) . We demonstrated a similar sensitization effect of acrA-PPMO against K . pneumoniae and S . enterica , which share 100% sequence homology with E . coli ( Fig 3C , top ) . Conversely , the acrA-PPMO had no activity against A . baumannii , P . aeruginosa , or B . cenocepacia , consistent with their lower ( 36%–45% ) sequence homology ( Fig 3C , bottom ) . Importantly , this demonstrated that the sensitization effect of acrA-PPMO was sequence-specific , and our strategy has the potential for being used against other pathogens . We quantified the AcrA protein expression in E . coli in increasing concentrations of acrA-PPMO ( Fig 4A , top panel ) . AcrA protein levels decreased nearly 30-fold at acrA-PPMO concentrations greater than 3 μM ( Fig 4A , middle panel ) . Control-PPMO had no effect on AcrA protein levels at 2 and 10 μM ( S5 Fig ) . Residual expression of AcrA ( ~2% compared to untreated cells ) is still detected even at the highest acrA-PPMO dose . We have also verified this effect by measuring growth rates of E . coli at different subinhibitory clindamycin concentrations using increasing concentrations of acrA-PPMO . Growth of E . coli , incubated with constant clindamycin concentrations , gradually decreased as acrA-PPMO concentration was increased ( Fig 4A , bottom panel ) . This indicated that clindamycin susceptibility was correlated with the AcrA expression in E . coli ( Fig 4A ) . Clindamycin sensitivity of E . coli , even at the highest concentrations of acrA-PPMO , was still lower than the sensitivity of the E . coli mutant with acrA deletion ( Fig 4A , bottom panel ) , which is consistent with the residual AcrA expression even at the highest concentrations of acrA-PPMO ( Fig 4A , top panel ) . To directly demonstrate that the acrA-PPMO’s inhibition of AcrA translation leads to reduced antibiotic efflux , we measured efflux of a DNA-binding dye , Hoechst 33342 , which is also a substrate for the AcrAB-TolC complex [28] . The rate of fluorescence accumulation inside bacterial cytoplasm reflects the difference between concentration-dependent influx of Hoechst dye and the AcrAB-TolC-related efflux of the Hoechst dye . We found that E . coli cells treated with 2 and 10 μM of acrA-PPMO had significant increases in final fluorescence levels and fluorescence accumulation rates , comparable to the acrA deletion mutant ( S6 Fig ) . This surrogate measure suggests that the efflux of antibiotic compounds is qualitatively similar to the efflux of the Hoechst dye; however , the magnitude of efflux will be specific to the chemical structure of particular antibiotics . Finally , we tested the toxicity of acrA-PPMO against human lung epithelial cells using a cell viability assay ( Fig 4B ) . Even at 19 . 2 μM , acrA-PPMO had no significant toxicity at the end of 4 d . These observations provide clear evidence that acrA-PPMO is a promising agent that works as an efficient antibiotic adjuvant by preventing AcrA translation and therefore preventing efflux in a sequence-specific way . One strategy often employed for treatment of severe bacterial infections is the combined use of two or more antibiotics with different mechanisms of action [29] . Particularly , the use of antibiotic pairs that display synergy is considered to be advantageous in clinical practice [30 , 31] . One risk of this approach is that several synergistic antibiotic pairs may promote evolution of multidrug resistance if they have overlapping resistance mechanisms [5 , 32 , 33] . Conversely , several antibiotic pairs that are less likely to promote resistance cannot be used in combination due to antagonistic drug—drug interactions [5 , 30] . Therefore , strategies that could rescue the use of antagonistic drug combinations could be of significant benefit [34] . Successfully enhancing antibiotic susceptibility by blocking efflux activity has three potential outcomes in antibiotic combination therapies ( Fig 5A ) . It could increase susceptibility to either antibiotic independently ( Fig 5A , left and middle ) , or it could increase susceptibility to both drugs simultaneously ( Fig 5A , right ) . We tested the acrA-PPMO together with antibiotic pairs to see if we could improve their antimicrobial efficacy . Here , we demonstrate that sensitizing bacteria against antibiotics by targeting acrA with the acrA-PPMO can increase sensitivity to both synergistic and antagonistic pairs . We quantified pairwise interactions between trimethoprim and sulfamethoxazole versus trimethoprim and piperacillin-tazobactam , in the presence and absence of acrA-PPMO ( Fig 5B ) . Trimethoprim and sulfamethoxazole are antifolate antibiotics that block the activity of dihydrofolate reductase and dihydropteroate synthase , respectively . Trimethoprim is often used together with sulfamethoxazole due to their synergistic interaction [30] . Conversely , using trimethoprim with piperacillin-tazobactam could be problematic since these two drugs were previously reported to antagonize each other’s activities [30] . We created two-dimensional gradients of trimethoprim-sulfamethoxazole ( Fig 5B , left ) or trimethoprim-piperacillin/tazobactam ( Fig 5B , right ) and determined MIC values for the wild-type E . coli in the presence ( Fig 5B , blue lines ) and absence of acrA-PPMO ( Fig 5B , black lines ) or with the acrA deletion ( Fig 5B , cyan lines ) . We compared the enhancement of drug combinations by acrA-PPMO by integrating the AUC of the resulting MIC curves ( Fig 5C ) . We found that acrA-PPMO increases the efficacy of both synergistic and antagonistic pairs by nearly 5-fold and 15-fold for trimethoprim-sulfamethoxazole ( Fig 5C , left ) and trimethoprim-piperacillin/tazobactam combinations ( Fig 5C , right ) , respectively . This observation clearly indicates that even though trimethoprim and piperacillin-tazobactam have antagonistic interactions , acrA-PPMO significantly ( p < 0 . 001 ) increases the efficacy of the trimethoprim-piperacillin-tazobactam combination in E . coli . This finding has the potential of making the trimethoprim-piperacillin-tazobactam combination a promising candidate for treating infections since trimethoprim and piperacillin-tazobactam have independent resistance mechanisms that make the emergence of cross-resistance less likely [5 , 7] . PPMO treatment did not change the shape of the MIC curves for both of the trimethoprim-sulfamethoxazole ( Fig 5B , left ) and trimethoprim-piperacillin/tazobactam ( Fig 5B , right ) combinations , but rather rescaled the MIC curves towards the origin compared to the wild type ( Fig 5B ) , as was previously demonstrated for other antibiotics by Chait et al . [34] . We conclude that the acrA-PPMO did not affect the drug interaction mechanisms , but rather , the decreased efflux of both antibiotic compounds resulted in increased effective antibiotic concentrations inside bacterial cells .
In this study , we demonstrate that we can strategically induce antibiotic hypersensitivity in pathogenic bacteria by targeting the genes that encode for the AcrAB-TolC efflux system with PPMOs . Antibiotic molecules that traverse the bacterial membrane can therefore remain intracellular for longer time periods leading to increased antibiotic activity . This could have also been achieved by using efflux pump inhibitor molecules , such as phenyl-arginine-β-naphthylamide , that block AcrAB-TolC activity [35–39] . However , efflux pump inhibitors are known to have significant toxicities and currently have limited use as therapeutic agents [35 , 36] . The method we introduced for inducing antibiotic hypersensitivity with the use of PPMOs does not exhibit cytotoxicity in the human cell line we tested and does not require editing the genome of the targeted bacteria in the human host . There are several advantages associated with increasing antibiotic susceptibility of pathogens by using PPMOs . First , though this is an in vitro demonstration , it suggests the potential of using lower doses of existing antibiotics , which may lead to fewer adverse effects of those agents . Second , being able to sensitize a specific bacterial pathogen and inhibit its growth by lower antibiotic doses could have the potential to minimally perturb beneficial members of healthy human microbiota . Third , by using PPMOs , we may have the opportunity to use several drugs for treating infections against which they are normally not effective , such as oxacillin against gram-negative bacteria ( Fig 3B and S4 Fig ) . Finally , increasing antibiotic efficacy with PPMOs may change the way we typically design combinatorial antibiotic therapies: by using PPMOs to minimize the antibiotic doses necessary in antibiotic combinations , we may be able to use antibiotic pairs even if the two drugs somewhat dampen each other’s inhibitory effects . However , pharmacokinetics of PPMOs and potential antibiotic combinations should be considered when designing combination therapies for maximum antimicrobial activity [40] . In addition to the benefits described above , the sequence-specificity of PPMOs allows for the ability to target a single genus , or multiple genera , if the PMO target sequence is conserved ( Fig 3C ) . As we have previously reported in Burkholderia , significant reductions in efficacy ( > 8-fold ) can occur with even single base mismatches in the PMO sequence [11] . Additionally , Tilley et al . demonstrated that four silent mutations were sufficient to render a targeted PMO ineffective in E . coli [41] . However , the relationship between mismatches , including number and where they occur spatially in the oligomer sequence , and impact on efficacy have not been thoroughly described and warrant future study . We conclude that targeting resistance genes with PPMOs is a plausible strategy to increase antibiotic susceptibility in pathogenic bacteria . Further studies are needed to extend our in vitro experiments to animal models of infection to bridge the gap between our in vitro experiments and translational studies in humans . Utilizing sequence-specific PPMOs that do not have antimicrobial activity when used alone has the potential advantage of avoiding classic selection pressure exhibited by traditional antimicrobials . Importantly , acquired bacterial resistance to PPMOs has thus far only been described in the context of PPMOs designed against essential genes and was found to be related to the peptide moiety and not the oligomer sequence [42] . Attachment of a different peptide to the same oligomer was able to rescue PPMO activity , indicating possible paths towards dealing with the development of resistance . Given the narrow pipeline for new antibiotics and the increasingly urgent worldwide problem of antibiotic resistance , innovative therapeutic approaches such as utilizing PPMOs could serve an important medical need in the future . Future studies will be conducted to systematically test the strategies we propose in this paper in preclinical in vivo models to bridge the gap between in vitro experiments and human studies .
Bacterial cells were grown at 37°C in M9 minimal medium ( 248510 , Difco ) supplemented with 0 . 4% glucose ( 50-99-7 , Fisher Scientific ) and 0 . 2% amicase ( 82514 , Sigma ) , if not stated otherwise . All E . coli strains were wild type K-12 derivatives of the parent strain BW25113 . The deletion strains were generated using the Keio Collection [43]; ΔacrA , ΔacrB , Δcmr , ΔemrB , ΔmarB , ΔompF , ΔtolC were obtained from the E . coli Genetic Stock Center with stock codes 11843 , 8609 , 8865 , 10099 , 9314 , 8925 , 11430 , respectively . These strains were used for making P1 lysates and generating mutant strains with multiple gene deletions by P1 phage transduction [44] . Kanamycin resistance marker genes were removed after every cloning step [43 , 45] . The antibiotics used in this study are: Ampicillin ( A1593 , Sigma-Aldrich ) , Carbenicillin ( C3416 , Sigma-Aldrich ) , Cefotaxime ( 454950050 , Acros Organics ) , Cefoxitin ( C4786 , Sigma-Aldrich ) , Chloramphenicol ( C0378 , Sigma-Aldrich ) , Ciprofloxacin ( F17850 , Sigma ) , Clindamycin ( 21462-39-5 , RPI ) , Doxycycline ( D9891 , Sigma-Aldrich ) , Erythromycin ( E5389 , Sigma-Aldrich ) , Fusidic acid ( F0881 , Sigma-Aldrich ) , Gentamycin ( PRX1002-Premier Pro RX ) , Kanamycin ( 60616 , Sigma-Aldrich ) , Levofloxacin ( 28266 , Sigma-Aldrich ) , Lomefloxacin ( L2906 , Sigma ) , Meropenem ( NDC6332350720 , Fresenius Kabi LLC ) , Nitrofurantoin ( N7878 , Sigma-Aldrich ) , Oxacillin ( NDC25021-162-24 , Sagent Pharmaceuticals ) , Penicillin ( P8396 Sigma-Aldrich ) , Piperacillin/tazobactam ( NDC60505-0688-4 , Apotex Corp . ) , Rifampicin ( R3501 , Sigma-Aldrich ) , Spectinomycin ( 85555 , Sigma-Aldrich ) , Spiramycin ( S9132 , Sigma-Aldrich ) , Sulfamonomethoxine ( 32091 , FLUKA ) , Tetracycline ( 87128 , Sigma-Aldrich ) , Tobramycin ( T4014 , Sigma-Aldrich ) , Trimethoprim ( T7883 , Sigma-Aldrich ) , Vancomycin ( NDC67457-340-00 , Mylan ) . All antibiotic solutions were prepared by following manufacturers’ instructions . PMOs were synthesized as previously described [46] . The cell-penetrating peptide ( RXR ) 4XB , where R is arginine , X is aminohexanoic acid , and B is beta-alanine , was synthesized using standard FMOC chemistry and purified to >95% purity at CPC Scientific ( Sunnyvale , CA ) and used without further purification . The peptide was conjugated to the nitrogen of a piperadine ring at the 5′-terminus of the PMO . First , a C-terminally reactive peptide-benzotriazolyl ester was prepared by dissolving the peptide acid with O- ( Benzotriazol-1-yl ) -N , N , N' , N'-tetramethyluronium tetrafluoroborate ( TBTU ) in 1-methyl-2-pyrrolidinone ( NMP ) . The concentration of the peptide was 50 mM . Diisopropylethylamine ( DIEA ) was added to the peptide solution at molar ratios of peptide acid:TBTU:DIEA of 1 . 0:1 . 5:1 . 5 , respectively . Immediately after the addition of DIEA , the peptide solution was added to a DMSO solution containing the PMO ( 20 mM ) at a 1:0 . 8 molar ratio . After stirring at 25°C for 3 h , the reaction was stopped by adding a 4-fold volumetric excess of water . 1 M H3PO4 was added to crude conjugated PMO in 50 μL aliquots until pH 3 was reached . After stirring at 25°C for 30 min , the reaction was neutralized by adding 1 M Na2HPO4 in 100 μL aliquots until pH 7 was reached . The resulting solution was loaded onto a Source 30s ( Sigma , St . Louis , MO ) column . The unconjugated PMO and other reaction products were purified by elution with a 1 . 5 M Guanidine-HCl solution in 20 mM NaH2PO4 with 25% MeCN in Milli-Q water at pH 6 . 5 from 0%–50% over 12 columns volumes . Fractions were selected and pooled based on UV absorbance . Pooled fractions were diluted by adding a 5-fold volumetric excess of water and the conjugate/salt solution was then loaded onto a SPE column ( Amberchrom CG300M , Dow Chemicals , MI ) , which was subsequently washed three times with two-column volumes of water to remove salt . Finally , the ( RXR ) 4XB—PMO conjugate was eluted off the SPE column with two-column volumes of 50% MeCN and lyophilized . The final products were analyzed by matrix-assisted laser desorption ionization time of flight mass spectrometry and HPLC . The purities of the final products were >85% . The nucleotide sequence for the control-PPMO is ATCGTTGCATC , for acrA-PPMO is GTTCATATGTA , for acrB-PPMO is TAGGCATGTCT , and for tolC-PPMO is TTCATTTGCAT . Master plates of each bacterial strain were prepared in a 96-well plate format using overnight cultures in ~15% glycerol ( ~5 x 108 CFU/mL ) and stored at −80°C . The master plates were thawed prior to experiments and then used to inoculate the antibiotic plates with a pinner ( VP Scientific , VP409 ) , which transfers ~5 x 104 CFUs into each well containing ~200 μL of growth media . MIC values were determined using either end-point ( final OD600 ) analysis or calculating the AUC [47] . Briefly , for end-point MIC determination ( Figs 1 and 3 and S2 and S3 Figs ) , the 96-well plates were incubated for 22 h in a shaker operated at 37°C and then OD600 of each plate was measured using a plate reader ( Infinite M200 Pro , Tecan ) . For each strain and antibiotic pair , the MIC value was defined as the lowest antibiotic concentration at which the final OD600 was below ~0 . 04 after background correction , which is slightly above the lower detection limit of our plate reader . Preliminary experiments were conducted with four replicates for clindamycin and fusidic acid . The remaining antibiotics were run twice with biological replicates . A Pearson correlation coefficient test was used to confirm the repeatability of the measurements and the p-values for MIC reduction significance were calculated using Wilcoxon rank sum test . For MIC determination using AUC values , plates were incubated under similar environmental conditions , but in an automated robotic system so that OD600 of growing cultures were recorded as a function of time ( Fig 2D and 2E ) . Linear interpolations of the resulting growth curves ( OD600 versus time , Fig 2D ) were then integrated to calculate the AUC as a metric for growth using a custom MATLAB code ( r2016a , MathWorks ) . MIC values were defined as the concentration of antibiotic where the AUC was reduced by at least 95% compared to the AUC without antibiotic . Although both methods gave qualitatively similar results , we used the AUC method whenever possible because it is more robust to experimental noise [47] . E . coli BW25113 , K . pneumoniae F45153 ( clinical urine isolate ) , S . enterica enterica serovar Typhimurium 14028S ( a generous gift from Dr . Sebastian Winter , UTSW Medical Center ) , A . baumannii AYE , P . aeruginosa PAO1 , and B . cenocepacia complex K56-2 ( cystic fibrosis clinical isolate ) were grown overnight in cation-adjusted Mueller-Hinton II broth ( MHII , Becton , Dickinson and Co . , Sparks , MD ) at 37°C , 220 rpm . Cultures were diluted to 5 x 105 CFU/mL in fresh MHII and incubated with 2-fold dilutions of piperacillin/tazobactam alone or in combination with 10 μM control-PPMO or acrA-PPMO in a 96-well plate and incubated for 18 h at 37°C , 220 rpm . Growth controls included H2O , 10 μM control-PPMO , and 10 μM acrA-PPMO alone . The MIC was evaluated at 18 h using the final optical density values as described above; for reference , the MIC values were 4 , 2 , 2 , 128 , 4 , and 64 μg/mL for E . coli BW25113 , K . pneumoniae F45153 , S . enterica 14028S , P . aeruginosa PAO1 , A . baumannii AYE , and B . cenocepacia K56-2 , respectively . Growth controls and wells at 1- , 0 . 5- , and 0 . 25-fold the MIC of piperacillin/tazobactam alone or in combination with PPMO were serially diluted in PBS and plated on trypticase soy agar with 5% sheep blood ( Remel , Lenexa , KS ) for CFU enumeration . Experiments were performed with at least four replicates . Experiments with some of the strains were carried out with six or nine replicates , because the dose-response curves of piperacillin-tazobactam are very steep in general and it is difficult to precisely measure MIC values with 2-fold dilutions . Western blots measuring AcrA levels after PPMO addition ( 1 to 12 μM ) were performed following standard procedures using an AcrA antibody ( 1:30 , 000; generous gift from Dr . Helen I . Zgurskaya , University of Oklahoma ) and cAMP receptor protein antibody ( 1:1 , 000; BioLegend: 664304 ) . E . coli cells ( BW25113 ) were grown overnight , and final OD600 was adjusted to unity . These cells were then diluted by 104 fold in 5 mL of M9 minimal media ( supplemented with 0 . 4% glucose and 0 . 2% amicase ) and grown for 6 h at 37°C ( 220 rpm ) in the presence of increasing acrA-PPMO concentrations ( 1–12 μM final concentration ) . Cells were then washed three times with cold PBS buffer ( pH 7 . 4 ) , and bacterial pellets were lysed in 1X Laemmli sample buffer ( 5 mL/O . D . ) . Equivalent amounts of the cell lysates ( 10 μL of the above sample ) from each set were electrophoresed in a 4%–15% precast polyacrylamide gel ( 561081; BIO-RAD ) , and western blotting was performed following standard procedures . IR-labeled secondary antibodies ( IRDye 800CW ( 926–32213 ) and IRDye 680RD ( 925–68072 ) ; Li-COR ) were used for detection . AcrA protein amount was quantified using an ODYSSEY infrared imaging system ( LI-COR ) . E . coli ( BW25113 ) and the acrA gene deletion E . coli strain were grown overnight , and final OD600 was adjusted to unity . The cells were then diluted by 103-fold in M9 minimal media ( with 0 . 4% glucose and 0 . 2% amicase ) and grown for 6 h at 37°C ( 100 rpm ) in the presence of 0 , 2 , and 10 μM acrA-PPMO concentrations until the OD600 reached ~0 . 25 . The cells were washed twice in PBS buffer ( pH 7 . 4 ) and were diluted to a final OD600 of 0 . 2 and 0 . 4 ( three replicates for each cell density ) . Hoechst 33342 dye ( Thermo Fisher , 62249 ) was then added to final dye concentration of 10 μM in a 96-well plate . Fluorescence and OD600 measurements were immediately recorded every ~75 s for 10 h ( S6A Fig ) . Fluorescence of Hoechst 33342 was measured by excitation at 361 nm and emission centered at 497 nm . Fluorescence values were corrected by subtracting fluorescence of PPMOs mixed with Hoechst 33342 dye in the absence of bacterial cells . This step was crucial , since acrA-PPMO produced significant fluorescence due to its AT-rich sequence yielding high fluorescence quantum yield for the Hoechst 33342 dye . Fluorescence intensity ( FI ) values were normalized with the optical density of bacterial cultures . Fluorescence accumulation rates were calculated by fitting a line to the normalized FI values that were recorded within the first 5–10 min of the experiment when FI linearly increases ( S6B and S6C Fig ) . Final FI levels were calculated by averaging fluorescence within a time window where fluorescence and OD values remain constant ( S6D Fig ) . For reversing the antibiotic sensitivity phenotype of E . coli with efflux gene deletions , we cloned acrA , acrB , and tolC genes into the arabinose inducible pJMK001 plasmid ( Addgene ) and introduced them into gene deletion strains . The efflux pump genes ( acrA , acrB , and tolC ) were PCR amplified from the wild type ( BW25113 ) E . coli strain using the following primer sets ( acrA-forward: CATGCCATGGGGATGAACAAAAACAGAGGGTTTACG , acrA-reverse: AGCTTTGTTTAAACTTAAGACTTGGACTGTTCAGGCTG ) ; ( acrB-forward: CATCAGTCATGATGCCTAATTTCTTTATCGATCG , acrB-reverse: AGCTTTGTTTAAACTCAATGATGATCGACAGTATG ) and ( tolC-forward: CATGCCATGGGGATGAAGAAATTGCTCCCCATTC; tolC-reverse: AGCTTTGTTTAAACTCAGTTACGGAAAGGGTTATGA ) . These fragments were cloned into pJMK001 after restriction digestion ( NcoI and PmeI ) followed by ligation . These plasmids were then transformed into E . coli strains that had relevant gene deletions . For expression of the efflux genes , bacterial cultures with and without plasmids were grown in the presence of 0 . 2% arabinose in M9 minimal medium . | Antibiotic resistance is a global health threat . While genome sequencing and genetic manipulation tools have elucidated many resistance mechanisms , these tools have not yet been developed into successful therapeutics . One tool with such potential are peptide-conjugated phosphorodiamidate morpholino oligomers ( PPMOs ) , which are synthetic DNA/RNA mimics that function as antisense mRNA translation inhibitors . In this paper we use PPMOs to increase antibiotic susceptibility of bacteria . First , we identify the AcrAB-TolC efflux system as a major intrinsic resistance mechanism in E . coli . Then by targeting the mRNA of each component of this efflux system with PPMOs , we identify an acrA-PPMO as the most effective antisense molecule . Treatment of bacteria with acrA-PPMO resulted in a 2- to 40-fold increase in antibiotic efficacy , prevented translation of the AcrA protein , and inhibited efflux of antibiotic molecules without being cytotoxic to human cells . Finally , we demonstrate that acrA-PPMO is efficacious in several pathogenic bacterial genera and enhances activity of both synergistic and antagonistic antibiotic pairs when used together . This work establishes that PPMOs can potentially be used to treat infections caused by antibiotic resistant bacteria . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"antimicrobials",
"deletion",
"mutation",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"drugs",
"messenger",
"rna",
"microbiology",
"pigments",
"antibiotic",
"resistance",
"mutation",
"antibiotics",
"materials",
"science",
"pharmacology",
"oligomers",
"dyes",
"bacterial",
"pathogens",
"materials",
"by",
"structure",
"antimicrobial",
"resistance",
"medical",
"microbiology",
"microbial",
"pathogens",
"materials",
"by",
"attribute",
"fluorescent",
"dyes",
"biochemistry",
"rna",
"nucleic",
"acids",
"genetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"drug",
"interactions"
]
| 2016 | Sequence-Specific Targeting of Bacterial Resistance Genes Increases Antibiotic Efficacy |
Noncoding RNAs are integral to a wide range of biological processes , including translation , gene regulation , host-pathogen interactions and environmental sensing . While genomics is now a mature field , our capacity to identify noncoding RNA elements in bacterial and archaeal genomes is hampered by the difficulty of de novo identification . The emergence of new technologies for characterizing transcriptome outputs , notably RNA-seq , are improving noncoding RNA identification and expression quantification . However , a major challenge is to robustly distinguish functional outputs from transcriptional noise . To establish whether annotation of existing transcriptome data has effectively captured all functional outputs , we analysed over 400 publicly available RNA-seq datasets spanning 37 different Archaea and Bacteria . Using comparative tools , we identify close to a thousand highly-expressed candidate noncoding RNAs . However , our analyses reveal that capacity to identify noncoding RNA outputs is strongly dependent on phylogenetic sampling . Surprisingly , and in stark contrast to protein-coding genes , the phylogenetic window for effective use of comparative methods is perversely narrow: aggregating public datasets only produced one phylogenetic cluster where these tools could be used to robustly separate unannotated noncoding RNAs from a null hypothesis of transcriptional noise . Our results show that for the full potential of transcriptomics data to be realized , a change in experimental design is paramount: effective transcriptomics requires phylogeny-aware sampling .
Genome sequencing has transformed microbiology , offering unprecedented insight into the physiology , biochemistry , and genetics of Bacteria and Archaea [1]–[4] . Equally , careful examination of transcriptional outputs has revealed that bacterial and archaeal transcriptomes are remarkably complex [5] . Roles for RNA include regulation , post-transcriptional modification and genome defense processes [6]–[10] . However , our view of the microbial RNA world still derives from a narrow sampling of microbial diversity [11] . Additional bias comes from the fact that many microbes are not readily culturable [12] . The development of metagenomics and initiatives such as the Genomic Encyclopedia of Bacteria and Archaea ( GEBA ) project have sought to redress these biases , generating genomes spanning undersampled regions of the bacterial and archaeal phylogeny [1] , and sequencing uncultured or unculturable species through metagenomics [2] , [13]–[16] . A further source of bias in our genome-informed view of microbes derives from a protein-centric approach to genome annotation . The majority of genome sequences deposited in public databases carry limited annotation of noncoding RNAs and cis-regulatory elements , yet it is rapidly becoming clear that RNA is essential to our understanding of molecular functioning in microbes [17] . The paucity of annotations is understandable , as RNA gene annotation is non-trivial [18] , [19] . However , the increasing number of roles for RNAs uncovered through experimental and bioinformatic studies make illuminating this “dark matter” all the more urgent . Among the remarkable discoveries made are: riboswitch-mediated regulation [9] , [20] , transcriptional termination by RNA elements [21]–[23] , identification of novel natural catalytic RNAs [24]–[27] , CRISPR-mediated acquired immunity [28] , [29] , temperature-dependent gene regulation [30] , [31] , and sno-like RNAs in Archaea [32]–[34] . The Rfam database [22] , [35] provides a valuable platform for collating and characterising these and other families of noncoding RNA . However , a recent comparative analysis [36] revealed that fewer than 7% of RNA families within Bacteria and less than 19% in Archaea show a broad phylogenetic distribution ( that is , presence in at least 50% of sequenced phyla ) . Crucially , that analysis revealed that underlying genome sequencing biases were a major contributor to this pattern , and that the wider genomic sampling provided by the GEBA dataset [1] did help improve identification of broadly-conserved RNA families [36] . Tools such as RNA-seq [37] and transposon insertion sequencing [38]–[40] promise to complement comparative genomics tools for RNA family discovery , and it may be possible to use a mix of data types in the identification of RNA elements . However , to date , no systematic analysis of available data has been undertaken , suggesting ncRNAs may be hidden in the deluge of published data . We have therefore assessed the value of RNA-seq data for identification of unannotated non-coding and cis-regulatory RNA elements in bacterial and archaeal genomes . We show that numerous , hitherto uncharacterised , expressed RNA families are lurking in publicly available RNA-seq datasets . We find that poor sequence conservation for RNA families limits the capacity to identify evolutionarily conserved , expressed ncRNAs from existing genomic and transcriptomic data . Our results suggest that maximising phylogenetic distance , a sampling strategy effective for identification of novel protein families [1] , [2] , is not the most effective strategy for ncRNA identification . Instead , our results show that , for RNA element identification , sequencing clusters of related microbes will generate the greatest benefit .
To assess the relative contribution of noncoding RNAs and protein-coding genes to transcriptional output , we collected all publicly-available bacterial and archaeal RNA-seq datasets ( available as of August 2013 ) , spanning 37 species/strains and 413 datasets . For all datasets , we supplemented publicly available genome annotations with screening for additional loci against the Pfam and Rfam databases [22] , [35] , [41] , [42] , followed by manual identification of expressed unannotated regions that have previously been dubbed RNAs of Unknown Function ( RUFs ) [43] . This latter annotation yielded 922 expressed RUFs . We next examined the relative abundance of transcripts within each RNA-seq dataset , yielding an expression rank for individual transcripts . This analysis reveals that most transcriptomes are dominated by highly expressed non-coding RNA outputs ( Figure 1 ) ( P-value , Chi-square test of observed vs . expected ratios and Fisher's Exact test on the counts ) . In addition to well-characterised RNAs ( rRNA , tRNA , tmRNA , RNase P RNA , SRP RNA , 6S and sno-like sRNAs ) , and known cis-regulatory elements ( riboswitches , leaders and thermosensors - Table S1 ) , the top 50 most abundant transcriptional outputs ( Figure 1 ) across the 32 Bacteria and 5 Archaea in our dataset included a total of 308 RUFs . To assess whether highly expressed RUFs possess features commonly associated with function , we employed three criteria: 1 ) evolutionary conservation , 2 ) conservation of secondary structure , 3 ) evidence of expression in more than one RNA-seq dataset . For this analysis , we compared and ranked transcriptional outputs across species/strains ( see Methods for details ) . Based on the relative rank across RNA-seq datasets and the maximum phylogenetic distance observed across all genomes , each transcript was classified as high , medium or low expression , and high , medium or low conservation . This yielded a set of highly expressed transcripts consisting of 162 Rfam families , 568 RUFs and 1429 Pfam families . As expected [44]–[46] , conserved , highly expressed outputs are dominated by protein-coding transcripts ( Figure 2B&C ) . In contrast , transcripts that are highly expressed but poorly conserved are primarily RUFs ( Figure 2A ) . Of the 568 RUFs identified , only 25 are supported by all three conservative criteria ( conservation , secondary structure and expression ) ( Figure 2D ) , a further 138 RUFs are supported by two criteria ( Figure 2D ) . Consequently , on these criteria , the vast majority of RUFs appear indistinguishable from transcriptional noise . However , as these RUFs are among the most highly expressed transcripts in public RNA-seq data , we next considered whether our criteria were sufficiently discriminatory to identify functional RNAs . It is well established that not all functional RNAs exhibit conserved secondary structure – antisense base pairing with a target is common , and does not require intramolecular folding [47] . This indicates that criterion 2 will apply to some , but not all functional RNA elements . Criteria 1 and 3 both derive from comparative analysis: criterion 1 requires an expressed RUF to be conserved in some other genome , while criterion 2 requires an expressed RUF to be expressed in another of the datasets in our study . We therefore sought to examine how effective our comparative analyses are given that the available data represent a small sample ( transcriptomes from 37 strains ) and given that biases in genome sampling across bacterial and archaeal diversity impact comparative analysis of RNAs [36] . Effective comparative analysis requires appropriate phylogenetic distances between species under investigation [48] . For discovery of protein-coding gene families , maximising phylogenetic diversity across the tree of life has proven very effective [1] , [2] . For non-coding RNA , underlying biases in genome sampling do affect the assessment of ncRNA conservation , and adding phylogenetic diversity improves the identification of broadly conserved ncRNA families [36] . However , few ncRNAs appear conserved across broad evolutionary distances [36] . We have therefore considered how species selection impacts comparative analysis as a tool for the identification of conserved ncRNAs . To assess the effect of strain selection on our capacity to identify RNA families using comparative analysis , we first generated F84 phylogenetic distances between 2562 bacterial strains and 154 archaeal strains using SSU rRNA sequences from each strain ( see Methods for details ) . Next , for each Rfam RNA family and Pfam protein family , we identified the maximum phylogenetic distance between any two species/strains that encode a given family . We then calculated the fraction of conserved RNA and protein families for a given phylogenetic distance . This reveals a dramatic difference in evolutionary conservation of Rfam and Pfam families ( Figure 3 ) . While 80% of protein families are still conserved at the broad evolutionary distances that separate Bacteria and Archaea , the phylogenetic distance at which 80% of RNA families are conserved lies somewhere between the taxonomic levels of genus and family ( Figure 3 ) . The explanation for this rapid decay of RNA family conservation across long evolutionary time-scales is likely to be a combination of the limited abilities of existing bioinformatic tools to correctly align RNA sequences [49] and rapid turnover of non-coding RNAs during evolution [36] . These results in turn indicate that appropriate evolutionary distances for optimal comparative analysis differ greatly for protein- and RNA-coding genes . Figure 3 confirms the utility of the GEBA sampling strategy [1] , [2] for protein-coding gene identification , since maximising phylogenetic diversity permits effective identification of conserved protein-coding genes . In contrast , at the largest phylogenetic distances , less than 40% of the RNA families are amenable to comparative analysis . These results define a ‘Goldilocks Zone’ ( an evolutionary distance neither too close nor too distant ) for ncRNA analysis through comparative analysis . In order to assess the potential for existing RNA-seq data to be used for ncRNA analysis , we mapped the pairwise distances between strains covered by the RNA-seq datasets in this study . Of the 506 possible pairs ( excluding Bacteria vs Archaea ) , only 11 are in the Goldilocks Zone for RNA ( phylogenetic distance between 0 . 0118 and 0 . 0542 ) covering 9 species/strains . While five pairs of datasets are ‘too hot’ ( i . e . too close phylogenetically ) , the remaining 490 comparisons are ‘too cold’ for effective comparative RNA analysis ( Figure 3 ) . The datasets in the Goldilocks Zone span three distinct clades covering five Enterobacteria , three Pseudomanada , and two Xanthomonada ( Figure 4 ) . We next calculated the percentage of conserved RUFs for all Enterobacterial strain pairs . On average , 83% of RUFs are conserved across the Goldilocks Zone . The two E . coli strains are extremely similar , and share 99% of their RUFs , suggesting that these strains are too similar for us to robustly separate expression of bona fide RNAs from noise . While these outputs could be genuine RNAs , these strains are in the ‘too hot’ region , meaning if everything is conserved , comparative power is lost . In contrast , only 12% of RUFs are conserved between strains/species pairs in the ‘too cold’ region ( spanning clades; Figure 4 ) and of the 197 RUFs found through comparative analysis of transcriptomes within the Goldilocks Zone , only 19 show evidence of expression in another transcriptome outside of this zone . This suggests that the low number of RUFs from Figure 2D showing both conservation and expression is primarily a consequence of limited sampling . That said , mining RNA-seq data within the Goldilocks Zone permits a higher confidence in the identification of novel ncRNAs . Three examples of this are illustrated in Figure 5 . These RUFs exhibit sequence and secondary structure conservation and are expressed at high levels across multiple Goldilocks Zone transcriptomes . In summary , the Goldilocks Zone for RNA is surprisingly narrow , and suggests that optimal strain selection for RNA comparative analyses should comprise strains of the same species , members of the same genus , and closely related taxonomic families ( Figure 3 ) . Thus , the Goldilocks Zone for RNA is not encompassed by the sampling regimes currently being employed for protein family discovery .
Our analyses of over 400 publicly-available bacterial and archaeal RNA-seq datasets reveal that there is evidence for large numbers of RNAs of unknown function in public data . We find evidence for close to 1000 unannotated noncoding transcriptional outputs , but , given that RNA-seq experiments provide a snapshot of gene expression under specific experimental conditions , this number is likely to be far lower than the complete set of transcriptional outputs . Thus , the dataset we assembled for this project , which includes data generated by a number of labs and derives from various species and strains grown under a range of experimental conditions , is expected to represent a broad , though partial , census of total expression outputs across the species represented . Equally striking is the fact that , for the 922 RUFs identified in our study , over half ( 568 ) are among the most abundant transcripts . These results suggest that ncRNA may play an even greater role in the molecular workings of Bacteria and Archaea than hitherto realised . This use of transcriptome data clearly improves our capacity to identify noncoding outputs: applying three criteria ( sequence conservation , conservation of secondary structure , and expression in multiple strains/species ) we have identified 163 high-confidence expressed RUFs from public data ( Figure 2 ) . An additional 405 RUFs are highly expressed across the transcriptomes we have examined , yet these do not show clear signs of sequence or structural conservation in other sequenced genomes . Given their high expression level , these seem unlikely to be transcriptional noise . Some may represent technical artefacts , but many could be bona fide lineage-specific ncRNAs with potentially novel functions . Our results indicate that the greatest gain in analytical power for ncRNA discovery will come from phylogenetically-informed experimental design . Indeed , we find that this is critical to successful element identification , since the ‘Goldilocks Zone’ for optimal comparative analysis of RNA elements is surprisingly narrow . Hence , existing efforts to maximise phylogenetic coverage of genome space [1] , [2] need to be complemented with fine-scale sampling of the tips ( Figure 4 ) . Indeed , analysing the few transcriptomes that span the Goldilocks Zone reveals a remarkable enrichment of transcripts showing evidence of structure , conservation and expression in other strains/species . Furthermore , it is worth noting that the RNA family conservation decays as the phylogenetic distance increases ( shown in Figure 3 ) . There is a possibility that the Rfam families used for this are biased . However , if a bias exists , it is towards families with higher conservation ( as the families are constructed from published ncRNAs that are often discovered based upon sequence conservation [22] , [35] ) . Thus , we might actually be overestimating RNA element conservation , making phylogenetically informed sampling even more important . Given that isolation , cultivation and study of individual bacterial and archaeal strains can be extremely challenging [12] successful phylogeny-informed comparative RNA-seq will be a demanding endeavour , requiring complex sets of expertise spanning advanced culturing and isolation techniques , functional genomics capability and RNA bioinformatics . This places such a project beyond the reach of most individual labs . We therefore propose that comprehensive resolution of the comparative RNA-seq problem can best be resolved via a community-driven initiative: in recognition of the success of the GEBA project , we have dubbed this An RNA Encyclopedia of Bacteria and Archaea ( AREBA ) . The appropriateness of this acronym will be especially clear to Japanophones , as , in Japanese , the phrase ‘areba’ ( ) translates to ‘if there’ .
All available bacterial and archaeal genomes were downloaded from the European Nucleotide Archive ( ENA ) ( 2 , 562 and 154 genomes , respectively ) [50] . RNA-seq datasets published as of August 2013 were collected , spanning 37 species/strains , 44 experiments and 413 lanes of sequencing data ( Table S2 ) . Most of these datasets were generated on the Illumina platform [51] , with a few lanes from the SOLiD platform [52] and the 454 platform [53] . Where possible , FastQ files were downloaded , scanned for residual adapter sequences using AdapterRemoval ( v1 . 5 . 4 ) [54] , and mapped to the reference genome using Bowtie2 ( v2 . 1 . 0 ) [55] for Illumina and 454 data and BFAST ( v0 . 7 . 0a ) [56] for SOLiD data . All genomes were re-annotated for both RNA genes and protein coding genes . Non-coding RNA genes were annotated using cmsearch ( v . 1rc4 ) [57] to identify homologs of RNA families from the Rfam database ( v11 . 0 ) using the default “gathering threshold” ( cmsearch –cut_ga ) [22] , [35] . Protein coding genes were annotated using three approaches: First , annotations were parsed from the ENA files . Secondly , Glimmer ( v3 . 02 ) was run on all genomes to predict open reading frames ( with parameters “-o7 -g45 -t15” ) [58] . Thirdly , all genomes were translated into all possible amino acid sequences of length 15 or more and scanned for homologs of entries in the Pfam database of protein families using hmmsearch ( v3 . 1dev and the parameter “–cut_ga” ) [41] , [42] . From the mapped RNA-seq data , potential novel RNA genes ( designated RNAs of Unknown Function , or RUFs ) were picked manually by locating regions in the genomes that showed high levels of expression without overlapping annotated protein coding or RNA genes . Only RUFs of lengths 50 to 400 nucleotides were included , yielding a total of 844 RUFs in Bacteria and 78 RUFs in Archaea . Homologs of the identified RUFs were found in all the downloaded genomes using nhmmer [59] in an iterative fashion: First , the RUF sequence alone was used in the scan; then , all hits with E-value were included and a HMM built . This was iterated 5 times . The alignments from the RUF homology search were analyzed further by investigating the potential for secondary structure formation using RNAz [60] and alifoldz [61] . Protein coding potential of the RUFs was assessed using RNAcode [62] . Overlaps between potential RUF homologs in other strains/species and all the annotations in the respective genomes were also assessed . For each strain , the available RNA-seq datasets were pooled and a list was created of transcripts showing expression in that strain in at least one experiment ( defined as a transcript having a median depth of at least 10 reads in any experiment ) . A RUF homolog was defined as being expressed if the median read depth of the homologous region was at least 10X . Transcripts were ranked for each strain based on median expression ( i . e . the most highly expressed transcript will have rank 1 ) , which makes relative comparison across strains and datasets possible . The final set comprises 452 different Rfam families , 922 different RUFs , and 7249 different Pfam domains . For comparative analysis , if a gene was found to be expressed in more than one strain/species , the minimum rank was used ( i . e . showing the relatively most abundant expression of the gene ) . This ensures that transcripts that are always low abundance will remain low abundance , whereas genes that are highly abundant in at least one of the sampled time points and conditions will be treated as such . The ranking is used as a measure of expression . “Family conservation” is based on SSU rRNA alignments of all Bacteria and Archaea , respectively . For each genome , the best hit to the Rfam model of SSU rRNA was extracted ( RF00177 for Bacteria and RF01959 for Archaea ) . The sequences were aligned to the model using cmalign [57] . Finally , a distance matrix was calculated using dnadist [63] with the F84 model [64] , [65] which allows for different transition/transversion rates and for different nucleotide frequencies . The pairwise strain/species distances produced in this manner estimate the total branch length between any pair of strains/species . For any gene found in two or more strains/species , the maximum pairwise distance is used as the conservation score . Upper and lower quartiles of the distributions are used to define sets of high , medium and low expression and conservation , respectively . ( Expression , upper quartile: 204 . Expression , lower quartile: 1660 . Conservation , upper quartile: 0 . 478 . Conservation , lower quartile: 0 . 267 ) . We ranked datasets based on the following quality control metrics ( values reported in Table S3 ) . | We have analysed more than 400 public transcriptomes , generated using RNA-seq , from almost 40 strains of Bacteria and Archaea . We discovered that the capacity to identify noncoding RNA outputs from this data is strongly dependent on phylogenetic sampling . Our results show that , for the full potential of transcriptomics data as a discovery tool to be realized , a change in experimental design is critical: effective comparative transcriptomics requires phylogeny-aware sampling . We also examined how comparative transcriptomics experiments can be used to effectively identify RNA elements . We find that , for RNA element discovery , a phylogeny-informed sampling approach is more effective than analyses of individual species . Phylogeny-informed sampling reveals a narrow ‘Goldilocks Zone’ ( where species are not too similar and not too divergent ) for RNA identification using clusters of related species . In stark contrast to protein-coding genes , not only is the phylogenetic window for the effective use of comparative methods for noncoding RNA identification perversely narrow , but few existing datasets sit within this Goldilocks Zone: by aggregating public datasets , we were only able to create one phylogenetic cluster where comparative tools could be used to confidently separate unannotated noncoding RNAs from transcriptional noise . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"next-generation",
"sequencing",
"genome",
"expression",
"analysis",
"bacterial",
"genomics",
"genomics",
"genome",
"evolution",
"genome",
"analysis",
"transcriptome",
"analysis",
"genome",
"annotation",
"biology",
"and",
"life",
"sciences",
"comparative",
"genomics",
"molecular",
"evolution",
"computational",
"biology",
"evolutionary",
"biology",
"genetics",
"microbial",
"genetics",
"evolutionary",
"genetics",
"microbiology",
"microbial",
"genomics"
]
| 2014 | Robust Identification of Noncoding RNA from Transcriptomes Requires Phylogenetically-Informed Sampling |
The major weaknesses of subunit vaccines are their low immunogenicity and poor efficacy . Adjuvants can help to overcome some of these inherent defects with subunit vaccines . Here , we evaluated the efficacy of the newly developed water-in-oil-in-water multiphase emulsion system , termed PELC , in potentiating the protective capacity of dengue-1 envelope protein domain III . Unlike aluminum phosphate , dengue-1 envelope protein domain III formulated with PELC plus CpG oligodeoxynucleotides induced neutralizing antibodies against dengue-1 virus and increased the splenocyte secretion of IFN-γ after in vitro re-stimulation . The induced antibodies contained both the IgG1 and IgG2a subclasses . A rapid anamnestic neutralizing antibody response against a live dengue virus challenge was elicited at week 26 after the first immunization . These results demonstrate that PELC plus CpG oligodeoxynucleotides broaden the dengue-1 envelope protein domain III-specific immune responses . PELC plus CpG oligodeoxynucleotides is a promising adjuvant for recombinant protein based vaccination against dengue virus .
Dengue is the most important mosquito-borne flavivirus disease . People living in the tropical and subtropical areas are at risk of dengue virus infection , and more than 50 million dengue infected cases occur worldwide each year [1] , [2] . Vaccine inoculation is a cost-effective way of combating the threat of infectious diseases . In the past six decades , tremendous effort has been made to develop a dengue vaccine [3]–[5] . However , despite these efforts , no licensed dengue vaccines are currently available . Many advanced biological technologies have been applied to dengue vaccine development , and numbers of vaccine approaches are currently in pre-clinical or clinical development . These approaches include chimerization with other flaviviruses or the deletion of portions of the genomes to obtain live attenuated dengue vaccines , viral vector vaccines , DNA vaccines , and recombinant subunit vaccines [3]–[5] . All of the approaches are associated with different advantages and disadvantages . Among these approaches , the recombinant subunit vaccine provides the greatest degree of safety . Dengue envelope protein domain III has been shown to be involved in host receptor binding [6] , [7] , and several neutralizing epitopes have been identified within this domain [8]–[13] . These characteristics of the envelope protein domain III indicate that it would be a promising dengue vaccine candidate [14] . Several subunit vaccines based on recombinant dengue envelope protein domain III have been developed to protect against dengue viral infection [15]–[23] . Formulating dengue subunit vaccine candidates with proper adjuvants [15]–[17] , [21]–[23] or expressing vaccine candidates in a lipoprotein [18]–[20] was necessary to enhance their immunogenicity . These results indicate that one of the major weaknesses of subunit vaccines is their low immunogenicity and that appropriate adjuvants or delivery systems are required to overcome this weakness . Adjuvants and delivery systems have noticeably improved over the past several years . We previously developed a bioresorbable diblock tri-component copolymer poly ( ethylene glycol ) -block-poly ( lactide-co-ε- caprolactone ) mixed with squalene and Span®85 to produce homogeneous nano-particles ( PELC ) . This water-in-oil-in-water multiphase emulsion system can be utilized for vaccine delivery [24]–[26] . In addition , we demonstrated that a formulation of inactivated influenza virus and CpG oligodeoxynucleotides ( CpG ) could enhance both the overall immune response and cross-clade protective immunity [27] . These results indicate that PELC-formulated vaccines has improved potential efficacy . In this study , we evaluated the potential of aluminum phosphate , CpG , PELC , and PELC plus CpG as adjuvants to enhance the immunogenicity of recombinant dengue-1 envelope protein domain III ( D1ED III ) . We demonstrated that recombinant D1ED III formulated with PELC plus CpG induced stronger and broader immune responses than using other adjuvant formulations . These results provide valuable information for future clinical studies .
Animal studies were carried out in strict accordance with the recommendations from Taiwan's Animal Protection Act . The protocol was approved by the Animal Committee of the National Health Research Institutes ( Protocol No: NHRI-IACUC-098014 ) and were performed according to their guidelines . A consensus sequence for D1ED III from dengue-1 viruses was obtained by aligning five amino acid sequences from different isolates of the dengue-1 virus [21] . According to the amino acid sequence of D1ED III , the DNA sequence of the D1ED III gene was derived using codon usage of Escherichia coli and was fully synthesized using the assembly PCR method [28] . The product of the assembly PCR was then amplified by conventional PCR . To generate an expressing plasmid for recombinant D1ED III , the following primers were used: forward primer , 5′-GGAATTCCATATGaaaggcatgagctatgtgatgt -3′ ( Nde I site , underlined ) ; reverse primer , 5′- CCGCTCGAGgctgctgccttttttaaa -3′ ( Xho I site , underlined ) . The PCR product was cloned into the expression vector pET-22b ( + ) ( Novagen , Madison , WI ) , using Nde I and Xho I sites to produce the pDen1E3 plasmid . As a result , the C-terminus of the recombinant protein contained an additional hexahistidine tag ( HisTag ) . The Escherichia coli strain BL21 ( DE3 ) ( Invitrogen , Carlsbad , CA ) was transformed with the expression plasmid pDen1E3 for protein expression . The transformed cells were cultured at 37°C overnight , and protein expression was induced by adding 1 mM isopropylthiogalactoside for 20 hours at 20°C . The recombinant D1ED III was purified by disrupting the harvested cells in a French Press ( Constant Systems , Daventry , UK ) at 27 Kpsi in homogenization buffer ( 20 mM Tris ( pH 8 . 0 ) , 50 mM sucrose , 500 mM NaCl and 10% glycerol ) . The cell lysate was clarified by centrifugation ( 80 , 000× g for 40 min ) . The majority of recombinant D1ED III was found in the soluble fraction . The recombinant D1ED III was purified using immobilized metal affinity chromatography ( IMAC ) columns . The eluent from the IMAC column was then polished using an anion exchange column ( Q sepharose fast flow; GE ) after dialysis against Q buffer ( 20 mM Tris-Cl , 1 m MEDTA , pH 8 . 0 ) . E membrane ( Pall , USA ) was used to remove the endotoxin . The endotoxin levels of the purified recombinant D1ED III were determined by the Limulus amebocyte lysate assay ( Associates of Cape Cod , Inc . Cape Cod , MA ) , and the resulting endotoxin levels were found to be below the detection limit of the kit ( <3 EU/mg ) . The fractions from each step were analyzed by SDS-PAGE gel stained with Coomassie blue ( Coomassie Brilliant Blue R-250 ) and were immunoblotted with anti-HisTag antibodies . The purified recombinant D1ED III was dialyzed against 5 mM ammonium bicarbonate , pH 8 . 5 . Dialyzed samples were mixed with trypsin ( Promega Co . , Madison , WI , USA ) at a 50∶1 ratio ( wt/wt ) in 25 mM ammonium bicarbonate , pH 8 . 5 . The reaction allowed to continue for 2 hours and stopped by adding formic acid at a final concentration of 1 . 2% . The tryptic peptides were analyzed by MALDI-TOF ( Burker ) mass spectrometry . Murine CpG was synthesized by Invitrogen Taiwan Ltd and was given as a 10 µg dose dissolved in the sterile water or in the antigenic media . The CpG sequence used was 5′-TCCATGACGTTCCTGACGTT-3′ with all phosphorothioate backbones . Aluminum phosphate suspension was kindly provided by the Taiwan CDC and given as a 300 µg dose in acidic media ( pH 6 ) . PELC is a squalene W/O/W nanoemulsion stabilized by Span®85 ( sorbitan trioleate , Sigma-Aldrich , Steinheim , Germany ) and PEG-b-PLACL , the latter consisting of 75 wt-% of hydrophilic bioabsorbable PEG and 25 wt-% of lipophilic biodegradable PLACL with molecular weight of 7 , 000 daltons as previously described [24]–[26] . Briefly , 120 mg of PEG-b-PLACL , 0 . 8 mL of phosphate buffer saline ( PBS ) , and 1 . 1 mL of oily solution consisting of squalene ( Sigma-Aldrich , Steinheim , Germany ) and Span®85 ( 85/15 v/v ) were emulsified using Polytron®PT 3100 homogenizer ( Kinematica AG , Switzerland ) at 6 , 000 rpm for 5 min . The emulsified PELC formulation was stored at 4°C until use . PELC-formulated vaccine was investigated by re-dispersing 0 . 2 mL of stock emulsion into 1 . 8 mL of aqueous solution and mixed with a test-tube rotator ( Labinco LD-79 , Netherlands ) at 5 rpm for at least 1 hour before injection . Recombinant D1ED III and/or CpG were introduced in the aqueous solution , respectively . Dengue-1 ( Hawaii ) was used for this study . The virus was propagated in Vero cells , and viral titers were determined by focus-forming assays with BHK-21 cells . Five BALB/c mice ( 6–8 weeks of age ) were immunized subcutaneously with recombinant D1ED III ( 10 µg per dose , unless indicated ) . Mice were given one or two immunizations at a two-week interval with the same regimen . To detect the anamnestic response generated by immunization , immunized mice were inoculated intraperitoneally with 3×106 focus-forming units ( FFU ) of live dengue-1 virus . Blood was collected from each mouse at different time points , as indicated . Sera were prepared and stored at −20°C until use . The numbers of IFN-γ- and IL-4-producing cells were determined using mouse IFN-γ and IL-4 ELISPOT kits ( eBioscience ) , respectively . All of the assays were performed according to the manufacturer's instructions . Briefly , 96-well plates with PVDF membranes ( Millipore ) were coated with capture antibody and incubated at 4°C for 18 hours . The plates were washed twice and blocked with RPMI medium supplemented with fetal bovine serum ( 10% ) for one hour to prevent nonspecific binding in later steps . Splenocytes were seeded at a concentration of 5×105 cells/well and stimulated with D1ED III ( 10 µg/mL ) for 3 days at 37°C in a 5% CO2 humidified incubator . After incubation , the cells were removed from the plates by washing three times with 0 . 05% ( w/v ) Tween 20 in PBS . A 100 µL aliquot of biotinylated detection antibody was then added to each well . The plates were incubated at 37°C for 2 hours . The washing steps were repeated as above , and after a 45-minute incubation at room temperature with the avidin-horseradish peroxidase ( HRP ) complex reagent , the plates were washed three times with 0 . 05% ( w/v ) Tween 20 in PBS and then three times with PBS alone . A 100 µL aliquot of 3-amine-9-ethyl carbazole ( Sigma-Aldrich ) staining solution was added to each well to develop the spots . The reaction was stopped after one hour by placing the plates under tap water . The spots were counted using an ELISPOT reader ( Cellular Technology Ltd . ) . The values presented in the results are mean ± standard deviation of each group . The levels of anti-D1ED III IgG in the serum samples were determined by titrating the samples . Sera were diluted using 3-fold serial dilutions ( starting at 1∶33 ) . Briefly , purified D1ED III was coated on 96-well plates . In some experiments , supernatant obtained from dengue-1 virus infected Vero cells was coated on 96-well plates ( 2×104 ffu virus/well ) . Bound IgG was detected with HRP-conjugated goat anti-mouse IgG Fc . After the addition of 3 , 3′ , 5 , 5′-tetramethylbenzidine ( TMB ) , the absorbance was measured with an ELISA reader at 450 nm . For measurement of IgG1 and IgG2a anti-D1ED III subclass , biotin-conjugated rat anti-mouse IgG1 and rat anti-mouse IgG2a were used as detectors , and avidin-HRP was then added . Color was developed as described above . ELISA end-point titers were defined as the serum dilution that gave a 0 . 5 OD value . The serum dilution was obtained from the titration curve by interpolation , unless the OD value was less than 0 . 5 at the starting dilution ( 1∶33 ) . BHK-21 cells were infected with dengue-1 virus . Three days after infection , the cells were fixed for 15 min in 3 . 7% formaldehyde/PBS . After washing with PBS , the cells were permeabilized with 0 . 1% Nonidet P40/PBS for 15 min and blocked with 3% bovine serum albumin ( BSA ) /PBS for 30 min . Viruses in the infected cells were detected by mouse pre-immune and immune sera ( from D1ED III-immunized mice ) . After washing with PBS , antibody-labeled cells were detected using a secondary antibody conjugated with fluorescein isothiocyanate ( FITC ) . Cellular DNA was labeled by Hoechst stains . Sera were diluted using 2-fold serial dilutions ( starting at 1∶8 ) and the sera were heat inactivated prior to testing . A monolayer of BHK-21 cells in 24-well plates was inoculated with dengue-1 virus that had been pre-mixed at 4°C overnight with pre-immunization or post-immunization sera to a final volume of 0 . 5 mL . The virus titer prior to pre-mixing was about 20–40 FFU per well . Viral adsorption was allowed to proceed for 3 hours at 37°C . An overlay medium containing 2% fetal bovine serum and 0 . 8% methylcellulose in DMEM was added at the conclusion of adsorption . The infected monolayer was incubated at 37°C . After 72 hours of infection , the overlay medium was removed from the wells , and the BHK cells were washed with cold PBS . The cells were fixed for 15 min in 3 . 7% formaldehyde/PBS . After washing with PBS , the cells were permeabilized with 0 . 1% Nonidet P40/PBS for 15 min and blocked with 3% bovine serum albumin ( BSA ) /PBS for 30 min . Infected cells were detected by a monoclonal anti-dengue antibody ( American Type Culture Collection , No . HB-114 ) . After washing with PBS , antibody-labeled cells were detected using a secondary antibody conjugated to HRP . The labeling was visualized using TMB . The FFUs were counted , and the neutralizing antibody titer FRNT50 ( or FRNT80 ) was calculated as the reciprocal of the highest dilution that produced a 50% ( or 80% ) reduction of FFU compared with control samples containing the virus alone . For calculation purpose , the neutralizing antibody titer was designated as 22 when neutralizing antibody titer was less than 23 . To test whether D1ED III blocks dengue virus infection of BHK-21 cells , virus was pre-mixed with different amount D1ED III or control BSA protein as indicated for 10 min at 4°C . Viral adsorption was allowed to proceed for 3 hours at 37°C . The FFUs were determined as described above . The statistical analysis was conducted using GraphPad Prism version 5 . 02 ( GraphPad Software , Inc . ) . Data from D1ED III blocking dengue-1 viral infection were processed by a two-tailed Student's t-test . Data from ELISPOT assay were analyzed by the Mann-Whitney test . Data from ELISA and FRNT were performed by the ANOVA Bonferroni post test . Differences with a p value of less than 0 . 05 were considered statistically significant .
The D1ED III amino acid sequence is a consensus sequence of dengue virus type 1 aligned from selected target sequences ( Accession number: P27909 , P27913 , P17763 , P33478 and P27912 ) [21] . The DNA sequence of the D1ED III gene was derived using codon usage with Escherichia coli and was fully synthesized using the assembly PCR method . The PCR product was cloned into the pET22b vector for recombinant D1ED III expression . The recombinant protein contained an additional HHHHHH sequence ( HisTag ) at the C-terminus and was expressed under the control of the T7 promoter ( Figure 1A ) . The recombinant D1ED III was purified using an immobilized metal affinity chromatography ( IMAC ) column and an anion exchange column ( Figure 1B , lanes 1–4 ) . Recombinant D1ED III was detected with anti-HisTag antibodies ( Figure 1B , lanes 5–8 ) . After the lipopolysaccharide ( LPS ) was removed ( less than 3 EU/mg ) , purified recombinant D1ED III was comparatively analyzed for its immunogenicity and efficacy in animal models . Recombinant D1ED III was also digested with trypsin to assess its peptide mass fingerprinting . The molecular weight of recombinant D1ED III is 12377 . All major peaks in the spectrum at m/z 983 . 41 , 1050 . 51 , 1184 . 68 , 1310 . 55 , 1314 . 58 , 1442 . 68 , 1637 . 88 , and 2559 . 29 cover over 80% of recombinant D1ED III . The results confirmed that the major peaks in the mass spectra were derived from recombinant D1ED III ( Figure 1C ) . Dengue envelope protein domain III has been show to be involved in cellular receptor binding [6] , [7] . We hypothesized that if purified recombinant D1ED III exists in a suitable conformation , then soluble D1ED III should interfere with dengue viral infections . As shown in Figure 2 , the ability of dengue-1 virus to infect BHK-21 cells was inhibited in the presence of D1ED III in a dosage-dependent manner . Greater than 80% reduction of focus number was observed when D1ED III added to cells at a concentration of 0 . 15 mg/mL . BSA did not inhibit dengue-1 focus formation at concentrations as high as 1 . 5 mg/mL , which is 10-fold higher than D1ED III . These results suggest that D1ED III can block the cellular binding sites of the dengue-1 virus . The purified recombinant D1ED III vaccine candidate was formulated with different adjuvants and then tested for its ability to induce both T- and B-cell immune responses in mice . Groups of BALB/c mice were immunized with the different formulations two times with a two-week interval between immunizations . Animals immunized with PBS alone served as controls . One week after the second immunization , splenocytes were harvested and examined for IFN-γ and IL-4 secretion in response to three days of D1ED III stimulation . The ability of the different formulations to induce cytokine secretion varied greatly , as shown in Figure 3 . The frequency of IFN-γ spots per 106 splenocytes in mice immunized with recombinant D1ED III alone was 4 . 9±4 . 4 ( n = 5 ) spots . Splenocytes from mice immunized with recombinant D1ED III formulated with aluminum phosphate ( 21 . 6±17 . 4 , n = 9 , p<0 . 05 compared to D1ED III alone ) , CpG ( 7 . 6±4 . 3 , n = 4 , p>0 . 05 compared to D1ED III alone ) , and PELC ( 24 . 9±14 . 3 , n = 4 , p>0 . 05 compared to D1ED III alone ) showed modest increases in IFN-γ spots . Mice immunized with recombinant D1ED III formulated with PELC plus CpG elicited the highest number of IFN-γ spots ( 85 . 9±70 . 3 , n = 9 , p<0 . 05 compared to D1ED III alone , alum , or CpG ) ( Figure 3A ) . Although the formulation using PELC plus CpG elicited fewer IL-4 spots than the aluminum phosphate formulation ( 89 . 7±48 . 6 vs . 140 . 7±87 . 0 ) , the difference was not statistically significant ( p = 0 . 1903 ) . Interestingly , the PELC plus CpG formulation elicited more IL-4 spots than the formulations using PELC ( 28 . 3±9 . 7 , p<0 . 05 ) or CpG ( 25 . 1±10 . 9 , p<0 . 05 ) alone ( Figure 3B ) . Next , we evaluated the IgG antibody responses following three immunizations with two-week intervals between immunizations . Serum samples were collected from the immunized mice at different time points , as indicated in Figure 4A . Formulations of recombinant D1ED III with PELC or PELC plus CpG were highly immunogenic and generated stronger antibody responses than the other formulation groups ( p<0 . 05 by the ANOVA Bonferroni post test ) . All of the antibody responses peaked between week 4 and week 8 after the first immunization . Importantly , substantial levels of anti-D1ED III IgG antibodies were detectable for over 20 weeks after the initial priming . Sera collected from different formulation groups at week 6 were analyzed for presence of IgG1 ( Figure 4B ) and IgG2a ( Figure 4C ) . Mice immunized with the CpG formulation generated lower levels of IgG1 antibody than mice that received the aluminum phosphate , PELC , and PELC plus CpG formulations ( p<0 . 05 by the ANOVA Bonferroni post test ) . The PELC and PELC plus CpG groups had significant levels of IgG1 in comparison with the other groups ( p<0 . 05 by the ANOVA Bonferroni post test ) . The combination of PELC and CpG induced the highest levels of IgG2a ( p<0 . 05 by the ANOVA Bonferroni post test ) . As shown in Figure 4D , the ratios of IgG2a to IgG1 in mice receiving PELC plus CpG formulation were higher than those observed in mice receiving the antigen alone , aluminum phosphate , or PELC formulations ( p<0 . 05 by the ANOVA Bonferroni post test ) . These results indicate that combination of CpG and PELC could improve the IgG2a response . As the preceding experiments showed that the various formulations of recombinant D1ED III used could elicit considerable antibody responses , we wondered whether these antibodies would recognize the dengue-1 virus . To address this , we employed indirect immunofluorescence staining to evaluate antibody specificity in the sera of mice immunized with the various formulations of recombinant D1ED III . As shown in Figure 5A , naïve serum did not produce immunofluorescent reactivity with dengue-1 virus infected cells . Weak immunofluorescence signals were observed in sera obtained from mice immunized with recombinant D1ED III alone or formulated with CpG . In contrast , strong immunofluorescence signals were observed in sera obtained from mice immunized with the aluminum phosphate , PELC , and PLEC plus CpG formulations . These results suggest that the antibodies induced by recombinant D1ED III can react with dengue-1 virus . To further examine whether D1ED III raised antibodies can recognize native dengue-1 virus , ELISA was performed by using dengue-1 virus coated 96-well plates . In comparison with naïve serum , sera obtained from mice immunized with D1ED III alone or with various adjuvants produced significantly higher OD values ( p<0 . 05 by the ANOVA Bonferroni post test ) . These results demonstrated that antibodies induced by D1ED III can recognize native dengue-1 virus ( Figure 5B ) . The major objective of this study was to explore whether any of the formulations of recombinant D1ED III could induce neutralizing antibodies responses . To evaluate the dengue-1 virus neutralizing ability of the antibodies induced by vaccination , antisera from the mouse immunized with the various formulations were collected , and the neutralizing antibody titers were assessed by focus reduction neutralization tests ( FRNT ) . As shown in Table 1 , mice immunized with recombinant D1ED III formulated without adjuvant or with aluminum phosphate , CpG or PELC could not generate significant neutralizing antibody responses ( neutralizing antibody titers FRNT50<23 ) . In contrast , mice received PELC plus CpG formulation elicited detectable neutralizing antibody titers ( FRNT50 = 24 . 6 ) . As the formulation of recombinant D1ED III with PELC plus CpG could elicit the strongest cellular and humoral immune responses and neutralizing antibody responses of all the formulations we tested , we further evaluated the protective efficacy of the D1ED III with PELC plus CpG formulation . Groups of BALB/c mice were immunized with various amounts ( 3 , 10 , or 30 µg/dose ) of D1ED III three times at two-week intervals . The animals were then injected intraperitoneally with dengue-1 virus ( 3×106 FFU/mouse ) 26 weeks after the first immunization . Serum samples were collected from the mice at the indicated time points , and the neutralizing capacity against dengue-1 virus was examined . As shown in Table 2 , no significant neutralizing activity was detected in the sera obtained from naïve mice before and after viral infection ( neutralizing antibody titers <23 ) . Mice immunized with 3 , 10 , or 30 µg of D1ED III produced neutralizing antibody responses at week 6 after initial priming . The neutralizing antibody titers waned at week 24 after the first immunization . At 6 days post-viral challenge , the neutralizing antibody titers FRNT50 were 23 . 8 , 24 . 4 , and 24 . 2 in mice received vaccination with 3 , 10 , or 30 µg/dose , respectively . These results provide tangible evidence that an efficient anamnestic neutralizing antibody response was induced in mice immunized with recombinant D1ED III formulated with PELC plus CpG .
Adjuvants containing aluminum ( alum ) are currently the most widely used adjuvants in human vaccines . However , formulations of dengue subunit vaccines using alum were unable to induce complete protection against dengue virus infection [15] , [29]–[32] . In the present study , we prepared D1ED III as a dengue subunit vaccine candidate ( Figure 1 ) . The purified D1ED III formed in the proper conformation , which could occupy the cellular binding sites to reduce dengue virus infection ( Figure 2 ) . We also found that D1ED III formulated with aluminum phosphate could not induce a significant neutralizing antibody response ( Table 1 ) . Altogether , these results suggest that alum may not be suitable for dengue subunit vaccines . In the present study , we evaluated the suitability of PELC-based adjuvants to potentiate the neutralizing antibody capacity of the D1ED III in the mouse model . All the mice immunized with D1ED III formulated without or with various adjuvants could induce a D1ED III-specific antibody response ( Figure 4A ) . We also noticed that antibodies elicited in all of the immunized groups could recognize dengue-1 virus infected cells ( Figure 5A and 5B ) . However , none of antibodies exhibited a significant neutralizing capacity aside from the antibodies obtained from PELC plus CpG immunized mice ( Table 1 ) . In addition , sizeable anamnestic neutralizing antibody responses were observed in mice immunized with D1ED III formulated with PELC plus CpG ( Table 2 ) . These results indicate that PELC plus CpG is a promising potential adjuvant for dengue subunit vaccines . The antibody-dependent enhancement ( ADE ) of flavivirus infection can be inhibited by complement protein C1q . The inhibition effect is IgG subclass-specific . ADE induced by an IgG2a monoclonal antibody but not by an epitope-matched IgG1 monoclonal antibody could be inhibited by purified C1q [33] . Our results show that different adjuvant formulations can alter the ratio of IgG1/IgG2a ( Figure 4D ) . Interestingly , a significant level of IgG2a was induced in mice immunized with the PELC plus CpG formulation ( Figure 4C ) . These results indicate that D1ED III formulated with PELC plus CpG induces an IgG2a-biased response that may possibly reduce the risk of ADE when sufficient serum C1q levels are present . Alum predominantly induces a Th2 polarized response [34]–[37] featuring IL-4 production . Consistent with these findings , D1ED III formulated with aluminum phosphate induced higher IL-4 production than any of the other adjuvant formulations that we tested ( Figure 3B ) . Interestingly , the PELC plus CpG formulation induced both vigorous IFN-γ and IL-4 responses ( Figure 3 ) . IFN-γ has been shown to play an important role in antiviral activity against dengue virus [38] , [39] . The induction of strong IFN-γ responses by the PELC plus CpG formulation will be greatly advantageous in protecting hosts against dengue virus . There are some of limitations for in vivo protection studies due to the lack of a relevant mouse model of dengue infection . In the present study , we utilized dengue-1 virus as an antigenic challenge model to evaluate memory neutralizing antibody responses . Our results show that the low neutralizing antibody titers were induced and diminished at 24 weeks after immunization of D1ED III formulated with PELC plus CpG ( Table 2 ) . Interestingly , quick anamnestic neutralizing antibody responses were evoked when stimulated with dengue-1 virus in mice immunized with D1ED III formulated with PELC plus CpG but not naïve mice ( Table 2 ) . These results suggest that memory neutralizing antibody responses were induced in mice immunized with D1ED III formulated with PELC plus CpG . Our findings show that the PELC plus CpG formulation improves both the intensity and quality of the immune responses against D1ED III . Moreover , the immune responses induced by the PELC plus CpG formulation are beneficial to host protection against dengue viral infection . In conclusion , PELC plus CpG is an attractive adjuvant for dengue subunit vaccines based on recombinant envelope protein domain III . Future work should expand to test the suitability of PELC plus CpG formulations in non-human primate studies . | Dengue is a mosquito-borne disease . Infection of dengue virus can cause clinical manifestations ranging from self-limiting dengue fever to potentially life-threatening dengue hemorrhagic fever or dengue shock syndrome . In recent years , dengue has spread to most tropical and subtropical areas , making it a global health concern . Specific approaches for dengue therapy do not exist; the development of a dengue vaccine would represent a major advance in the control of the disease . Currently , no licensed dengue vaccine is available . Subunit vaccines provide a great safety strategy for developing dengue vaccine . However , the major weaknesses of subunit vaccines are low immunogenicity and poor efficacy . Here we employed dengue-1 envelope protein domain III as a model vaccine candidate and described a newly developed water-in-oil-in water multiphase emulsion system to overcome the inherent defect of subunit vaccines . We showed that emulsification of dengue-1 envelope protein domain III and CpG oligodeoxynucleotides synergistically broadened immune responses and potentiated the protective capacity of dengue-1 envelope protein domain III . These results provide valuable information for development of recombinant protein based vaccination against dengue virus and future clinical studies . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"humoral",
"immunity",
"medicine",
"viral",
"vaccines",
"immunology",
"microbiology",
"dengue",
"adaptive",
"immunity",
"immune",
"defense",
"infectious",
"disease",
"control",
"immunizations",
"infectious",
"diseases",
"biology",
"immune",
"response",
"immunity",
"virology",
"viral",
"diseases"
]
| 2012 | Dengue-1 Envelope Protein Domain III along with PELC and CpG Oligodeoxynucleotides Synergistically Enhances Immune Responses |
A fundamental stage in viral infection is the internalization of viral genomes in host cells . Although extensively studied , the mechanisms and factors responsible for the genome internalization process remain poorly understood . Here we report our observations , derived from diverse imaging methods on genome internalization of the large dsDNA Paramecium bursaria chlorella virus-1 ( PBCV-1 ) . Our studies reveal that early infection stages of this eukaryotic-infecting virus occurs by a bacteriophage-like pathway , whereby PBCV-1 generates a hole in the host cell wall and ejects its dsDNA genome in a linear , base-pair-by-base-pair process , through a membrane tunnel generated by the fusion of the virus internal membrane with the host membrane . Furthermore , our results imply that PBCV-1 DNA condensation that occurs shortly after infection probably plays a role in genome internalization , as hypothesized for the infection of some bacteriophages . The subsequent perforation of the host photosynthetic membranes presumably enables trafficking of viral genomes towards host nuclei . Previous studies established that at late infection stages PBCV-1 generates cytoplasmic organelles , termed viral factories , where viral assembly takes place , a feature characteristic of many large dsDNA viruses that infect eukaryotic organisms . PBCV-1 thus appears to combine a bacteriophage-like mechanism during early infection stages with a eukaryotic-like infection pathway in its late replication cycle .
A fundamental and general stage in viral infection is the transfer of the viral genome into the host cell . After attachment to the cell membrane , viruses that infect animal cells depend on various entry pathways , mainly consisting of endocytosis , pinocytosis , phagocytosis and variants of these strategies [1 , 2] . Thus , un-coating of viral genomes occurs inside the host cell . In contrast , most bacteriophages eject their genome into their bacterial host through the cell wall and membrane layers [3 , 4] , eventually leaving an empty capsid at the periphery of the bacterial cell . Paramecium bursaria chlorella virus-1 ( PBCV-1 ) is the prototype of the genus Chlorovirus ( family Phycodnaviridae ) that infects chlorella-like green algae and along with viruses in the Mimiviridae , Asfarviridae , Poxviridae , Iridoviridae and Marseilleviridae families , is a member of the nucleocytoplasmic large eukaryote-infecting dsDNA viruses clade [5 , 6] . Viruses belonging to this clade have recently attracted interest due to their unusual size , structural complexity , large genomes and elaborate infection cycles [7 , 8] . PBCV-1 is an icosahedral virion ( 190 nm in diameter ) that , like bacteriophages , needs to penetrate a thick host cell wall and cellular membranes to initiate infection [9 , 10] . The virus contains a single spike-like structure at one vertex [11] , which makes the first contact with the wall of its host cell [12] , the unicellular photosynthetic alga Chlorella variabilis NC64A . PBCV-1 attachment is followed by host cell wall degradation at the point of contact by a virus-packaged enzyme ( s ) [9] . As reported here , following wall degradation the viral internal membrane fuses with the host membrane , thus generating a membrane-lined tunnel through which the ~331kbp linear dsDNA viral genome and viral proteins are ejected into the host cytoplasm [10] , leaving an empty viral capsid on the cell surface [9] , a trait characteristic of bacteriophages . Once ejected , the viral genome is rapidly translocated to the host nucleus , as indicated by the finding that transcription of viral genes is detected in infected cells at 7 min post infection ( PI ) [13] . This finding , along with the fact that the virus neither encodes nor packages a recognizable RNA polymerase support the notion that at least initial viral DNA replication and transcription processes occur in the host nuclei . This notion is also consistent with recent observations revealing major morphological modifications of the host nucleus during PBCV-1 infection [14] . Indeed , no extensive morphological changes of host nuclei are detected during the replication cycle of the giant Mimivirus or the Vaccinia virus whose entire replication cycles take place in the cytoplasm [15 , 16] . These observations raise several fundamental questions . The large internal pressure generated by the highly condensed genome in bacteriophages , along with pull forces exerted by bacterial DNA-binding proteins such as RNA polymerases present in the cytoplasm , have been suggested to contribute to viral DNA ejection [4 , 17–21] . Neither of these factors can account for the ejection of the PBCV-1 genome , as the pressure generated by the PBCV-1 genome , although substantial , is significantly less than that characteristic of bacteriophages [22] , and no DNA-binding proteins are expected to be present in chlorella cytoplasm . Thus , what are the mechanisms responsible for PBCV-1 genome ejection ? Moreover , PBCV-1 genomes ejected into the host cytoplasm are rapidly translocated to the host nucleus [23] . The translocation issue is intriguing since , as shown in this study , PBCV-1 host cells are packed with thylakoid membranes that surround most of the cell periphery and hence generate a formidable barrier for DNA translocation , as are all intracellular membrane structures [24] . To obtain insights into the initial events of the PBCV-1 infection cycle we used advanced super-resolution fluorescence and electron microscopy techniques , including Stochastic Optical Reconstruction Microscopy ( STORM ) that enables sub-diffraction resolution [25] , Scanning-Transmission Electron Microscopy ( STEM ) tomography , and specific DNA labeling technologies such as Electron Microscopy In Situ Hybridization ( EMISH ) . We demonstrate that shortly after attachment to the host cell wall , PBCV-1 perforates the wall and generates a membrane-lined tunnel through the fusion of viral membrane and host cytoplasmic membrane . Viral genomes are then ejected through this tunnel and rapidly translocated to the host nucleus , possibly through viral-induced perforations of thylakoid membranes . Previous studies revealed that at late infection stages PBCV-1 generates cytoplasmic organelles , termed viral factories , in which viral assembly takes place , a feature characteristic of many eukaryote-infecting dsDNA viruses [8 , 14 , 26–36] . These findings , along with those reported here , which underline the bacteriophage-like traits revealed by PBCV-1 at early infection stages , imply that PBCV-1 uniquely combines a bacteriophage-like infection mechanism during early infection stages with a eukaryotic-like infection pathway in its late replication stages .
Previous studies of PBCV-1-infected chlorella cells demonstrated that shortly after attachment , PBCV-1 degrades the host cell wall and ejects its genome into the host cytoplasm [9] . To obtain deeper insights into viral DNA delivery , we used double-tilt Scanning Transmission Electron Microscopy ( STEM ) tomography of high-pressure-frozen and freeze-substituted ( HPF-FS ) PBCV-1-infected chlorella cells . Our tomography studies revealed that degradation of the cell wall at the virion attachment site is followed by the extension of the viral membrane towards the host cell ( Fig 1 ) . This deformation is accompanied by the protrusion of the host cellular membrane outwardly at the viral attachment site ( Fig 1A–1G; white arrowheads ) , which is likely to result from the large turgor pressure within the host cells [10] . The concomitant deformation of viral and host membranes leads to a tight proximity between these membranes ( Fig 1A and 1B ) , thus enabling fusion of the two membranes , which in turn results in the formation of a narrow membrane-lined tunnel of ~5 nm in its inner diameter and ~32 nm long ( Fig 1 , blue arrowheads ) . In addition , the PBCV-1 genome appears to undergo massive reorganization during its ejection ( asterisks in Fig 1B and 1D ) , rearranging from an apparently homogenous morphology that is spread throughout most of the internal viral core ( Fig 1A and 1B ) to a mass that is positioned at the center of the capsid ( Fig 1C and 1D ) . Upon completion of DNA ejection , empty capsids are left attached to the cell wall ( Fig 1E–1G ) , frequently near multi-layered thylakoid membranes ( Fig 1G; red arrowheads ) . In addition , the STEM tomograms revealed that the viral tunnel persists throughout the entire course of DNA delivery into the cell without changing its internal diameter . Our STEM data do not provide , however , an unambiguous answer on the fate of the tunnel following genome delivery . Fig 1H , I shows a 3-D surface reconstruction derived from the STEM tomogram depicted in Fig 1A ( S1 Movie ) . The infecting virus is attached to the cell wall ( brown layer ) and creates a hole in the wall , presumably using viral packaged enzymes . The tunnel generated following host and viral membrane protrusions and subsequent fusion is highlighted in Fig 1H and 1I ( white arrowheads and blue structures , respectively ) . Previous studies of the PBCV-1 infection cycle provided circumstantial evidence that following ejection of the viral genome into the host cytoplasm , PBCV-1 DNA and viral proteins are rapidly translocated towards and into the host nucleus [10 , 23] . Details on how this translocation occurs remain unknown . Specifically , the inherent hurdles associated with trafficking of the large PBCV-1 genomes through the crowded cytoplasm are highlighted in the STEM-derived model ( Fig 1H and 1I , S1 Movie ) that reveals numerous vesicles , cisternae and Golgi stacks in the host cytoplasm . The multiple , densely-packed chloroplast membrane stacks that surround most of the host cell ( Fig 2 ) are likely to impose an additional and particularly demanding hurdle that viral DNA needs to overcome in its trajectory towards the host nucleus . As indicated above , the internal diameter of the membrane tunnel generated by the fusion of the host and internal viral membranes is very narrow ( ~5 nm ) , presumably allowing concomitant transfer of only a single dsDNA helix along with putative viral DNA-binding proteins [22] . This result was unexpected given that the entire 331 kbp dsDNA PBCV-1 genome is translocated through the tunnel in only a couple of minutes . Such a rapid genome transfer is indicated by the finding that empty capsids attached to chlorella cells are detected already at two minutes following exposure of the cells to the PBCV-1 viruses ( Fig 1E–1G ) . To localize viral genomes and follow their trajectories we used both immuno-DNA labeling of cryo-preserved specimens as well as Electron Microscopy In Situ Hybridization ( EMISH ) technology , which allows to specifically identify viral DNA . Briefly , EMISH methodology relies on hybridization of digoxygenin-labeled DNA probes with viral DNA sequences , followed by treatment with anti-digoxygenin antibodies . In addition to detecting cellular DNA localized in the host nucleus and chloroplast , anti-DNA antibodies revealed labeled DNA extending from infecting PBCV-1 particles ( Fig 3A and 3B , black arrowhead ) , suggesting a viral genome in the process of being ejected into the cytoplasm and translocated towards the nucleus . Indeed , DNA was also detected near the host nucleus ( Fig 3C and 3D and S1A and S1B Fig ) . Within 6 min , PBCV-1 genomes were detected in the vicinity of the nucleus and inside it ( Fig 3E and 3F , S2 Fig ) . Significantly , this is the first visual evidence that PBCV-1 DNA actually enters the nucleus . The results depicted in Fig 3 imply that viral genomes are translocated as condensed structures rather than as extended , linear molecules ( note the dense labeling of viral DNA in Fig 3D and 3F ) . Specifically , analysis of EMISH sections derived from 18 PBCV-1-infected chlorella cells revealed 16 cases of condensed DNA morphologies and two extended structures . In addition , general antiDNA immunoTEM studies of nine infected cells indicated seven cells with clearly condensed morphologies and two cells with in which the structure of viral genomes could not be precisely defined . Since slices used for EMISH and immunoTEM studies were obtained from random sectioning of infected cells at diverse cell volumes , these results strongly support the notion that viral genomes are translocated as condensed structures . This finding is consistent with the notion that condensed DNA conformations enable trafficking in the dense cytoplasm milieu by facilitating the bypass of cellular obstacles [8] . Notably , in immuno-DNA labeling assays , mock-infected cells revealed DNA labeling in the nucleus and chloroplasts , consistent with the presence of DNA in these organelles ( Fig 4A and 4B ) , but no cytoplasmic DNA labeling . In addition , EMISH analysis of mock-infected cells hybridized with PBCV-1 DNA probes did not reveal any viral DNA sequences in the host cytoplasm or nuclei ( Fig 4C and 4D ) . Further validation of the specificity of viral DNA probes was obtained with PCR and hybridization assays on thin transmission electron microscopy sections of mature virions ( S3A and S3B Fig ) . Attempts to detect viral genomes inside infected cells using conventional fluorescence microscopy were unsuccessful due to diffraction resolution limit . Therefore , we used the Stochastic Optical Reconstruction Microscopy ( STORM ) technology that allows for the localization and identification of single-emitting fluorophores and reconstruction of high resolution images [25] . Our STORM studies consisted of immuno-labeling PBCV-1 capsids at 1 . 5–2 min PI with anti-capsid antibodies , followed by counterstaining with SYTOX Orange for DNA detection and localization . Fig 5A shows a capsid attached to the cell wall at the opposite side of the nucleus , and a condensed DNA structure extending from the capsid . Further STORM analyses of PBCV-1-infected chlorella cells reveal condensed viral DNA extending from capsids towards the nucleus ( Fig 5B ) . It should be noted that the viral capsids depicted in Fig 5 are either empty or almost empty , as no DNA staining was detected . Altogether , the STORM results support our immuno-DNA labeling Electron Microscopy studies , which imply that , following ejection , viral genomes effectively overcome cellular obstacles in their trajectory towards the host nucleus , presumably by assuming a condensed morphology . Empty viral capsids attached near the host chloroplast are frequently observed . This observation underscores the question how do large viral genomes bypass the multilayered thylakoid membranes ? An intriguing answer is provided by the observations that viral DNA is detected inside chloroplasts and that a discontinuity of the thylakoid membrane stacks is discerned at the point of viral DNA localization ( Fig 6 ) , implying either the use of preexisting gaps in the thylakoid membrane stacks or direct degradation of the membranes . As implied by TEM thin sections and STEM tomograms , the diameter of the perforations in the thylakoid membranes is ~5 nm . As our EMISH studies revealed viral genomes in the host chloroplasts , we conducted immuno-fluorescence assays at late PI time points to examine the notion that viruses attached to the cell wall next to chloroplasts are indeed capable of inserting their DNA through the multilayer thylakoid membranes ( Fig 7 ) . Our imaging observations , which comprise the entire volume of infected chlorella cells , demonstrate that even virions adsorbed near the host chloroplasts appear to generate viral factories . These virions are therefore infective , thus supporting the notion that viral genomes can be translocated through the thylakoid membrane stacks in their trajectory towards the host nucleus . To exclude the notion that viral DNA can bypass the chloroplast to reach the nucleus , we carried out Focused Ion Beam Scanning Electron Microscopy ( FIB-SEM ) studies . This methodology enables capturing the entire volume of a chlorella cell at high resolution . S2 Movie reveals volume imaging of a mock-infected cell , demonstrating that the chloroplast occupies the entire volume , from top to bottom on one side of the cell , implying that a virus attached near the chloroplast must deliver its DNA through it , without the ability to bypass it in order to reach the nucleus . This finding supports the results of our immuno-fluorescence assays , suggesting that these viruses are infective ( Fig 7 ) .
A model summarizing the early stages of PBCV-1 infection and highlighting the similar patterns of bacteriophage and PBCV-1 early infection cycles is depicted in Fig 8 . We demonstrate that within a couple of minutes after exposing the unicellular photosynthetic chlorella cells to PBCV-1 , the virus attaches to the host wall , using a spike located at a unique PBCV-1 icosahedral vertex [11] ( Fig 8A ) , and digests the wall at the attachment site ( Fig 8B ) . The spike complex is subsequently removed [12] , thus creating an opening in the vertex and enabling the generation of a portal . The viral genome is then ejected into the host cytoplasm through a 32 nm-long tunnel that , as shown here , is generated by the fusion of the virus internal membrane with the host membrane ( Fig 8C and 8D ) . Notably , such an infection process , including a spike-dependent viral-host attachment that is followed by the removal of the spike complex and formation of a membrane tube was reported for some bacteriophages , such as PRD1 that , like PBCV-1 , contains an internal membrane [37 , 38] . We suggest that membrane fusion is promoted by the large internal pressure within the host cell ( discussed below ) , which enables protrusion of the host cellular membrane towards the viral membrane through the virus-generated aperture in the host wall ( Fig 8C ) . The inner diameter of the tunnel , which persists throughout the process of genome delivery , is ~5 nm . Such a narrow portal is intriguing as it enables transfer of only a single double-helix DNA at a time . This linear , base-pair by base-pair DNA translocation represents an additional feature characteristic of bacteriophage genome ejection that proceeds through a narrow ‘nanotube’ membrane [38 , 39] . Significantly , this process differs from genome release pathways of other members of large dsDNA viruses such as Mimivirus , which proceeds through a large portal that allows a concomitant release of the entire genome [32] , or Vaccinia virus that similarly has a single-step release of the entire genome [40] . Genome ejection in many bacteriophages proceeds through a two-step process , whereby a first ‘push’ stage is promoted by the very high pressure generated by the tight genome packaging within capsids , which amounts to 60 atmospheres ( atm ) ( ~6 MPa ) [4 , 20 , 41] . The second stage has been suggested to involve genome pulling mediated by several putative mechanisms , including transcription-based internalization and hydrodynamic effects [3 , 4 , 18 , 21 , 42 , 43] . Estimates of the internal pressure within capsids that are based on DNA packaging densities imply that although the pressure in PBCV-1 virions is lower than typical pressures in bacteriophages , it is substantial [22] and as such is likely to contribute to the internalization process of the PBCV-1 genome . An additional barrier to PBCV-1 genome ejection is the turgor pressure inside chlorella cells , which is higher than that of bacteria [44] , suggesting that host internal pressure represents an additional barrier to PBCV-1 genome ejection . However , this hurdle is at least partially mitigated by the viral-encoded potassium ion channels ( Kcv ) located in the viral internal membrane [45] . Fusion of the virus membrane with the host plasma membrane ( Fig 8C and 8D ) results in a rapid depolarization of the host membrane , thus enabling efflux of ions and water out of the chlorella cells [46] , thus reducing the host cell turgor pressure . Significantly , several studies demonstrated that the initial infection stages of diverse bacteriophages involve depolarization of the bacterial host membranes , leading to massive efflux of positively charged ions ( mainly K+ , presumably accompanied by the efflux of additional cations as well as of diverse anions that have not yet been characterized ) and water molecules [3 , 37 , 47–49] . The findings that bacteriophages as well as the eukaryotic-infecting PBCV-1 utilize a membrane-depolarization pathway to overcome host turgor pressure further supports the notion of a bacteriophage-like process of PBCV-1 infection . The finding that the viral genome maintains a condensed morphology located at the center of the virion ( Fig 1C and 1D ) is intriguing , as thermodynamic considerations would have predicted a genome conformation dispersed throughout the viral core . While the source of this unique structure remains unclear , we propose that the multiple and abundant DNA-binding proteins previously shown to be present in the PBCV-1 core ( 19 ) promote a condensed DNA morphology . TEM immuno-labeling and super-resolution fluorescence studies reported here indicate that shortly after being released into the host cytoplasm , the PBCV-1 genome assumes a condensed morphology . Such condensation is implied by the dense and highly clustered DNA labeling that is depicted in Fig 3 and is characteristic of condensed , rather than of dispersed , morphologies , as indeed demonstrated by the heavy DNA labeling revealed by virions shown in this figure . This condensation , which is shown here for the first time to occur during genome internalization of a eukaryotic-infecting virus , presumably plays a crucial role in PBCV-1 infection . In addition to promoting genome internalization , a compact DNA morphology is likely to facilitate passage of the large PBCV-1 genome towards and into the host nucleus within the crowded host cytoplasm . Notably , DNA condensation was proposed to represent a significant pulling force during bacteriophage genome ejection [19 , 43 , 50 , 51] . Once released into the host cytoplasm , the PBCV-1 genome is rapidly translocated towards and into the host nucleus where it is replicated and subsequently released into a cytoplasmic factory where viral assembly occurs [14 , 52] . As shown in this study , chlorella cells are packed with chloroplasts containing thylakoid membranes that surround most of the cell periphery , underlining the question how do PBCV-1 genomes overcome this major hurdle during their trajectory towards the host nucleus . Our studies reveal that shortly after PBCV-1 infection , host thylakoid membranes are perforated , thus paving a pathway for the virus genome towards the host nucleus . Indeed , viral DNA sequences are present in the host thylakoid membranes ( Fig 6 ) . Significantly , a proteome study revealed that PBCV-1 packages two viral-encoded putative phospholipases [53] . It is tempting to speculate that PBCV-1 uses these phospholipases to perforate the thylakoid membranes in order to generate a trajectory towards the host nucleus . This notion is consistent with reports indicating that almost immediately after PBCV-1 infection , substantial reduction in photosynthesis occurs [54 , 55] . Notably , after ejecting their genome , empty PBCV-1 virions remain attached to the host wall , as is the case for bacteriophages . Taken together , the results reported here reveal that the initial infection process of the chlorovirus PBCV-1 genome is remarkably similar to the process used by many tailed-bacteriophages yet differs from the process used by eukaryotic-infecting viruses that initiate infection through internalization of the entire virion or a substantial part of the particle . The PBCV-1 infection cycle proceeds through perforation of the host cell wall , cell plasma membrane and thylakoid membranes , thus overcoming the obstacle imposed by these cellular components on the translocation of viral DNA towards the nucleus . Previous studies established that at late infection stages , PBCV-1 generates cytoplasmic organelles , termed viral factories , where viral assembly takes place , a feature characteristic of many eukaryotic-infecting large dsDNA and ( + ) RNA viruses [14 , 15 , 27 , 30 , 31 , 34–36 , 56–58] . Thus , PBCV-1 uniquely combines a bacteriophage-like mechanism during its early infection stages with a eukaryotic-like virus infection pathway in its late replication stages .
Chlorella variabilis NC64A cells were grown under continuous light and shaking on a modified Bold’s basal medium ( MBBM ) [54] . PBCV-1-infected as well as mock-infected cells were prepared for electron microscopy studies , including STEM tomography , FIB-SEM and immuno-electron microscopy [14] . Multiplicity of infection ( MOI ) was 10 in all experiments with the exception of the STORM studies in which it was 20 and immuno-flouresence studies in which it was 1 . To detect PBCV-1 in immuno-fluorescence assays we raised antibodies against the major capsid protein , Vp54 using a short peptide sequence , NDDRYNYRRMTDC , derived from the full-length protein . The synthetic peptide conjugated to KLH ( Keyhole Limpet Hemocyanin ) , a carrier protein extensively used to enhance antibody production on Cys residues was used to immunize 5 mice . Chlorella cells were infected with PBCV-1 at MOI of 1 for 3h and fixed with 4% paraformaldehyde ( EMS ) for 15 min at RT . Cells were washed in PBS and transferred to Poly-Lysine coated glass microwell dishes ( Mat-Tek corp . ) . Cells were then blocked with 4% BSA-PBS solution for 30 min at RT and exposed to anti-capsid antibody diluted in 4% BSA-PBS for 1h at RT . Following washes in PBS , a secondary antibody , goat anti-mouse IgG conjugated to Alexa488 ( Life Technologies ) diluted in 4% BSA-PBS solution was added for 30–45 min . Cells were then counterstained with 500 nM SYTOX Orange ( Life Technologies ) in DDW for 15 min . Fluorescence images were visualized and photographed using a Deltavision system ( Applied Precision ) equipped with X100 UPlanSApo NA 1 . 40 objective . Fluorochromes were excited at 490 nm for Alexa488 , 555 nm for SYTOX Orange and 640nm for chlorophyll ( auto-fluorescence ) . Images were acquired with a photometrics coolSNAP HQ2 CCD ( Roper Scientific ) and de-convoluted with SoftWorx package using high noise filtering and 10 iterations . Image analysis and processing were conducted with Image J and Adobe Photoshop CS4-extended softwares . Chlorella cells were infected with PBCV-1 for 1–2 min at MOI of ~20 and fixed with 4% paraformaldehyde for 15 min at RT . Cells were washed with PBS and transferred to Poly-Lysine coated glass dishes ( Mat-Tek corp . ) , blocked with 4% BSA-PBS solution for 30 min at RT and exposed to anti-capsid antibody for 1h . Following washes in PBS , goat anti-mouse IgG conjugated to Alexa488 ( Life technologies ) diluted in 4% BSA-PBS solution was added for 30 min for labeling viral capsids . Cells were washed in PBS and counter-stained with 5 nM SYTOX Orange ( Life technologies ) in DDW . Images were collected on a Vutara SR200 STORM microscope ( Bruker ) . Before performing super resolution imaging , virus locations were identified by conventional fluorescence using Alexa488 labeling and 488 nm laser excitation ( ~5kW/cm2 ) . DNA structures labeled with SYTOX Orange were imaged in super resolution using 561 nm laser excitation power of ~15kW/cm2 . Images were recorded using a X60 , NA 1 . 2 water immersion objective ( Olympus ) and Evolve 512 EMCCD camera ( Photometrics ) . Data were analyzed with Vutara SRX software . | Although extensively studied , the mechanisms responsible for internalization of viral genomes into their host cells remain unclear . A particularly interesting case of genome release and internalization is provided by the large Paramecium bursaria chlorella virus-1 ( PBCV-1 ) , which infects unicellular eukaryotic photosynthetic chlorella cells . In order to release its long dsDNA genome and to enable its translocation to the host nucleus , PBCV-1 must overcome multiple hurdles , including a thick host cell wall and multilayered chloroplast membranes that surround the host cytoplasm . Our observations indicate that these obstacles are dealt with perforations of the host wall , the host cellular membrane , and the host photosynthetic membranes by viral-encoded proteins . Furthermore , our results highlight a bacteriophage-like nature of early PBCV-1 infection stages , thus implying that this virus uniquely combines bacteriophage-like and eukaryotic-like pathways to accomplish its replication cycle . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"viral",
"transmission",
"and",
"infection",
"plant",
"cell",
"biology",
"microbiology",
"chloroplasts",
"plant",
"science",
"membrane",
"fusion",
"viral",
"genome",
"cellular",
"structures",
"and",
"organelles",
"microbial",
"genomics",
"viral",
"genomics",
"viral",
"packaging",
"viral",
"replication",
"cell",
"membranes",
"cytoplasm",
"plant",
"cells",
"host",
"cells",
"cell",
"biology",
"virology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"genomics"
]
| 2017 | Structural studies demonstrating a bacteriophage-like replication cycle of the eukaryote-infecting Paramecium bursaria chlorella virus-1 |
Schistosome cercariae only elicit high levels of protective immunity against a challenge infection if they are optimally attenuated by exposure to ionising radiation that truncates their migration in the lungs . However , the underlying molecular mechanisms responsible for the altered phenotype of the irradiated parasite that primes for protection have yet to be identified . We have used a custom microarray comprising probes derived from lung-stage parasites to compare patterns of gene expression in schistosomula derived from normal and irradiated cercariae . These were transformed in vitro and cultured for four , seven , and ten days to correspond in development to the priming parasites , before RNA extraction . At these late times after the radiation insult , transcript suppression was the principal feature of the irradiated larvae . Individual gene analysis revealed that only seven were significantly down-regulated in the irradiated versus normal larvae at the three time-points; notably , four of the protein products are present in the tegument or associated with its membranes , perhaps indicating a perturbed function . Grouping of transcripts using Gene Ontology ( GO ) and subsequent Gene Set Enrichment Analysis ( GSEA ) proved more informative in teasing out subtle differences . Deficiencies in signalling pathways involving G-protein–coupled receptors suggest the parasite is less able to sense its environment . Reduction of cytoskeleton transcripts could indicate compromised structure which , coupled with a paucity of neuroreceptor transcripts , may mean the parasite is also unable to respond correctly to external stimuli . The transcriptional differences observed are concordant with the known extended transit of attenuated parasites through skin-draining lymph nodes and the lungs: prolonged priming of the immune system by the parasite , rather than over-expression of novel antigens , could thus explain the efficacy of the irradiated vaccine .
The radiation-attenuated schistosome ( RA ) vaccine remains the most effective way of inducing high levels of protective immunity against Schistosoma mansoni in rodent and primate hosts ( reviewed by Coulson ) [1] . However , an effective recombinant vaccine based upon it , for use in humans , has thus far proved elusive [2] . Few differences have been reported between irradiated and normal larvae apart from an altered morphological phenotype at the lung stage of development [3] that produced subtle differences in motility . This accorded with a key feature of the vaccine that attenuated larvae must undergo a truncated migration , as far as the lungs , to prime the immune system [4] . Furthermore , extensive parasite tracking [5] and immunological investigations [6] have revealed the lung schistosomulum to be the principal target of immune effector responses in the murine host . The requirement for CD4+ T cells [7] means that antigens must be released by , or exposed on , the surface of target larvae for processing and presentation by accessory cells to trigger such effector responses . Targets of protective immunity have historically been identified by screening crude antigen preparations and expression libraries with sera from putatively immune hosts [8] , [9] . In the schistosome context such screens have , in the main , produced a catalogue of abundant cytoplasmic proteins that one would not ordinarily expect to be secreted or surface-exposed and thus available to the immune system . Indeed , the abundance and antigenicity of cytoplasmic proteins appears to pose a major obstacle to identifying truly protective antigens . Abundant transcripts can dominate the content of cDNA libraries; equally , highly expressed proteins may mask attempts to identify bona fide vaccine candidates using proteomics [10] or immunoproteomics [11] . Clearly alternative approaches are needed to pinpoint antigens relevant to protection in this model and the sequencing of the schistosome transcriptome [12] and genome ( www . GeneDB . org ) now provide unparalleled opportunities for rapid progress using post-genomic techniques [13] . We have previously constructed a microarray comprising cDNAs derived from normal lung stage schistosomula and used it to identify genes highly expressed in the migrating parasite relative to six other life cycle stages [14] . We found genes encoding six membrane , six membrane-associated and five secreted proteins that were preferentially expressed at the lung or skin and lung stage . However , when considered in isolation it is difficult to predict which of these proteins , if any , will make suitable vaccine candidates . Their site of expression in the complex parasite body is unknown and some are hypothetical proteins with no ascribed function except at the motif or domain level . We now report use of the same lung stage array to pinpoint transcripts differentially expressed between normal and irradiated parasites cultured to the lung stage . This experiment was designed to identify the molecular changes underlying the altered phenotype , primarily using Gene Set Enrichment Analysis ( GSEA ) to delineate groups of genes with associated functions , which could explain the enhanced immunogenicity of the irradiated larvae .
A Puerto Rican isolate of S . mansoni was maintained by passage through NMRI strain mice and Biomphalaria glabrata snails , the animal work being approved by the Biology Department Ethics Committee , University of York . The microarray [14] was screened with mRNA from schistosomula , derived from mechanically transformed cercariae and grown in vitro for four , seven or ten days [15] . The times were chosen on the basis of previous parasite tracking [4] and lymphadenectomy experiments [16] . Attenuated schistosomula begin to accumulate in the lymph node and lung at day four , reaching a plateau in both sites at day seven [4] . Excision of skin-draining lymph nodes at , or prior to , day ten has a major ablative effect on subsequent protection [16] The cercariae were obtained by exposing snails with a patent infection to a bright light . Prior to culture one half of each cercarial shed was exposed to 200 Gray of radiation from an X-ray source at Cookridge Hospital , Leeds . The microarray ( ArrayExpress A-SGRP-2/E-TABM-408 ) containing approximately 6000 features printed in duplicate ( accession numbers AM042715-AM048613 ) , the hybridisation protocol , and array scanning were as described in Dillon et al . ( 2006 ) . The array represents 3088 unique sequence contigs and singlets , encompassing an estimated 44% of the lung worm transcriptome [12] . At each of the day four , seven and ten sampling times total RNA was extracted from parallel cultures of normal and irradiated schistosomula with Trizol ( Invitrogen ) according to manufacturer's instructions . Each total RNA was labelled with Cy3 or Cy5 dyes ( Perkin Elmer ) , without amplification , before hybridisation to the array at 20μg per channel [14] . Analysis of the normal and irradiated treatments , in pairs , at three time points encompassed twelve slides , comprising three biological replicates per treatment and one technical replicate ( i . e . one of the biological replicates was split and repeated in order to control for experimental error ) . Dye swaps were balanced across treatments to limit bias resulting from differential dye incorporation and intensity , i . e . 50% of irradiated samples were labelled with Cy3 and 50% with Cy5 . One sample from day ten failed to label so only 3 slides in total contributed to that time point . The quantative dataset obtained using the GenePix 4000B instrument ( Axon Instruments Inc . ) , was analysed with the GenePix Pro software and the R language for statistical computing ( www . r-project . org ) [17] . Specifically , the data was processed with the microarray analysis tools available from the Bioconductor Project , a tool for the analysis and comprehension of genomic data ( www . bioconductor . org ) [18] . The background was subtracted from array data using a Bayesian model-based method [19] . Array data were normalized using the LIMMA component ( Linear Models for Microarray Data ) of the Bioconductor package [20] with printtip loess to correct for spatial and other artefacts generated during the printing process . ( Loess is a locally weighted polynomial regression; see LIMMA documentation . ) Linear models were applied and significance statistics generated using empirical Bayesian methods to assess differential gene expression . This has the effect of borrowing information from the ensemble of genes to aid with inference about each individual gene [20] . An observation was classed as significant if it exceeded a natural log-odds ( lods ) cutoff of 3 . To determine the effects of radiation , irrespective of sampling time , normal and irradiated results were pooled and reanalysed as a two way comparison . Detailed description of the methods used can be found in the LIMMA documentation: http://bioconductor . org/packages/2 . 1/bioc/vignettes/limma/inst/doc/usersguide . pdf . As each EST is duplicated on the array , mean red and green values for the 6528 probes were generated from background-subtracted red and green fluorescence values . The LIMMA function “normalizeQuantiles” *was applied to these mean fluorescence values to normalize between arrays . Thus each quantile of each EST is adjusted to its mean across all arrays , irrespective of channel , normalising the data by ensuring the signal intensities within each treatment have the same empirical distribution . In those instances in which two or more ESTs on the array were members of the same Sm contig , the mean normalised values were taken , resulting in a single value for each contig . The normalised signal intensities were combined into tables containing all Sm contigs and singlets , with their GO/Protein analyst annotation ( as described in Dillon et al . 2006 ) , test channel signals and reference channel signals , prior to submission to the GSEA package . GSEA statistically assesses whether expression of groups of genes correlates with a given phenotype , and requires those groups to contain 15 or more members to function [21] . It quantifies the enrichment of individual members at the top and bottom of a ranked list of gene expression . The enrichment score ( ES ) is calculated by parsing the ranked gene list for members of a single category , and increasing a running-sum statistic when one of those genes is encountered or decreasing that sum if it is not . The enrichment score is then normalized by adjusting for the number of genes in a category and the GSEA package estimates the significance of each normalized enrichment score ( NES ) by calculating a false discovery rate ( FDR ) . Gene sets were deemed to be enriched when the FDR ≤ 0 . 25 , this apparently relaxed cut-off being used because the primary goal of GSEA , as specified by Subramanian et al . ( 2005 ) , is to generate hypotheses rather than exclude every last false positive . The FDR is calculated by comparing the tails of the observed and null distributions for the NES . The null is produced by randomly assigning phenotype labels and producing a reordered gene list; this is done 1000 times to generate a null ES for each set . The LES is defined as the core grouping of genes contributing to the enrichment score; this generally represents approximately 30–50% of genes in an enriched category [21] . Three ESTs deemed to be differentially expressed , using the LIMMA package of Bioconductor , plus one on the threshold of significance were chosen for validation of array predictions by real time PCR analyses . The ESTs and primers used are outlined in Table S1 . The Primer Express package ( Applied Biosystems ) was used to design primers to the four ESTs and the 18S ribosomal RNA control . A dissociation plot was performed for each primer to determine specificity . Comparable amplification was confirmed and assays performed in triplicate , on an ABI 7300 PRISM instrument using SYBR green dye , according to the manufacturer's instructions . All data were normalized to the lowest level of expression as determined by real time PCR .
The LIMMA package analysis of changes in single genes across the three time points highlights only seven significant differences between irradiated and normal parasites ( Table 1 ) . In all cases genes reaching our stringent statistical cut-off are conspicuously down-regulated in the irradiated parasite . Two differentially regulated transcripts encode proteins destined for the plasma membrane . One of these is the previously characterised Sm25 ( also known as Gp18–22 ) and the other codes for a hypothetical protein . A third transcript encoding Tetraspanin D ( Sm-TSP-2 ) , known to be present at the tegumental surface [22] , is down-regulated at all three time points . Of the remaining genes revealed , JF-2 codes for a membrane-associated cytoskeletal component thought to link actin filaments to the plasma membrane , cdc2 is a key control enzyme of the cell cycle and two code for hypothetical proteins ( Table 1 ) . The level of expression of the four selected ESTs was determined using real time RT-PCR and compared with that estimated from the array hybridisations ( Figure S1 ) . Plotting the data as a histogram highlights the broad level of agreement between the two techniques , and shows that in contrast to previous work [14] , differences in sensitivity are not as pronounced . This is likely due to the smaller variations in expression measured . A scatter plot of the same data ( data not shown ) demonstrates that the two methods exhibited high concordance , with a correlation coefficient R = 0 . 80 . The expression of genes , grouped by biological function or subcellular location , correlating with a specific phenotype , was assessed using the GSEA package . A heat map recording the differential expression of the 1769 unique features on the array is presented in Figure 1A . A symmetrical distribution of expression profiles , can be discerned whereby two thirds of the genes are visibly associated with a phenotype . Approximately one third are up-regulated in the irradiated parasite ( the red-dominated top left corner ) and a different third in the normal parasite ( the red-dominated bottom right corner ) ; the remaining central third display no obvious pattern . A list of genes ranked by the intensity of their expression was derived from the heat map ( Figure 1B ) and used to produce a graphical plot of the running sum statistic ( Figure 1C , E and G ) . This running sum statistic increases every time a member of a given gene set ( i . e . a GO category ) is encountered in the ranked gene list and decreases when it is not encountered . Where no correlation occurs between a gene set and the N or I phenotype the genes appear randomly in the ranked list producing a plot of the running sum statistic that fluctuates either side of zero ( e . g . Figure 1C ) . The running sum for expression of the gene sets that correlate with the irradiated phenotype ( e . g . Figure 1E ) is skewed to the left by the abundance of numerous members in that region of the ranked gene list ( Figure 1F ) . Conversely , correlation with the normal phenotype is skewed to the right ( e . g . Figure 1G and H ) . The leading edge subset ( LES ) represents the core of genes most strongly associated with the N or I phenotype ( Figure 1E and G ) . The most prominent enriched gene sets at day four include ‘protein modification’ in the irradiated parasite and ‘RNA-directed DNA polymerase activity’ ( root GO term is ‘Molecular Function’ , which is identical to the ‘Biological process’ category ‘RNA-dependent DNA replication‘ ) in the normal parasites ( Table 2 ) . The former category is noteworthy for containing the E1-3 ubiquitinating enzymes in its LES ( Table S2 ) while the latter appears to consist primarily of retrotransposon transcripts ( at least 17 members , data not shown ) . The ‘GTP binding’ LES associated with the irradiated phenotype contains no fewer than 17 ras/rab/rac small G-protein homologues , together with a stimulatory and an inhibitory heterotrimeric G-protein alpha subunit ( Table S2 ) . A fifth ‘calcium ion binding’ category , also associated with the irradiated phenotype , possesses numerous transcripts encoding proteins of disparate motor or structural function including EF hand-containing proteins such as Sm22 . 6 , myosin , at least two annexins and severin ( Table S2 ) . At day seven , only the irradiated phenotype shows gene set enrichment ( Table 2 ) . Gene categories associated with ‘metabolism’ , ‘mitochondrion’ and ‘electron transport’ are overrepresented and an analysis of LES overlap reveals commonalities between the three subsets of enriched genes . The genes shared are specifically involved in the respiratory electron transport chain . A number of cytochrome subunits , NADH metabolising and antioxidant thioredoxin enzymes all contribute to the enrichment score of the three gene sets ( Table S3 ) . Protein synthesis and degradation also appear to be prominent processes in the day seven irradiated parasite . The categories ‘protein biosynthesis’ , ‘protein folding’ , ‘proteolysis’ , ‘ribosome’ , ‘structural constituent of ribosome’ , ‘cysteine-type peptidase activity’ and ‘isomerase activity’ encompassing genes encoding translation initiation factors , isomerases and chaperones all correlate with the irradiated parasite phenotype ( Table 2 ) . Intriguingly , transcripts encoding extracellular proteins also appear to be enriched , although the heterogeneous nature of this LES makes it difficult to discern a biological pattern ( Table S3 ) . Nevertheless , the presence of the antigen 5 transcript , protease inhibitors and a lipoprotein receptor is noteworthy . The transcriptional divergence of the irradiated and normal parasites is even more apparent by day ten; 32 of the 89 gene sets submitted to GSEA show enrichment correlating with one or other phenotype . For the irradiated parasite the enrichment of ‘ribosome’ components persists into day ten and ‘RNA-directed DNA synthesis’ is again comparatively depressed with respect to the normal parasite . Indeed , at this stage the irradiated parasite differs from the normal parasite in many biological systems ( Table 2 ) . Categories for ‘transcription regulation’ , ‘RNA binding’ and ‘helicase activity’ are under-represented in the irradiated parasite and there is also a relative shortfall in ‘intracellular kinase signalling’ and ‘structural proteins’ , specifically cytoskeletal transcripts . The comparative paucity of receptor-encoding transcripts is particularly striking in the irradiated parasite , as is the general dearth of transcripts from the gene sets ‘endoplasmic reticulum’ , through the ‘golgi’ to the ‘plasma membrane’ . An analysis of the LES of the ‘receptor activity’ category reveals an overlap with other gene sets diminished in their own right , including ‘ion channel activity’ and ‘G-protein coupled receptor signalling’ . The functional overlap reveals that a significant proportion of these transcripts are neuroreceptors or channels , including acetylcholine , purinergic , nicotinic , glutamate and aspartate receptors plus voltage and ligand-gated ion channels ( Table S4 ) . Examining the pooled data for differential enrichment of categories , in normal versus irradiated , emphasises the apparent importance of protein synthesis and degradation in the irradiated parasite . Categories associated with protein metabolism , including ‘protein folding’ are prominent as is the ‘cysteine-type peptidase activity’ GO set , containing a number of cathepsins , and a ‘cytosol’ set that contains proteosome activators and some 20S proteosome components in its LES ( Table S5 ) . While Golgi-related transcripts do not meet the FDR cut-off , the overlap between the ‘golgi’ , ‘GTP binding’ and ‘small GTPase mediated signal transduction’ LES together with deficiencies in the ‘ER’ category is noteworthy ( Table S5 ) . Although the comparative paucity of ‘receptor activity’ transcripts in the irradiated parasite is not obvious at days four and seven , the receptor activity is depressed at all time points when irradiated versus normal parasites were compared ( Table 2 ) . Analysis of the category ‘ion channel activity’ comprising transcripts encoding receptors associated with ion flux across membranes is also diminished in the irradiated parasite . Signalling cascades , particularly ‘kinase activity’ ( Table 2 ) are also less abundant in the irradiated parasite . The kinases may well interact with the ‘cell adhesion’ and ‘cytoskeleton’ categories contributing to the observed differences ( Table 2 ) but overlap analysis does not indicate shared genes in their respective LES .
Lung stage schistosomula of S . mansoni are a validated target of protective immunity induced in the murine host by exposure to RA cercariae . However , attempts to identify the antigens responsible , a key step in the development of a recombinant vaccine , have met with limited success [1] , [8] . Microarrays offer a route to antigen identification by pinpointing subtle differences in gene expression between irradiated and normal worms , irrespective of transcript abundance . Characterising the underlying transcriptional differences should highlight changes at the parasite-host interface that explain why irradiated larvae can elicit protective immunity when normal larvae do not . In addition , by shifting the focus away from antibody-based technologies , microarrays may identify genes encoding non-immunogenic proteins that are nevertheless fundamental to parasite migration and development . Analysis of the normal and irradiated parasite transcriptomes at day four , seven and ten revealed only seven genes that showed significant differences in expression . All were down-regulated as a result of radiation . Proteome Analyst predicted two as plasma membrane proteins ( Sm25; hypothetical protein ) . A third , a tetraspanin ( Sm-TSP-2 , CD63-like , tetraspanin D ) , mispredicted as lysosomal , is known to be exported to the tegument surface plasmamembrane [22] , as is Sm25 [23] , [24] . Biotinylation studies on adult worms indicated that tetraspanin D may play a role in maintaining tegumental membrane structure and organisation and could be accessible to the immune system [22] . Indeed , this particular tetraspanin , identified using a signal sequence trap [25] , elicited protective immunity when the major extracellular loop was used to vaccinate mice [26] . Conversely Sm25 , or its decorating glycans , may actually protect the parasite by subverting the host immune response as , despite eliciting high antibody titres , the recombinant protein does not protect vaccinated animals [27] . On the basis of membrane association and immunofluorescence studies it has been suggested that the actin binding protein JF-2 may be available at the tegument surface [28] . However , a sizeable proportion of patients infected with S . japonicum possess antibodies to JF-2 [28] , yet continual chemotherapy is still required to limit the impact of reinfection [29] . This observation argues that JF-2 normally confers little or no resistance and may simply be another cytoplasmic protein albeit one associated with plasma membranes [30] . Cdc2 , the final protein with an ascribed function , is a crucial cell cycle control enzyme . While the down-regulation of a single gene should not be over interpreted , suppressed levels of the cdc2 protein may reflect the inability to re-enter the cell cycle [31] . Migrating schistosomes are in a semi-quiescent metabolic state ( Lawson and Wilson , 1980 ) with no cell division taking place [32] . However , they are primed to enter cell cycle upon reaching the portal vein and beginning to blood feed [33] . Thus , down-regulation of cdc2 may be part of the explanation why irradiated parasites never mature . It is clear from numerous studies on the RA vaccine ( reviewed by Coulson 1997 ) that attenuated parasites must persist in the host for 1–2 weeks to elicit effective protection . Furthermore they must also migrate beyond the skin to its draining lymph nodes , and to the lungs . As anticipated , large transcriptional changes were not evident four days or more after the radiation insult , since the acute stress response has long subsided by that time [34] . Therefore , the ability to detect small coordinated changes , using the GSEA package developed by Subramanian et al . was particularly important as a means of dissecting out the longer-term effects of radiation exposure . Schistosomula undergo marked phenotypic changes while resident in the skin soon after penetration , which include remodelling of the tegument surface and ablation of penetration glands [35] . Subsequently , mid-body spines are lost and the larval body elongates to facilitate intravascular migration beyond the lungs [36] . At day four , approximating to the skin stage , it was difficult to detect meaningful differences in transcript abundance , suggesting that the delayed effects of radiation were very subtle . However , RNA-directed DNA synthesis , indicated by retrotransposon transcription , was more prominent in the normal parasite . Why this should be depressed in the irradiated parasite is unclear but could reflect long term suppression by DNA repair mechanisms [37] . The ‘protein modification’ category provided an early indication of enhanced protein metabolism in the irradiated parasite . By day seven the protein metabolism categories specified by GSEA revealed a more pronounced effect but this distinction was diminished by day ten . Despite the lack of obvious morphological differences between early normal and irradiated parasites ( Mastin et al . , 1983 ) our data are consistent with the observations of Wales et al ( 1992 ) that protein synthesis is temporarily inhibited by irradiation . It seems likely that body remodelling has been delayed so the enhanced protein metabolism may reflect a catch-up process relative to the normal parasite . In a similar vein the switch to anaerobic respiration [38] may be retarded as evidenced by the enrichment of energy metabolism categories at day seven , in the irradiated parasite . By day ten the divergence between normal and irradiated parasites was greatest , with the majority of highlighted gene sets down-regulated in the latter . We consider that these represent biological processes damaged beyond recovery by the now-distant radiation event . The decreased prominence of categories involving intracellular signalling ( e . g . ‘G-protein coupled receptor signalling pathway’ ) may indicate a reduced ability to respond to external developmental cues; the down-regulation of cdc2 , already noted , should be viewed in this context . In addition , deficiencies in structural categories such as ‘cytoskeleton’ may further impede the irradiated parasite's capacity for locomotion . This apparent inability to detect and respond appropriately to the surroundings is further reinforced by the comparative paucity of receptor transcripts , especially those encoding components of neurotransmitter pathways . All these categories identified by GSEA accord with the visible phenotype revealed by SEM studies [3] . Although the irradiated parasite is in most respects morphologically similar to the normal parasite , elongating and losing mid-body spines [39] , it nevertheless displays abnormal constrictions of circular muscle fibres in the body wall , resulting in uncoordinated movement [3] . It is this compromised locomotion that leads to the persistence of irradiated parasites in the host lymph nodes and lungs for five weeks or more [4] , [39] . In the lymph nodes the parasites drive lymphocyte proliferation [40] and in the lungs they act as a long-term stimulus to recruit lymphocytes that arm that organ against challenge parasites [41] . The reason that RA parasites in general elicit protective immunity when normal parasites do not has long been the subject of speculation and investigation [42] . Our study strongly indicates that the up-regulation of specific gene products to provide elevated immune stimulation is not the key . Indeed an expressed fragment of the tetraspanin gene that we detected as down-regulated by single gene analysis , was recently shown to have protective potential in the mouse [26] . This underlines our thesis that even if gene expression is reduced , the extended stay of attenuated parasites in the skin draining lymph nodes may still result in enhanced immune priming against exposed antigens . Equally once the host has been primed by the vaccine , antibody or cell mediated effector responses could act early upon the incoming parasite , after cercaria-schistosomulum transformation has been completed; from previous microarray experiments we already know that tetraspanin D is strongly expressed in the two day old schistosomulum [14] . We cannot rule out that parasites in vivo respond differently to some host factor , not present in vitro , by up-regulating specific genes as suggested by ex vivo experiments [43] . However , the subtle nature of differences between normal and irradiated parasites leads us to believe that changes in protein expression are poor indicators of potential antigenicity; it is likely that the accessibility rather than abundance of an antigen is the important factor . In this context , retarded development increasing the duration of immune stimulation appears to be the salient feature . In the long term , the radiation insult compromises the transcription of schistosome genes involved in neuromuscular activity and ultimately cell cycle progression . In this respect schistosomes are particularly suited to deliver a prolonged stimulus as they undertake a protracted migration from skin to portal system , lasting 7–21 days after penetration ( i . e . irradiation ) . Only when blood feeding and cell division begin in the liver [44] will DNA strand breaks prove lethal . This priming by larvae is quite distinct in both location and antigen load from the continuous priming over months to years provided by adult worms and their eggs . Furthermore , recent studies in the baboon model have shown that protective responses elicited by the irradiated vaccine are dissociated from both responses to chemotherapy and an ongoing chronic infection [45] . Exposure to irradiated metazoan and protozoan parasites has been widely used to study protective immunity , as the basis for vaccine development , but we believe this is the first attempt to interrogate the transcriptome of such a parasite . In addition to schistosomes , protective immunity is induced by radiation-attenuation of the nematodes Dictyocaulus and Ancylostoma spp . and the protozoa , Plasmodium , Eimeria and Theileria spp . [46]–[50] . Given the efficacy of radiation-attenuated parasites as vaccines , the findings of this study should provide pointers to the phenotypic changes that account for the success of these other parasites as inducers of protective responses . | Schistosoma mansoni is a blood-dwelling parasitic worm that causes schistosomiasis in humans throughout Africa and parts of South America . A vaccine would enhance attempts to control and eradicate the disease that currently relies on treatment with a single drug . Although a manufactured vaccine has yet to generate high levels of protection , this can be achieved with infective parasite larvae that have been disabled by exposure to radiation . How these weakened parasites are able to induce protective immunity when normal parasites do not , is the question addressed by our experiments . We have used a technique of gene expression profiling to compare the patterns in normal and disabled parasites , over the period when they would trigger an immune response in the host . We found that only a handful of genes were differentially expressed , all of them diminished in the disabled parasite . However , a more sensitive technique to examine groups of genes revealed that those involved in nervous system and muscle function were depressed in the disabled parasites . We suggest that reduced mobility of these larvae permits them longer contact with the immune system , thus enabling a strong protective immune response to develop . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"genetics",
"and",
"genomics/gene",
"expression",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"cell",
"biology",
"infectious",
"diseases/helminth",
"infections",
"public",
"health",
"and",
"epidemiology/preventive",
"medicine",
"microbiology/parasitology"
]
| 2008 | Altered Patterns of Gene Expression Underlying the Enhanced Immunogenicity of Radiation-Attenuated Schistosomes |
The ventromedial nucleus of the hypothalamus ( VMN ) has an important role in diverse behaviours . The common involvement in these of sex steroids , nutritionally-related signals , and emotional inputs from other brain areas , suggests that , at any given time , its output is in one of a discrete number of possible states corresponding to discrete motivational drives . Here we explored how networks of VMN neurons might generate such a decision-making architecture . We began with minimalist assumptions about the intrinsic properties of VMN neurons inferred from electrophysiological recordings of these neurons in rats in vivo , using an integrate-and-fire based model modified to simulate activity-dependent post-spike changes in neuronal excitability . We used a genetic algorithm based method to fit model parameters to the statistical features of spike patterning in each cell . The spike patterns in both recorded cells and model cells were assessed by analysis of interspike interval distributions and of the index of dispersion of firing rate over different binwidths . Simpler patterned cells could be closely matched by single neuron models incorporating a hyperpolarising afterpotential and either a slow afterhyperpolarisation or a depolarising afterpotential , but many others could not . We then constructed network models with the challenge of explaining the more complex patterns . We assumed that neurons of a given type ( with heterogeneity introduced by independently random patterns of external input ) were mutually interconnected at random by excitatory synaptic connections ( with a variable delay and a random chance of failure ) . Simple network models of one or two cell types were able to explain the more complex patterns . We then explored the information processing features of such networks that might be relevant for a decision-making network . We concluded that rhythm generation ( in the slow theta range ) and bistability arise as emergent properties of networks of heterogeneous VMN neurons .
The ventromedial nucleus of the hypothalamus ( VMN ) is a large hypothalamic nucleus with an important role in diverse behaviours stretching beyond its classic role in appetite regulation and energy homeostasis [1] . The nucleus regulates glucose and lipid homeostasis [2–4] , appetite and energy expenditure [5–8]; but also sexual behaviour [9–11] , social behaviours and aggression [12 , 13] , and defensive and escape behaviours [14–16] . The diversity of functions regulated by the VMN and the common involvement in these of sex steroids , nutritionally-related signals , and emotional inputs from other brain areas , has led to the suggestion that subpopulations of VMN neurons “constitute a nutritionally sensitive switch , modulating the competing motivations of feeding and avoidance of potentially dangerous environments” [17] . It has similarly been suggested that a ‘switch’ in the VMN might underlie the reciprocal gating of sexual and feeding behaviour [18] . This suggests that the VMN is a multi-stable network–that , at any given time , its output is in one of a discrete number of possible states corresponding to discrete motivational drives . How might this behaviour arise in the neuronal networks within the VMN ? The neurons of this nucleus are densely interconnected , and the great majority of them are glutamatergic: mRNA for the vesicle glutamate transporter VGLUT2 is densely expressed throughout the VMN , while the VMN is virtually devoid of GAD65 and GAD67 mRNA , indicating that it contains few intrinsic GABA neurons [19] . This suggests that VMN neurons are extensively interconnected by mutually excitatory pathways . For mutual excitation to support stable firing in a subpopulation of neurons , such positive feedback must be restrained by activity-dependent inhibition . In the case of VMN neurons , there is good reason to think that activity-dependent inhibition arises from intrinsic neuronal mechanisms–from slow , spike-dependent hyperpolarising currents . A conspicuous feature of many VMN neurons is that , in vivo , spikes are followed by a prolonged relative refractory period , as evidenced by spontaneous spike patterning [20] . Other neurons have a brief refractory period followed by a brief period of hyperexcitability . Such spike-dependent hyperexcitability might arise either by intrinsic mechanisms ( depolarising after-potentials ( DAPs ) that can arise by multiple mechanisms over different time scales ) , or by recurrent excitatory pathways . Detailed analysis of spiking activity in neurons of the VMN in vivo [20] previously detected a number of distinctive electrophysiological “phenotypes”–consistent patterns of spiking activity that could be classified into about nine behavioural types . These types vary in complexity: some cells show apparently random spike intervals , shaped only by an initial refractory period , while others show short rapid bursts of spikes , or underlying oscillatory activity . Here we used modelling to explore what combinations of distinct cell properties and network connectivity might explain the heterogeneity observed in the VMN . We modelled single neurons of the VMN using an integrate-and-fire based model with post-spike excitability modified by simplified afterpotentials . We have previously used this approach to model hypothalamic oxytocin and vasopressin neurons: these have no direct or indirect , synaptic interactions between them , so we can directly infer intrinsic properties from their spiking patterns . We found that a model neuron with two spike-dependent mechanisms mimicking a large , brief post-spike hyperpolarisation ( a hyperpolarising afterpotential , HAP ) and a small but prolonged post-spike afterhyperpolarisation ( AHP ) can very closely match the spike patterning of oxytocin cells in vivo [21 , 22] and that these simplified activity-dependent potentials are consistent with a biophysically detailed Hodgkin-Huxley type model of those neurons [23] . In modelling vasopressin cells , a close quantitative match to their more complex phasic burst spiking patterns could be achieved with the addition of a DAP and a spike-suppressed hyperpolarisation , acting together to produce emergent bistability , and resulting in intrinsic bursting activity [24] . In the VMN , however , spiking patterns are the product of intrinsic mechanisms combined with network interactions . To model VMN neurons therefore requires assembling a network , but the intrinsic properties of the neurons comprising that network cannot be inferred directly from their observed spiking patterns . Accordingly , we began with minimalist assumptions about their intrinsic properties as inferred from studies of these neurons in vivo , using an integrate-and-fire based model as a flexible template constrained to be consistent with experimental observations . We then constructed network models with the challenge of finding simple explanations of those patterns . Finally , with a network model framework that seemed able to explain most observed patterns , we asked what information processing features of such networks are likely to be relevant for a decision-making network . The recorded VMN spike data , model source code , and software , compiled for Windows PC , are available at https://github . com/HypoModel/VMNNet/releases .
Using two post-spike potentials the model has 11 parameters , but only five of these are required to fit the model to data from a given VMN cell ( one for input rate , two for the HAP , and two for the optional AHP or DAP ) making it amenable to an automated fitting procedure based on a genetic algorithm ( GA ) , and enabling a more objective and thorough exploration of the parameter space . The GA based technique involves generating a population of random parameter sets , using each set to run the model and generate a fit score , and then ‘evolving’ these over a number of generations to find a best fit . For each cell , we fitted the model ( default parameters in Table 1 ) using a population size of 128 parameter sets in each generation; this was run for 40 generations , varying the parameters within a physiologically plausible range ( Table 2 ) . We have previously demonstrated with the same model and fit scoring that this is sufficient to make a robust exploration of the parameter space [23] . To choose a final fit for each cell , we ran the algorithm 100 times and chose the final parameter values as the median values from the 10 best fits . The best fits showed little variation in parameter values , giving confidence in the final fit parameters . For each cell we repeated this process using just an HAP , an HAP and an AHP , and an HAP and a DAP . We also tested the combination of HAP , AHP , and DAP but did not find any cells where the fit could be substantially improved by using all three . Of the nine cell types recognised previously , five types could be well-fitted by the single neuron model ( Fig 2 and S1–S5 Figs ) . Four of the five cells originally classified as “random cells” [20] were fitted by a fast HAP ( mean λHAP = 9 . 1 ms ) and a short AHP ( mean λAHP = 92 ms ) ; the fifth required a DAP for a good fit , to match high IoD values . Four other cell types ( “slow DAP” , “longtail1” , “longtail2” , and “broad” ) could also all be well fitted with the model , and the fit scores and parameters for all 25 cells are given in S1 Table . Fig 2 shows examples of the fits achieved for one cell of each of these five types , and the fits for all cells are shown in S1 to S5 Figs . The “slow DAP” cells have a peak in their hazard function that was previously attributed to a DAP . However , this could equally arise from a very fast HAP ( mean λHAP = 4 . 7 ms ) : if λHAP is less than the PSP half-life ( 7 . 5 ms ) then the accumulated EPSPs that have triggered the spike can have a depolarising effect that outlasts the HAP [26] . Only one of the “slow DAP” cells needed a DAP for the best fit . The two classes of “longtail” cells could all be fitted with a slow HAP ( “longtail1” mean λHAP = 39 ms and “longtail2” mean λHAP = 57 ms ) . Most also required a small DAP for their best fit , with none needing an AHP . The “broad” cells are characterised by a hazard function which has a very slowly decaying refractory period . This was fitted by a fairly slow but small HAP ( mean λHAP = 22 ms; mean kHAP = 20 mV ) . The most important factors in distinguishing between these cell types were the HAP parameters ( Fig 2 ) , and the ovals in Fig 3 illustrate the range for each cell type . The previous type classifications remained robust on the basis of model fitting but show some overlap in HAP parameters . The duration of the apparent relative refractory period observed in hazard functions mainly reflects the combined effects of the HAP magnitude and half-life . We estimated this duration as the time taken for the membrane potential of a model cell to return to within 1 mV of the resting potential after a spike in the absence of any synaptic input . The “random” cells and “slow DAP” cells have relative refractory periods of 48 ms and 31 ms respectively ( estimated from median parameter values for the cluster ) ; the “longtail1” and “longtail2” cells have relative refractory periods of 198 and 293 ms; and the “broad” cells a relative refractory period of 96 ms . Four of the five “broad” cells needed an AHP as well as an HAP to achieve the best fit , while six of the ten “longtail” cells needed a small but slow DAP as well as an HAP . In neurons with an AHP , the IoD decreases with increasing binwidth , as previously observed in oxytocin neurons [21 , 23] . The slow DAP has the opposite effect , producing a higher IoD with longer binwidths ( Fig 4 ) . Essentially , a slow DAP makes cells more ‘bursty’ by its positive feedback effects , while a slow AHP makes them more stable by its negative-feedback effects . Thus requiring model cells to fit IoD range data ensures that they capture these features of recorded neurons . The sixth cell type , “regular cells” , have a hazard function that rises monotonically to very high values and have a very low IoD . This is consistent with a resting potential which lies above the spiking threshold , and a slowly decaying post-spike hyperpolarization that acts as a pacemaker current . However , as we see below , this is not the only way that such patterning can arise . We conclude that a substantial portion of the heterogeneity in the spike patterning of VMN neurons is due to heterogeneity in their intrinsic mechanisms that shape post-spike excitability . The dominant effect of these is activity-dependent inhibition , and under this there are two substantial subpopulations ( Fig 3 ) , distinguished by differences in the duration of post-spike hyperpolarisation . The “random” and “slow DAP” cells displayed a relatively brief hyperpolarisation whereas the “longtail1” , “longtail2” , and “broad” cells displayed a long hyperpolarisation , and hereafter we call these consolidated types ‘fast HAP’ and ‘slow HAP’ cells respectively . Fitting these cells with a single neuron model subject to random synaptic input shows that the spike patterning in these cell types can be explained purely by intrinsic properties that shape post-spike excitability , using known mechanisms such as the HAP and DAP . It does not assume that they are disconnected from other neurons but rather that there is no coordinated patterning or feedback in the inputs they are receiving . The more complex patterned cell types which could not be fitted with the single neuron model indicate either some unknown intrinsic mechanism or some non-random structure in their inputs . Our subsequent studies tested the idea that this might be the consequence of local network interactions . Three VMN cell types—those originally classified as “doublet” , “doublet-broad” , and “oscillatory” cells—have multi-modal ISI histograms and hazard functions that we could not match with the single neuron model . To test the idea that these might be network generated , and based on the evidence that most of the VMN neurons are glutamatergic [19] , we began by constructing a network of neurons connected by excitatory synapses . The simplest cell types , those that could be well fit by the fast HAP and slow HAP single cell models , defined the building blocks for the subsequent network models . For initial testing we constructed networks of 50 model neurons ( Fig 5 ) with identical intrinsic parameters , based on ‘fast HAP’ cells ( kHAP = 40 , λHAP = 10 ) or ‘slow HAP’ cells ( kHAP = 20 , λHAP = 40 ) , both with no AHP or DAP ( parameters in Table 3 ) . We used a simple model of synaptic transmission where a single spike generates a single EPSP , of fixed amplitude , subject to a transmission delay . Summation of PSPs within a single time step is linear , although non-linearity is introduced by their exponential decay . The network is randomly generated , with any two neurons having a chance of connecting defined by parameter esyn1 . The connections have a fixed strength defined by parameter synweight which is used to modify the magnitude of PSPs triggered by spikes generated within the network . Such networks tended to over-synchronise , shifting suddenly from slow spiking to fast synchronised bursts . We therefore increased the noise and variation in the synaptic connectivity by adding a random chance of failure to synaptic transmission ( fixed at probability 0 . 5 ) and a variable transmission delay . This produced a more gradual evolution of spike patterning as the synaptic connectivity was progressively increased . In the VMN there are likely to be varied transmission delays between neurons , depending on varied propagation delays , dendritic tree structures , and synapse locations , although how much they might vary is very hard to estimate . To test this assumption we went back to our experimental data to find paired recordings of coupled VMN cells that could be analysed to measure their coupling latency ( S6 Fig ) . The two pairs shown in this Figure show latencies consistent with transmission delays in the range of 5 to 15 ms . The ‘fast HAP’ network , subject to the same random input signal , shows a gradual shift to much faster more regular spiking as the connectivity increases ( Fig 6 ) . There are no new matches to other types of VMN neurons . However , in the ‘slow HAP’ network , as the connectivity increases , the firing rate increases , and the ISI distributions of individual cells become less skewed until they develop a second mode , matching the spiking observed in “doublet” neurons ( Fig 7 ) . With further increases in connectivity , these neurons display regular short bursts of spikes . The change in patterning appears to depend on the combination of intrinsic properties and network connectivity , with the slow HAP forming a negative feedback to counter the positive feedback of the excitatory connections . The progressive changes in the ISI distribution with increasing esyn1 closely correspond to the ISI distributions of several VMN cell types ( Fig 7 ) . Thus , a fixed slow HAP neuron model with a single parameter change increasing network connectivity is sufficient to reproduce much of the heterogeneity observed in VMN spiking patterns . We further tested the ability of the network model to match VMN cell patterning , attempting precise fits to five sample cells from the “doublet” and “doublet-broad” cell types . We attempted to adapt the automated fitting to the network model with a network version , but this was not sufficiently robust in the quality and consistency of the fits it produced . However , using manual parameter adjustment , informed by the matches observed in Fig 7 , we could achieve close fits with 50-neuron network models to both the “doublet” and “doublet-broad” type neurons ( S7 and S8 Figs; parameters in S2 Table ) . The fits require the right balance between the random external input , determined by Ire and Iratio , and the network generated input , determined by esyn1 , synweight1 , and Δrange . Producing the first , mode of the ISI histogram , which corresponds to the short ISIs of the doublets ( and multiple spike short bursts ) , requires sufficient network input ( it cannot be produced by the random external input alone ) . The width of this first mode is determined by the variability in network transmission ( Δrange , ) . The second mode corresponds to the longer ISIs , and is mainly determined by features of the external input . The overall skew of the ISI distribution–the length of the tail–decreases with Ire . The position of the second mode is strongly influenced by the HAP half-life ( λHAP ) . All of these parameters influence the spike rate , and have to be compensated against each other . Increasing Ire for example increases the height of the first mode which might be compensated by reducing esyn1 . Thus fits fall within a consistent range , but are not unique . To study this we attempted multiple fits in an example “doublet-broad” cell ( S2 Table ) , testing fits fixed by different Iratio and synweight1 . A larger Iratio can be compensated by a smaller Ire and shorter λHAP . , and a larger synweight1 can be compensated by a smaller esyn1 . Thus manual fitting , while not exhaustive or fully objective , gives a good understanding of how the model is working , and the essential balance between activity-dependent and -independent input signals , and intrinsic properties . The network fits to the “doublet” cells use a neuron model with parameters in the range of the “longtail1” neurons . The “doublet broad” type fits mostly use a neuron model with parameters in the range of the “broad” neurons . However , some of the previously “doublet” classified cells do not have a second mode , and these cells ( doublets19 and doublets22 in S7 Fig ) can be fitted by a single neuron model . The fits thus far establish the intrinsic properties and network connectivity required to match VMN neurons but do not demonstrate how the heterogeneous cell types might co-exist . To test the ability for matches to multiple VMN cell types to co-exist in a single model network we generated a network of 200 randomly varied model neurons , with three parameters randomly varied on a normal distribution . The neurons are based on our ‘slow HAP’ model , but include random variation in λHAP ( mean = 40 , SD = 20 ) sufficient to produce some cells which fall in the ‘fast HAP’ range . We also applied random variation to the input rate ( Ire , mean 150 , SD 45 ) and network connectivity ( esyn1 , mean 0 . 12 , SD 0 . 15 ) parameters . Neurons with an esyn1 of 0 or less do not receive any connections from the network , but can still send connections to other neurons . The network was run for 2000 s and the resulting varied ISI distributions for each neuron are presented in S9 Fig . For comparison , in S10 Fig , we include ISI distributions for our library of recorded VMN cells , presented in the same scaling and format . The ISI distributions show matches to both the single mode , and multi-mode distributions observed in the VMN cells . The majority can be matched to “longtail” , “broad” , or “doublet-broad” type cells , but there are also examples of “doublet” , “random” , “slow DAP” , and “oscillatory” type cells . All of the model cell distributions are consistent with those observed in the VMN , including some silent cells . This a very over-simplified representation of the VMN networks where we would expect neurons to be much more structured and entrained than pure random heterogeneity , but it shows that the proposed variations in intrinsic neural properties and network connectivity are capable of explaining the range of spike patterning phenotypes observed in vivo . To begin investigating the function of such a network , we looked at the summed population spike activity of the 50 ‘slow HAP’ neurons ( Fig 7 ) , reasoning that the summed activity of a network cluster might form a signal to a downstream neuronal target population . When the network is highly connected and shows two distinct modes in the ISI histogram , spiking in individual neurons consists of short bursts . The period of these bursts is determined by the competing drives of the synaptic input rate and the duration of the HAP ( which accumulates across the very short ISIs of the burst ) . If the network is very highly synchronised then the summed activity consists of sharp distinct peaks , but with the combined noise of random variation in synaptic input , network connections , and transmission delays these peaks become more like an oscillatory waveform . Thus , as esyn1 increases , the summed activity shifts from flat random activity to strong oscillating peaks , indicating that the network can function as a signal generator , turning random synaptic input into a rhythmically oscillating signal . We explored this further by testing a 100-neuron network of the same ‘slow HAP’ neurons with fixed esyn1 = 0 . 35 ( equivalent to esyn1 = 0 . 7 in the 50-neuron network ) with increasing rates ( Ire ) of random synaptic input ( Fig 8 ) . Coherent oscillating output appears at ~2 . 3 Hz ( Ire = 130 ) . The oscillation frequency increases with the input rate , peaking at ~6 Hz ( Ire = 600 ) . To illustrate the collective network signal , we simulated the effect that the summed spiking activity might have on the membrane activity of a downstream neuron . This uses a reduced version of our network transmission model to model the input potential that would result from each spike generating an EPSP , essentially producing a smoothed version of the summed spike counts ( Fig 8C ) . We further tested the scalability of the network with up to 500 neurons and produced similar results with esyn1 scaled to match the increased population , i . e . a 500 neuron network used esyn1 = 0 . 07 ( S11 Fig ) . The low connection probability between two individual neurons is countered by the larger number of neurons providing a greater chance of indirect connections . The good fits to the “doublet” and “doublet-broad” neurons suggest that much of the more complex spike patterning observed in VMN neurons can be explained by excitatory networks of simple cells of the single ‘slow HAP’ type . However , a model network with a single simple cell type cannot explain the ISI distributions of “oscillatory” VMN cells ( Fig 1 ) . These have a multimodal ISI distribution , including modes at multiples of a period corresponding to a ‘fundamental frequency’ of 2 . 5–4 Hz , but also a prominent “early” mode at about 20 ms . A rhythmic signal sufficient to generate spiking on its own would produce only a single periodic mode , thus this periodic excitability suggested the idea that the oscillatory spiking activity is due to a subthreshold rhythmic signal overlaid by a random input signal . The multiple modes in the ISI histogram arise because , in any given cycle , whether or not a spike will be triggered is subject to this randomness . Experimental evidence for an underlying rhythmic signal is apparent in the average spike-triggered field potential of recorded oscillatory neurons [20] . The network of cells with a fixed slow HAP generates such a rhythm , but weakening the rhythmic signal to subthreshold by reducing network connectivity also breaks the rhythm generation , suggesting the need for heterogeneity of intrinsic neuron parameters . Initial attempts at matching the multi-modal ISI distribution using random parameter variation of slow HAP neurons produced multi-modal histograms in some slower firing cells , but these lacked the first short ISI mode . A second problem was variation between runs . If the random element in intrinsic properties was large enough to produce results different in interesting ways from that achieved with a network of identical neurons , then it also became less consistent . Only some runs produced cells which showed multi-modal histograms . Thus , a ‘slow HAP’ neuron network can produce a rhythmic signal at the expected frequency , but not the multi-modal ISI distribution . Heterogeneity is necessary , but needs to be more controlled . Neurons which are sufficiently connected and which have a HAP slow enough to generate the oscillatory signal are also not capable of producing the early ISI mode , indicating the need for additional cells with a shorter HAP . We therefore explored a two cell-type network of ‘slow HAP’ neurons generating a rhythmic input for ‘fast HAP’ neurons , with both cell types also receiving random input ( Fig 9 ) . We used an interconnected population of 100 neurons ( esyn1 = 0 . 35 ) with a slow HAP ( kHAP = 60 mV , λHAP = 50 ms; ) ( type 1 ) and fed their outputs to 100 fast HAP neurons ( kHAP = 60 mV , λHAP = 5 ms ) ( type 2 ) , parameters in Table 4 . The ‘fast HAP’ neurons receive connections from the ‘slow HAP’ neurons with probability esyn12 = 0 . 2 but are not connected to each other . We also tested interconnected ‘fast HAP’ neurons , and connections from the ‘fast HAP’ neurons to the ‘slow HAP’ neurons . As these made no substantial difference to the results , we retained the simpler network , but importantly the results are not dependent on such a specific structure . To reduce the sharpness of the rhythmic peaks , compared to Fig 8 , we increased the random component of the synaptic transmission delay ( synrange ) from 10 to 15 ms . The ‘slow HAP’ neurons received random input Ire = 200 , and the ‘fast HAP’ neurons received random input Ire2 = 80 . With this , the network output was an approximately sinusoidal rhythm in the low theta range . Fig 9 shows the ISI distributions for single neurons from the network of each cell type . The ISI distribution in the ‘slow HAP’ cells shows a single sharp mode at 300 ms , like the “regular” VMN neurons identified by Sabatier and Leng ( 2008 ) . The fast HAP distribution shows a close match to the “oscillatory” VMN neuron of Fig 1 , with a sharp early peak , followed by decaying modal peaks at 300-ms intervals . We applied the same spike waveform analysis as performed in [20] to the model’s recorded membrane potential , producing another close match to the experimental data ( Fig 9 ) . Thus it seems that the diverse firing patterns observed in the VMN in vivo can all be accounted for by two intrinsic cell types receiving random external inputs and with varying degrees of random excitatory synaptic interactions between them , possibly structured into multiple sub-networks . A prediction of the excitatory network model is that we should see cells in vivo which show a high degree of synchrony in their spiking activity . To look for evidence of this , we returned to the library of VMN neurons and inspected the original voltage recordings to find examples where , as well as the spikes from the cell analysed , smaller spikes from a second cell in the background that could be extracted by waveform analysis . From this , we found six examples of pairs of very tightly coupled cells , one of which is shown in Fig 10 . Many VMN neurons in vivo also show bistability , switching between prolonged periods of fast and slow spiking ( see Fig 9 in [20] , and in one case we recorded a pair of such cells for a prolonged period , and recognised synchronous changes in activity ( Fig 11A ) . These cells are also evidence for the VMN’s ability to act as a switching , decision-making network . Even under anaesthesia we would expect to see active decision-making mechanisms in neurons involved in regulating physiological processes; generally the homeostatic functions of hypothalamic neuronal circuits function normally under urethane anaesthesia . As well as spontaneous switching of activity , the switching can also be triggered by systemic injections of CCK [27] which mimic peripheral signals arising from the gut . CCK predominantly inhibits VMN neurons and can switch a bistable cell from stable high frequency firing to a prolonged low activity state which typically ends with an abrupt return to high-frequency firing ( Fig 12A ) . Changes to the input rate in the networks tested so far produce only gradual shifts in output spiking activity and brief input perturbations produce no sustained change in activity . Thus the strength and duration of the excitatory network connections is not sufficient for self-sustaining activity . However , adding a DAP to the neurons in a 100-neuron slow HAP network ( esyn1 = 0 . 25 ) ( parameters in Table 5 ) can make excitation self-sustaining . At Ire = 100 the network sits in a stable slow spiking state ( 0 . 85 spikes/s ) . At Ire = 103 or 104 the network switches to fast spiking state after a delay subject to the randomly timed PSPs . At Ire = 110 the network switches to a stable fast firing state ( ~6 spikes/s ) . For a given mean input rate , the network is typically stable in one or the other state , with no slowly decaying or accumulating element . The range of parameters which produce stability depend on a balance between input rate ( Ire ) , connection density ( esyn1 ) , and the DAP ( kDAP and λDAP ) . A lower input rate can be compensated by a higher connection density , and a lower connection density can be compensated by a larger DAP . However , a larger DAP or higher connection density also produces a higher plateau firing rate , and so matching this to the in vivo examples here constrained the parameters used . Fig 11B shows an example with a noisy Ire ( mean = 100 , tau = 120 s , amp = 0 . 05 ) where the network randomly switches between states , like the spontaneous in vivo bistable activity observed experimentally and shown in Fig 11A . We failed to reproduce bistable behaviour with just a DAP , or just the excitatory connections . Thus , the bistability requires both intrinsic and network generated positive-feedback mechanisms . We tested this further by using short negative and positive perturbations to a fixed Ire to simulate transient inhibitory and excitatory signals ( Fig 11 ) . To match the in vivo experiment ( Fig 12A ) which shows a spontaneous return to the high state following inhibition by a CCK injection we set Ire = 104 ( Fig 12B ) . To test switching by both excitatory and inhibitory pulses we set Ire = 100 ( Fig 12C ) . Here short ( 2-s ) perturbations are sufficient to switch the network between slow and fast stable states , demonstrating that the network can self-sustain both slow and fast spiking under the control of transient signals .
The aim of this study was to use spiking neural models to bridge between the electrophysiology of the VMN and its hypothesised function as a decision-making network . Previous studies had identified a set of neural subtypes based on patterning in recorded spike times [20] . Some of these can be fitted by a simple single neuron model , but the more complex cell types , showing patterning features such as short bursting and oscillatory activity , cannot . However , as we showed here , they can be well matched by a network of simple neuron models . The vast majority of synaptic connections within the VMN are glutamatergic ( see Introduction ) , and we therefore attempted to build network models using only excitatory connections . The single neuron model fits , making predictions about the intrinsic electrophysiological properties that underlie the spike patterning , divide the simpler cell types into two classes , fast HAP , and slow HAP . Using these as building blocks , we developed three types of network model which can explain the more complex spike patterning . We showed that a network of slow HAP cells with mutual excitatory connections can generate close matches to the patterning in the “doublet” and “doublet-broad” VMN cells and mimic the short bursting activity observed in some recorded VMN neurons . Looking at the progressive changes in spike patterning with increasing network connectivity also revealed close matches to “longtail1” and “longtail2” cells , suggesting that the variation observed between these cell types may be due to different degrees of connectivity rather than varied intrinsic properties . We have shown here that such networks can function as a signal generator , turning a random noise input into a rhythmic output in the high delta/low theta range . We showed that such a network of slow HAP cells projecting to fast HAP cells can closely match the distinctive multi-modal ISI distribution of “oscillatory” VMN neurons . Conventionally , rhythm generation in neural networks makes use of inhibitory and excitatory connections [28–31] , but in this network the only inhibitory influence is intrinsic , arising from the HAP . We showed that an approximately sinusoidal waveform can be generated by a network that is deliberately not over-synchronised: weakening the synchronisation of the network by adding noise to synaptic transmission results in more wave-like summed activity , producing an oscillating signal . These controlled examples using one or two types of homogeneous neuron models show how the more complex spike patterning observed in the VMN can arise , but leave the question of how the multiple patterning types might co-exist . The fitting of multiple recorded cells suggested that the intrinsic properties of the neurons in the VMN are highly heterogeneous . By generating a more heterogeneous network model , with variation in both intrinsic properties and network connectivity , we can produce a single network that shows matches to all of the observed VMN cell types . Producing the more simply patterned cell types requires that some cells receive less local input than others , making a prediction for the structure in the VMN . These neurons might serve as pacemakers for rhythm generation , or as a first layer that receives more of the external input signals . Less local inputs can occur either because of fewer actual connections , or because variations in input activity mean that afferent cells are silent . The library of recorded VMN neurons consists of only active cells; in fact very many VMN neurons are silent or very slow firing , such that we cannot discover their patterning . It has been proposed that slow rhythms are important for facilitating the transfer of information between brain regions [32] . For example , if a neuron ensemble A projects to a neuron ensemble B , a subthreshold theta input from ensemble C to both A and B will , by coherently enhancing presynaptic excitability in A and postsynaptic excitability in B , selectively enhance communication from A to B . In particular , theta oscillations have been reported to synchronize the basolateral amygdala with the hippocampus and medial prefrontal cortex during periods of conditioned and innate fear in mice [33] . The amygdala has a rich reciprocal interconnection with the VMN [34] and has a common involvement in the regulation of fear , appetite and sexual behaviour [35] . Theta rhythms are also notably present in regions close to the VMN–the posterior hypothalamus and the supramamillary nucleus [36] . As summarised in the Introduction , there is evidence that the VMN is involved in decision making . A network which makes decisions in response to transient signals must be able to sustain its new state beyond the duration of that signal , while remaining responsive to new signals that might cancel that decision . An example might be feeding in a wild environment where a sensed danger would require switching from feeding behaviour to fight or flight behaviour . This requires bistable network activity . Here , we found that a network of mutually connected neurons , each with a slow HAP , and a DAP , can match the bistable activity observed in some VMN neurons . Using an excitatory network to generate bistability uses similar principles to previous work such as [37] . However , here the presence of a DAP is critical for this behaviour: neither the DAP or excitatory connections are alone sufficient . The bistability is highly robust: it is possible to generate spontaneous switching through random variations in input , but only within a very small range of input activity . Reliably triggering a switch in state requires a signal that is sustained for ~ 2 s , and the network is thus sufficiently responsive without being vulnerable to noise . By tuning the noise input signal and using transient inhibitory and excitatory perturbations to simulate injected signals such as CCK and ghrelin , this network can match the observed spontaneous , and stimulated bistable activity of VMN neurons , in particular the switching between slow and fast spiking states that has been observed in vivo . Finally , we note that if two such clusters are interconnected by the sparse inhibitory neurons present in the VMN , this will generate reciprocal bistability . This suggests a natural mechanism by which the VMN can reciprocally regulate competing behavioural desires .
Single neurons are modelled using the integrate-and-fire based model previously described in [21] . For each neuron , an external input signal Iext is generated using twin random Poisson processes to generate EPSP and IPSP counts en and in at each time step ( dt = 1 ms in the results here ) , using mean rates Ire and Iri . We tested smaller time steps down to 0 . 1-ms to confirm that a 1-ms step was sufficiently accurate , while maintaining a practical runtime . The IPSP rate , Iri is defined as Iri = Iratio Ire and the external input rate is controlled using just Ire . The PSPs have fixed amplitudes eh = 3 mV and ih = -3 mV and are summed to give the input Iext: Iext=ehen+ihin In the single neuron model , Iext composes the entire input signal I such that I = Iext . This is summed to form the synaptic component of the membrane potential , Vsyn , decaying exponentially with half-life λsyn corresponding to time constant τsyn . dVsyndt=−Vsynτsyn+I Vsyn is initialised to 0 mV . Time constants are calculated from half-life parameters using: τx=λxln ( 2 ) where x is the variable concerned . The HAP variable decays exponentially with half-life parameter , λHAP , and is incremented by kHAP when a spike is fired: dHAPdt=−HAPτHAP+kHAPδ where δ = 1 if a spike is fired at time t , and δ = 0 otherwise . The AHP and DAP use the same form: dAHPdt=−AHPτAHP+kAHPδ dDAPdt=−DAPτDAP+kDAPδ At t = 0 , the three post-spike potential variables are initialised to their respective k parameter values , and remain cumulative , with no post-spike reset . The voltage components are summed with the resting potential , Vrest , to give the membrane potential V: V=Vrest+Vsyn−HAP−AHP+DAP When V exceeds the spike threshold , Vthresh , and the time since the previous spike exceeds the 2 ms absolute refractory period , a spike is fired , though its form is not modelled . To model a network we run multiple copies of the spiking model , calculating network input activity at each time step . Each spiking model independently generates its random external input signal . The network connections are static , and randomly generated before running the model , with each neuron connecting to each other neuron with probability esyn . In models with two neuron types , the connection probability is defined between each pair of types . The results here have type 1 neurons connected to each other with probability esyn1 and type 1 neurons connected to type 2 neurons with probability esyn12 . At each time step ( 1ms ) the network generated EPSPs are summed for each neuron to generate its network input signal , Inet . Each connection also has a varied transmission delay component , randomly generated in a range defined by Δrange . To model variable transmission delay , each neuron stores a 20-ms network input queue . When a neuron fires a spike , it sets a flag that it is active . At the beginning of each time step , each neuron checks for active flags on each of its connections . If a connection is active , the neuron generates a uniform random value rtrans in range 0 to 1 . If rtrans > P ( trans ) , where P ( trans ) is the transmission probability ( in all the results here set to 0 . 5 ) , then transmission is successful . The transmission delay Δtrans is calculated as the sum of a fixed base component Δmin and a uniform random component ranging from 0 to Δrange . Δtrans is either fixed for each connection at network generation or generated dynamically for each transmission event . Testing both methods showed no effect on the results . Δtrans is thus defined: Δtrans=Δmin+Δrangeurand where urand is a uniform random number between 0 and 1 . The input queue is then incremented at position Δtrans . After input processing , each neuron moves its queue forward one step and the first queue position then gives the current count of network EPSPs , nn . The network input potentials have fixed amplitude nh = 3 mV to give network input: Inet=nhnn Input I then becomes: I=Iext+Inet Most of the results here use a fixed rate external input , defined by parameter Ire . This is used to generate randomly timed EPSPs and IPSPs , producing an input signal with noise on a tens of millisecond time scale . A more noisy input signal can be generated by applying Gaussian noise to Ire using an Ornstein-Uhlenbeck process . Ire becomes a variable , defined: dIre=μnoise−Ireτnoisedt+knoisedtgrand where μnoise is the noise mean , τnoise is the noise time course or decay rate , and knoise is the noise amplitude . grand is a Gaussian ( or normal ) distributed random number with mean = 0 and standard deviation = 1 . The variable Ire is initialised to Ire = μnoise . The data from the model and from experimental recordings consist of series of spike times . These are used to calculate mean firing rates and to generate ISI distributions and hazard functions , calculated from the ISI distributions as described in [38] . The ISI distributions are calculated as histograms of all the ISIs calculated from the spike times , counted in 5-ms bins . To compare data with varied spike counts , the bin counts are normalised and scaled to total 10000 . The hazard function converts the absolute probabilities of the ISI distribution into conditional probabilities , so that each bin gives the chance of firing a spike in that time window ( or bin ) , ( hazard in bin [t , t+5] ) = ( number of intervals in bin [t , t+ 5] ) / ( number of intervals of length > t ) . The hazard thus shows how excitability change over time in the period following a spike . Index of dispersion ( IoD ) , calculated as the variance of a variable divided by its mean , is used here to measure the variation in binned spike rate across time , as previously described in [21] . Using spike times from a recorded cell or generated by the model , we count the number of spikes in successive bins . We then calculate the mean and variance of these bin counts to generate the IoD of spike rate across time . This is repeated for bin sizes of 0 . 5 s , 1 s , 2 s , 4 s , 6 s , 8 s , and 10 s to generate the IoD range . Purely random spike times , with no activity-dependent influence , will produce a flat IoD range . The previously published in vivo spike data for model fitting are from extracellular recordings of cells in the ventromedial nucleus ( VMN ) of urethane-anaesthetised rats , as detailed in [20] . The automated fitting used to fit the single neuron model to recorded cell data uses an evolutionary genetic algorithm ( GA ) based method , described in detail in [23] . A population of randomly generated ( within specified ranges ) parameter sets are run with the model and compared with the recorded cell data using a set of weighted fit measures based on the ISI distribution ( divided into head range and tail range ) , hazard function , and IoD range , to calculate a single value fit score . The best parameter sets ( the ‘parents’ ) are then interbred to create the next generation . The GA parameters ( Table 6 ) include the weights for the four fit measures , and the range parameters for the ISI distribution measures . Only the ISI range parameters were altered between fits , tuned to match the ISI distribution of individual cells . We ran the GA for 40 generations each with a population size of 128 . This was sufficient for the population to converge . The result is picked as the best scoring parameter set from the final generation . For each fit the GA was run 100 times , with the final fit calculated as the median parameter values of the best 10 . This was repeated for each fitted cell with HAP only , HAP + AHP , and HAP + DAP . The GA uses a GPU based implementation with the GPU code developed in Nvidia’s CUDA [39] . A single run of the GA takes 18s running on a GeForce GTX 960 GPU . The model is implemented in custom software developed in C++ , compiled in Microsoft Visual Studio 2010 . The graphical interface is developed in our own modelling software development toolkit , based on wxWidgets [40] and available at https://github . com/HypoModel/HypoModBase . At each 1-ms time step , the software processes input , membrane potential and spiking for each neuron in turn , using a single thread loop . A single run of a two cell type network with 200 neurons , simulated for 2000s , takes 43s . on an Intel i7-5960X processor running at 3 . 0GHz . The C++ source code for the model and the GA , and a working version of the software compiled for Windows PC is available at https://github . com/HypoModel/VMNNet/releases . The code for the model is specifically in file “vmnmod . cpp” . The software archive includes all the spike data and parameter files used to generate the figures in this paper . | When the needs of an animal require the execution of particular behaviours , the brain must decide which of these needs to prioritise–whether to flee from or fight an aggressor for example , or whether to hunt for food or pursue sex . The ventromedial nucleus of the hypothalamus is involved in such decisions , in the regulation of aggression , feeding behaviour and sexual behaviour . We began with evidence of the electrical activity patterns of these neurons , and from evidence of how they are interconnected . We built computational models to understand how the activity patterns could arise from mutually excitatory connections amongst simple neuron “types” . With a network framework that could explain the observed patterns , we asked what information processing features of such networks might be relevant for decision-making . Two important features arise as emergent properties of such networks; slow oscillatory rhythms–a phenomenon believed to be important for co-ordinating activity in different brain regions–and bistability . A bistable network has an upstate where neurons are active ( and , in this case , generating a coherent rhythm ) and a downstate , where neurons are inactive . Alternation between these states , we propose , reflects a switch between different behavioural states . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"action",
"potentials",
"medicine",
"and",
"health",
"sciences",
"neural",
"networks",
"membrane",
"potential",
"microbiology",
"signaling",
"networks",
"electrophysiology",
"neuroscience",
"network",
"analysis",
"computational",
"neuroscience",
"excitatory",
"postsynaptic",
"potentials",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"internal",
"ribosome",
"entry",
"site",
"viral",
"replication",
"cellular",
"neuroscience",
"cell",
"biology",
"virology",
"physiology",
"neurons",
"single",
"neuron",
"function",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"computational",
"biology",
"neurophysiology"
]
| 2019 | Emergent decision-making behaviour and rhythm generation in a computational model of the ventromedial nucleus of the hypothalamus |
The human T-lymphotropic virus type 1 ( HTLV-1 ) which is associated with the diseases of adult T-cell leukemia/lymphoma , HTLV-1 associated myelopathy / tropical spastic paraparesis ( HAM/TSP ) and HTLV-associated uveitis , can cause transfusion-transmitted infections . Although HTLV screening of blood donors was already routinely performed in developed countries , little is know about the HTLV prevalence among blood donors in developing countries which do not perform HTLV screening , such as China . To systematically characterize the prevalence of HTLV infection among bloods in southeast China . A 10-year survey for HTLV prevalence in blood donors was performed in Xiamen , southeast China , during 2004-2013 . The HTLV-1/2 of blood donations were screened by enzyme-linked immunosorbent assay , following with confirmation by western blot assay and 9nucleic acid testing . The HTLV-1 prevalences in donors from different cities were calculated . Viral sequences derived from identified HTLV-positive cases were sequenced and analyzed . Among 253 , 855 blood donors , 43 were confirmed to be seropositive for HTLV-1 ( 16 . 9 per 100 , 000 95% CI: 12 . 3-22 . 8 ) and none HTLV-2 infection was found . The HTLV-1 prevalence varied significantly in donors from different cities . Donors from cities in Fujian province ( 24 . 3 per 100 , 000 , 95%CI: 17 . 4-33 . 1 ) had a significantly higher ( p=0 . 001 ) HTLV-1 seroprevalence than those who were born in non-Fujian cities ( 3 . 4 per 100 , 000 , 95%CI: 0 . 7-9 . 8 ) . Among nine cities in Fujian province , the highest prevalence was found in blood donors from Ningde ( 171 . 3 per 100 , 000 , 95%CI: 91 . 3-292 . 8 ) which is a coastal city in the northeast of Fujian . Molecular characterization of viral sequences from 27 HTLV-1 carriers revealed 25 were Transcontinental subtype of genotype A and 2 were Japanese subtype of genotype A . Interestingly , 12 of 25 Transcontinental subtype sequences harbored a characteristic L55P mutation in viral gp46 protein , which was only presented in the Transcontinental subtype sequences from Japan and Taiwan but not in that from other countries . Although China is considered to be a non-endemic region for HTLV , the HTLV-1 prevalence in blood donors is significantly higher in Fujian province , southeast China . A higher prevalence of HTLV-1 in the Fujian may be attributed to endemic foci in the city of Ningde .
Human T-lymphotropic virus ( HTLV ) is a retrovirus which has been known for about 35 years since it was first isolated from a patient with a T-cell malignancy [1] . Previous studies had revealed the etiologic role of HTLV infection in the diseases of adult T-cell leukemia/lymphoma ( ATL ) , HTLV-1 associated myelopathy / tropical spastic paraparesis ( HAM/TSP ) and HTLV-associated uveitis ( HAU ) [2 , 3] . There are four HTLV related viruses had been identified: HTLV-1 , HTLV-2 , HTLV-3 and HTLV-4 [4 , 5] . However , only HTLV-1 has been convincingly linked to human diseases at present . The most important routes of HTLV transmission were found to be from mother to child and predominantly through breastfeeding and blood contact [3] . The efficiency of the mother-to-child transmission route is estimated to be 20% [6] . Besides vertical transmissions , horizontal transmissions of HTLV are also usual , possibly resulting from unprotected sex , multiple sexual partners , lifetime contact with an HTLV infected partner and transfusion of blood not tested for HTLV [7] . Compared to other transmission route , intravenous exposure to virus-contaminated blood is the most efficient model of HTLV transmission [7] . Globally , approximately 20 million people are estimated to be infected by HTLV-1 , and 90% of them remain asymptomatic carriers during their lives [8] . Previous studies had revealed that Japan , Central and Western Africa , the Caribbean islands and Central and South American were the regions with the highest HTLV-1 prevalence in the world [1] . The seroprevalence in these areas were reported to be higher than 5% of the population tested [3] . Because of the absences of effective treatment options and preventive vaccine for HTLV-associated diseases , current prevention approach of new HTLV infections are mainly depended upon the effective control of viral transmission , either blood transfusion or sexually intercourse . Since 1993 , HTLV screening of blood donors was already performed in all developed countries and in some developing countries where the virus is endemic . According to previous epidemiological data , most regions of China were considered as a non-endemic area for HTLV infection [9 , 10] . However , the Fujian province in southeast China was found to be a high endemic area for HTLV infection [11] . Hence , we performed a 10-year blood screening survey to systematically characterize the prevalence of HTLV infection among bloods in Fujian province in southeast China since 2004 .
The study was conducted in accordance with the guidelines of the 1975 Declaration of Helsinki and the principles of good clinical practice . All procedures were approved by the Medical Ethical Committee of Xiamen Blood Services . Written informed consent was obtained from all subjects . From 15 February 2004 and 31 December 2013 , all voluntary blood donors at the Xiamen Blood Service were regularly tested for HTLV-1/2 infection . An anti-HTLV-1/2 enzyme-linked immunosorbent assay ( Wantai , Beijing , China ) was used for initial screening of the infection . The assays were performed on an ELISA STARlet automated system ( Hamilton , Bonaduz , Switzerland ) according to the manufacturers’ instructions . The assay had been validated to be sensitive and specific as previously described [11] . All positive samples in the initial assay were repeated for twice using the same assay , followed by a confirmation testing using Western blot ( HTLV blot 2 . 4 , Genelabs Diagnostics , SciencePark , Singapore ) . An HTLV infection episode is defined as positive in both ELISA and Western blot testing . Other blood screening markers , including hepatitis B surface antigen ( HBsAg ) , anti-human immunodeficiency virus ( HIV ) , anti-Treponemapallidum ( the syphilisspirochete ) , and anti-hepatitis C virus ( anti-HCV ) antibodies were detected by the use of the Murex ELISA products ( Abbott Murex , Dartford , UK ) . After signing an informed consent , a follow-up samples were collected from all blood donors infected by HTLV . Peripheral blood mononuclear cells ( PBMCs ) were separated from whole blood and stored at -80°C . Proviral DNA of HTLV-1 in PBMCs were extracted by using QIAamp DNA blood kit ( Qiagen , Hilden , Germany ) , and then were subjected to real-time PCR detection as previously described [12] . Real-time PCR positive samples were further amplified for the coding region of the HTLV-1 gp46 gene ( nt5222-nt6151 , according to HTLV-1 reference sequence of M33896 ) by nested-PCR [13] . The PCR products were purified and directly sequenced on an ABI Prism 3130X automatic genetic analyzer ( Applied Biosystems ) . All HTLV-1gp46 sequences obtained in this study and 28 reference sequence in the GenBank were selected for phylogenetic analysis by neighbor-joining with Maximum Composite Likelihood corrected distances in the MEGA4 package , using 1000 bootstrap replicates . The sequences obtained in this study and another 251 HTLV-1 reference sequences derived from different geographical areas and different populations containing viral env gene , collected from a public HTLV-1 molecular epidemiology database ( http://htlv1db . bahia . fiocruz . br ) [14] , were analysis for amino acid variability . The new nucleotide sequence data reported in this paper have been deposited in the GenBank databases under the accession numbers ( Table 1 ) For donors with multiple donations during the period , only the screening results of the first samples were calculated . Statistical analyses were performed by the Mantel-Haenszel χ2 test and Fisher’s exact test for categorical variables . Differences were considered significant at a 2-tailed p<0 . 05 . SPSS software version 17 . 0 was used for the all statistical analyses .
A total of 253 , 855 donors were tested for HTLV infection , 43 among them were positive for antibody against HTLV-1 , none was positive for anti-HTLV-2 , as suggested by ELISA and confirmed by Western blot ( Fig 1 ) . All 43 HTLV seropositive donors were also positive in HTLV-1 proviral DNA by real-time PCR . Among them , 2 were co-infected with Treponemapallidum , 1 was co-infected with HCV , none was found to be co-infected with HBV or HIV . The overall prevalence of HTLV-1 infection was 16 . 9 per 100 , 000 ( 95% CI , 12 . 3–22 . 8 ) . The demographic characteristics of the donors were summarized in Table 2 . The HTLV-1 infected donors occurred every year during the study , without significant yearly variance ( Table 2 ) . Twenty eight ( 65% ) HTLV-1 carriers were men but the prevalence of men is similar as that of the women . The median age of the carriers is 27 . 5 ( range 18–47 , mean 28 . 4±7 . 1 ) . The prevalence in donors aged from 26–35 years ( 26 . 5 / 100 , 000 ) is significantly higher than that of the younger ( 12 . 2/100 , 000 ) . Most of the donors are born in Fujian province ( 40/43 , 93% ) . The prevalence of Fujianese ( 24 . 3/100 , 000 ) is 7 . 24 times ( 95% CI: 2 . 24–23 . 43 ) higher than that of non-Fujianese ( 3 . 4/100 , 000 ) . The HTLV-1 carries occurred in all nine major cities in Fujian Province , the highest prevalence ( 171 . 3 / 100 , 000 ) occurred in donors born in Ningde City , a coastal city in the northeast of Fujian ( Fig 2 ) . The HTLV-1 gp46 gene from 27 of the 43 HTLV-1 infected donors were successfully amplified and sequenced . Phylogenetic analysis indicated that 25 sequences ( 25/27 , 92 . 6% ) were clustered into Transcontinental subtype of genotype A and 2 remaining sequences ( FJ05-FZ and FJ19-ZZ ) were clustered into Japanese subtype of genotype A ( Fig 3 ) . The coded amino acid sequences of the isolated gp46 genes were further analyzed ( Fig 4 ) . Ten strains ( FJ10-ND , FJ12-PT , FJ13-LY , FJ14-XM , FJ15-ND , FJ16-SM , FJ17-SM , FJ22-PT , FJ23-XM and FJ26-PT ) presented identical amino acid sequence with the consensus sequence of HTLV-1 genotype A , whereas the remaining 16 strains harbored at least 1 amino acid substitution . A characteristic L55P mutation in the receptor binding domain ( RBD ) of gp46 protein [15] was presented in 12 sequences ( 48 . 0% ) of Transcontinental subtype . To further understand the polymorphism of the L55 site , we collected 251 sequences containing RBD region and their geographic origin information ( as shown in S1 Table ) from a public HTLV-1 database ( http://htlv1db . bahia . fiocruz . br ) . The comparison indicated that the L55P mutation was only observed in Transcontinental subtype strains originated from Japan ( 9/20 , 45 . 0% ) , Taiwan ( 4/11 , 36 . 4% ) and China ( 1/2 , 50% ) , but not appeared in strains isolated from Transcontinental subtype strains originated from other Asia countries ( 0/3 ) , Africa ( 0/6 ) , Central America ( 0/29 ) , Europe ( 0/22 ) , North America ( 0/12 ) , South America ( 0/38 ) , but also in other subtypes of genotype A ( 0/67 ) or genotype B/C ( 0/40 ) .
This was the first longitudinal and more complete coverage study about HTLV infection of blood donors in China mainland . The key findings include: HLTV-1 infection persist in many cities of Fujian Province , the southeast coast of China . None HTLV-2 was found . The prevalence in blood donors was 16 . 9 per 100 , 000 ( 95% CI: 12 . 3–22 . 8 ) , Ningde people up to 171 . 3 ( 95% CI: 91 . 3–292 . 8 ) ; molecular analyses demonstrated most of HTLV-1 isolates ( 25/27 , 92 . 6% ) in our study belong to Transcontinental subtype of genotype A; 12 of 25 Transcontinental subtype sequences harbored a characteristic L55P mutation in viral gp46 protein , which was only presented in the Transcontinental subtype sequences from Japan and Taiwan but not in that from other countries . The HTLV prevalence varies significantly in different geographic areas . The World Health Organization ( WHO ) has advised that decisions regarding the screening of blood donations for HTLV be guided by local epidemiological evidence . In most Asia countries except Iran and Japan , the HTLV infection appears rare . However , epidemiological data of most areas of Asia , particularly of the China mainland , were very little due to the lack of large-size and representative studies [10] . In China mainland , according the limited reports , the Fujian province in southeast China was suggested to be a relative HTLV endemic region . In 2005 , a cross-sectional study reported a total of 19 confirmed HTLV-1 positive cases among 145 , 293 donors from 13 provinces of China and indicated all 19 HTLV-1 carrier donors were from Fujian province [11] . A recent study conducted in the years of 2012 to 2013 identified 38 confirmed HTLV carriers among 122 , 468 blood donors from 9 provinces containing 19 blood banks in China . Of the 38 positives , 34 live in Fujian province and the remaining 4 cases were from the Guangdong and Zhejiang provinces that are neighboring to Fujian [10] . We started HTLV screening of blood donors in Xiamen city of Fujian province since 2004 to prevent HTLV associated transfusion-transmitted infection . Through a 10-year survey , as the results presented in this study , we found 43 confirmed HTLV-1 carriers among 253 , 855 blood donors that indicated a prevalence of 16 . 9 per 100 , 000 ( 95%CI: 12 . 3–22 . 8 ) in our study population . This prevalence was much lower than those in southern Japan , Sub-Saharan Africa , and the Caribbean area , similar with those in Taiwan , Europe and North American [16 , 17] . However , one limitation of our study which should be noted that it was conducted in blood donors instead of in general population . The prevalence data derived from blood donors which were subject to selection by the blood center and by self-selection for good health and altruism , introducing bias and likely underestimation of the HTLV prevalence in the general population . Thus , the true HTLV prevalence in general population of southeast China should be higher than that we observed . Our study revealed that the HTLV-1 prevalence in donors from Fujian province ( 24 . 3 per 100 , 000 , 95%CI: 17 . 4–33 . 1 ) was about 7-fold ( OR = 7 . 24 , 95%CI , 2 . 24–23 . 43 ) higher than that in donors from non-Fujian provinces ( 3 . 4 per 100 , 000 , 95%CI: 0 . 7–9 . 8 ) . These data confirmed previous findings regarding Fujian was a HTLV endemic region in China . Furthermore , for the first time , we found the HTLV-1 prevalence significantly varied among donors from nine different cities in Fujian province and it was the highest in donors from Ningde ( 171 . 3 per 100 , 000 95%CI: 91 . 3–292 . 8 ) . Statistical analysis demonstrated the HTLV-1 prevalence among donors from Ningde was significantly higher than any other cities of Fujian ( p<0 . 05 , respectively ) , and two neighboring cities ( Nanping and Sanming ) of Ningde also presented relatively higher prevalence ( Fig 2 ) . A new study published recently also noted that donors in Ningde had a very high HTLV-1 prevalence ( 22/5534 ) , which was consistent with our finding [18] . These data suggested this city was an endemic focus may essentially contribute to the high HTLV-1 prevalence of Fujian region . Further studies should be performed to investigate the risk factors and transmission routes of HTLV-1 infection in these regions , aiming to develop effective control strategies toward HTLV infection . Molecular analyses demonstrated most of HTLV-1 isolates ( 25/27 , 92 . 6% ) in our study belong to Transcontinental subtype of genotype A . Interestingly , two cases ( 2/27 , 7 . 4% ) carrying Japanese subtype of genotype A were found for the first time in China . Previous study had revealed that the Japanese subtype was a predominant viral strain in some populations of Japan and was also found in neighboring Taiwan [19] . The presence of Japanese subtype in blood donors of southeast China suggested a potential possibility of the introduction of this subtype into southeast China from Japan . In contrast to Japanese subtype , the Transcontinental subtype distributes worldwide . The proportion of the Transcontinental subtype was about 30% in Japan , whereas it was about 70% in Taiwan which is neighboring area of Fujian . Our results demonstrated the HTLV-1 sequences of Transcontinental subtype in Japan , Taiwan and China shared a unique gp46 L55P mutation which was not presented in sequences originated from other geographic regions . This frequencies of the L55P mutation among HTLV-1 in Japan , Taiwan and China were very similar ( about 36 . 4–50 . 0% ) , that suggested a close relationship in between Transcontinental subtype viral strains of these regions and these viruses possibly shared a common evolutionary ancestor . Furthermore , potential influence of such a common mutation on viral function phenotype and clinical course certainly warrant further investigation . In summary , the HTLV-1 prevalence in blood donors is significantly higher in southeast China , especially in the northern cities of Fujian province , such as Ningde . A surveillance system should be implemented to evaluate residual risk of transfusion-transmitted HTLV-1 infection in different regions of China . Moreover , similar molecular characteristics of prevalent HTLV-1 sequences in southeast China , Taiwan and Japan suggested a same origin of these viruses . | The human T-lymphotropic virus type 1 ( HTLV-1 ) which is associated with the diseases of adult T-cell leukemia/lymphoma , tropical spastic paraparesis etc . , can cause transfusion-transmitted infections , it also can be transmitted by sex or breastfeeding . Globally , approximately 20 million people are estimated to be infected by HTLV-1 , and 90% of them remain asymptomatic carriers during their lives . Previous studies had revealed that Japan , Central and Western Africa , the Caribbean islands and Central and South American were the regions with the highest HTLV-1 prevalence in the world . Little is know about the HTLV prevalence in China . We performed a 10-year blood screening survey to systematically characterize the prevalence of HTLV infection among bloods in Fujian province in southeast China since 2004 . The HTLV-1 prevalence in blood donors is significantly higher in southeast China , especially in the northern cities of Fujian province , such as Ningde . Moreover , similar molecular characteristics of prevalent HTLV-1 sequences in southeast China , Taiwan and Japan suggested a same origin of these viruses . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| []
| 2015 | The Prevalence of Human T-Lymphotropic Virus Infection among Blood Donors in Southeast China, 2004-2013 |
Deep-coverage metabolomic profiling has revealed a well-defined development of metabolic decay in human red blood cells ( RBCs ) under cold storage conditions . A set of extracellular biomarkers has been recently identified that reliably defines the qualitative state of the metabolic network throughout this metabolic decay process . Here , we extend the utility of these biomarkers by using them to quantitatively predict the concentrations of other metabolites in the red blood cell . We are able to accurately predict the concentration profile of 84 of the 91 ( 92% ) measured metabolites ( p < 0 . 05 ) in RBC metabolism using only measurements of these five biomarkers . The median of prediction errors ( symmetric mean absolute percent error ) across all metabolites was 13% . The ability to predict numerous metabolite concentrations from a simple set of biomarkers offers the potential for the development of a powerful workflow that could be used to evaluate the metabolic state of a biological system using a minimal set of measurements .
The data generated from deep coverage omics tools are becoming broadly available and thus their use is becoming more common [1 , 2] . With this data , researchers have begun to identify metabolomics biomarkers that can be used to describe systemic behavior with only a few inexpensive and reliable measurements [3–7] . In transfusion medicine , deep coverage metabolomics data sets for human red blood cells ( RBCs ) in cold storage are rapidly accumulating [8] and have been used to characterize the state of the RBC metabolic network during storage [9–13] . Big data analysis of RBC metabolomics data has yielded a well-defined three-phase pattern of metabolic storage lesion that has fundamental consequences for blood storage [10 , 13] . Recently , eight extracellular metabolic biomarkers have been identified that reliably define this three-phase decay process observed in RBCs [6] . These biomarkers ( adenine , glucose , hypoxanthine , lactate , malate , nicotinamide , 5-oxoproline , and xanthine ) recapitulate the qualitative trend of the entire metabolome . However , it has yet to be determined whether these biomarkers can be used to predict quantitative network behavior . In this study , we determine that five of the eight biomarkers ( glucose , hypoxanthine , lactate , malate , and xanthine ) are not only excellent qualitative predictors , but also accurate quantitative predictors of metabolic concentrations in the rest of the metabolic network . Using a simple computational formulation [14] prevalent in a variety of fields [15–18] , we extend the utility of these biomarkers by using them to quantitatively predict the concentration profiles of 91 other metabolites in the network . This added use of validated biomarkers offers the potential for a powerful workflow that utilizes five biomarkers to evaluate the state of RBC metabolism .
For this study , we used the metabolomics data set from Bordbar et al . [10] that measured 96 intracellular and extracellular metabolites in human red blood cells under storage conditions . The data set measured 14 time points over a 45 day time period for 20 biological replicates . For the purposes of modeling , we randomly divided these 20 replicates into equal sized training and testing sets of 10 samples . We observed a high amount of variability in the extracellular glucose measurement at Day 31 ( S1 Fig ) , a behavior which was not observed in the intracellular glucose measurement ( S2A Fig ) but was seen in other extracellular measurements at Day 31 ( S2B Fig ) . In order to avoid bias arising from the inclusion of potentially erroneous data , we excluded the measurements from Day 31 , resulting in 13 total time points spanning 45 days of storage . We trained multiple polynomial models of varying complexity on the concentration profiles of the biomarkers and the concentration profile of the target metabolite ( Fig 1 ) . The best performing model was a simple , linear Output-Error model [14] . Variation between blood bags is a known challenge , as both donor and technical factors contribute to sample heterogeneity [1] . Due to this variation , we noted that simply because these eight biomarkers are good qualitative predictors of systemic behavior does not imply that they are also good quantitative predictors . We therefore performed a feature selection and cross validation within the eight biomarkers , determining that adenine , nicotinamide , and 5-oxoproline were not able to quantitatively predict systemic behavior as well as the other five biomarkers ( see Methods ) . Thus , glucose , hypoxanthine , lactate , malate , and xanthine were used for the remaining analysis . In order to generate a prediction for each metabolite , we trained the model using the five biomarkers and a measured profile for the target metabolite as input ( Fig 1 ) . We used an ensemble modeling approach [19] to reduce bias arising from using either individual replicates or averaging replicates to train a single model . With 10 training replicates , this approach allowed us to generate an ensemble of trained models that inherently includes the biological variation of the training data ( S3 Fig ) . We then used this trained ensemble computational model to predict a consensus concentration profile of a target metabolite , this time only using the biomarkers as input ( Fig 1 ) . We tested the model’s capabilities by comparing the predicted profiles of the remaining 91 measured metabolites to their measured profiles ( Fig 2 ) . We calculated the symmetric mean absolute percentage error ( SMAPE ) for each predicted concentration profile , resulting in a median error of 0 . 1340 ± 0 . 1505 ( S4 Fig ) . See Supplementary Material for all predicted profiles . To further validate our model , we compared against 10 , 000 profiles generated using a naive random walk for each metabolite . The naive random walk model assumes that metabolite concentration changes over time are independent of each other and are normally distributed . The random walk is a widely used benchmark for dynamic forecasting models [20] . When a significant number ( ≥500/10 , 000 , i . e . , ≥5% ) of random walks outperform a trained model for a metabolite , we conclude that the dynamics of that metabolite are indiscernible from noise for the data given ( see Methods for details on the random walk comparison ) . Despite the complexity of RBC metabolism , we found that 84/91 ( 92% ) of RBC metabolites were predicted more accurately than random walks using five biomarkers as input ( p < 0 . 05 ) . In an effort to lend biological intuition to this surprising result , we viewed these results in the context of the complete RBC metabolic network ( Fig 2 , S5 Fig ) . The map highlights several points . First , the five biomarkers are largely distributed across key subsystems . Surprisingly , two biomarkers are adjacent in the network: xanthine and hypoxanthine . From inspection of the map , it becomes more intuitive that to unambiguously predict IMP levels ( Fig 2 ) , both biomarkers need to be quantitatively measured .
RBCs in storage undergo a series of morphological changes ( commonly referred to as “storage lesion” ) that become more pronounced throughout the storage process [1 , 21 , 22] . Recent studies have shown that blood transfused after being stored for longer than five weeks is associated with post-transfusion complications [23 , 24] , indicating the serious clinical implications of metabolic decay in transfused blood . With the recent identification of eight extracellular biomarkers that are able to define this decay , the field of transfusion medicine now has an opportunity to define the metabolic state of stored RBCs with just a few measurements . Thus , there is a need for predictive modeling methods that can extend the applicability of these biomarkers to provide deeper understanding of the metabolic state of RBCs collected and stored under blood banking conditions using current and future technologies ( e . g . , improved bags or storage solutions , pathogen reduction technologies ) . In this study , we have developed a statistical model that uses these biomarkers to predict the time series concentration profiles of other metabolites in the RBC metabolic network . This powerful tool was rigorously validated to avoid overfitting through model ( complexity ) and feature selection , and comparing against a standard forecasting baseline model ( i . e . , naive random walk ) . As with any data modeling approach , the performance of a model is dependent upon the quality of the input data; this is no exception here . We see that certain metabolites ( e . g . , ADP , inosine ) had higher prediction errors , which can be partially attributed to noise in the training data and to low concentrations ( S6 Fig ) . The results presented here have two important implications . First , we have shown that if good biomarkers are available for a given system ( like for the human RBC ) , then they can be used to make quantitative predictions about systemic behavior . Second , this provides the potential for a cost-effective workflow to monitor the metabolic state of a biological system since the only input under new conditions is the concentration profiles of biomarkers . Through the use of modeling and statistical analysis , the measured and predicted concentrations would enable a quantitative understanding of systems-level behavior . Thus , we have demonstrated the predictive power of biomarkers through the use of a statistical model for RBCs in storage . This data-driven statistical modeling approach performed remarkably well for the RBC system , even without a detailed kinetic model . These results are encouraging and provide a complementary approach for predicting metabolite dynamics in less characterized organisms . As our validation procedure indicates , a critical mass of high-quality data is required to extract meaningful signals from noise . Our workflow provides a valuable assessment on whether this critical mass has been satisfied; the results here indicate that as few as 20 biological replicates are sufficient to provide a training set capable of achieving >90% accuracy . Follow up studies should address the question of how many measurements need to be made during storage in order to provide a reliable assessment of the RBC metabolome during storage , as this question has direct clinical implications . As biomarkers are identified for new systems , there will be a need to analyze omics in an attempt to efficiently characterize complex biological systems using just these few informative measurements . Our workflow addresses this need by incorporating such biomarkers with a statistical model , offering broad utility in both the laboratory and the clinic .
An Output Error ( OE ) model [14] predicts system dynamics from past values , measured inputs , and unmeasured disturbances as follows: y ( t ) = ∑ i = 1 n B i ( q ) F i ( q ) u i ( t - n k i ) + e ( t ) ( 1 ) where y ( t ) is the output at time t , ui is an input ( i . e . , metabolite i ) , e is the unmeasured disturbance ( i . e . , system noise ) , and B ( q ) and F ( q ) are polynomials expressed in the time-shift operator q as follows: B ( q ) = b 1 + b 2 q - 1 + … + b n b q - n b + 1 ( 2 ) F ( q ) = 1 + f 1 q - 1 + … + f n f q - n f . ( 3 ) For this system , n = 5 ( i . e . , the five biomarkers ) , nb = 1 , nf = 0 , and there was no input delay ( nk = 0 ) . The B and F polynomials are estimated during the system identification step using least squares regression to minimize the difference between the measured signal and the predicted output . This OE model performed better than more complex OE models having higher nb and more complex polynomial models . It also performed better than simpler linear regression—the OE model thus represents an optimal degree of complexity . In order to evaluate the accuracy of the predicted concentration profiles for the various metabolites , we calculated the symmetric mean absolute percentage error ( SMAPE ) , given by: SMAPE = 1 n ∑ t = 1 n | y t - y ^ t | y t + y ^ t ( 4 ) where n is the number of time points , y is the measured concentration profile , and y ^ is the predicted concentration profile . For the global statistics reported in S4 Fig , the mean of the SMAPE of the 10 predicted profiles is given . We trained the OE model using a recently published metabolomics data set of RBCs under storage conditions at 4°C with 20 biological replicates from Bordbar et al . [10] . In order to predict the concentration of target metabolites , we used the eight extracellular biomarkers [6] as input since they are highly representative of the qualitative behavior of the rest of the system . In order to determine if these biomarkers are also good quantitative predictors , we performed a 10-fold cross validation on the set of 10 samples used for training the model to verify the generalization performance of the trained model . We ran our cross validation on all 56 combinations of five biomarkers ( i . e . , 8 choose 5 ) ; the five selected biomarkers had a mean SMAPE of 10 . 33% , which was within 1% of the top performing set of five biomarkers . Thus , we used glucose , hypoxanthine , lactate , malate , and xanthine as the final set of biomarkers input to the OE model . We trained an ensemble of OE models using the five biomarker profiles and each of the 91 measured metabolite profiles . Thus , we trained 91 ensemble models ( one ensemble for each metabolite ) . Each ensemble model consisted of 10 OE models , each trained on a biological replicate . We used Bags 1–10 as this training set . We combined the outputs of these 10 OE models into a single prediction for each metabolite by computing the median of the 10 predictions at each time point ( S3 Fig ) . This ensemble modeling approach captures the biological variability inherent among the samples used for training . We used Bags 11–20 as the testing data set . In order to assess the variability between the training and testing data , we performed a two-sample t-test at each time point for each metabolite . This showed that approximately 24% of the data rejected the null hypothesis ( FDR-adjusted p < 0 . 05 ) that the two data sets came from the same distribution and also showed greater than a 20% difference in the mean concentrations at a given time point ( S7 Fig ) . For each test replicate , the five biomarkers were input to the trained ensemble model . In addition to the prediction error , as computed by SMAPE , we also evaluated our model by comparing its performance against a benchmark model . We chose as a benchmark the random walk model , which assumes that metabolite concentration changes over time are independent of each other and are normally distributed with zero mean . The random walk model is commonly used to benchmark dynamic forecasting models [20] . To ensure that the random walk was representative of the metabolite concentration changes , we estimated the standard deviation of random changes from all 10 testing replicates across all time points for each metabolite . We further ensured that the random walk was an appropriate benchmark by initializing with a realistic concentration . To do so , we randomly chose from the pool of the 10 measured starting points of the testing replicates for each metabolite . We generated 10 , 000 of these random walk profiles for each metabolite . In order to compare these to our model predictions , our null hypothesis was that our trained model performed no better than the random profiles . We calculated the SMAPE for each of the random profiles and compared to the SMAPE for the predicted profiles; the given p value is the number of random profiles which had a lower SMAPE than the average of the predicted profiles for that metabolite . | While deep-coverage omics data sets are allowing for more complete characterization of biological systems , there has been a concerted effort to identify a subset of measurements that are representative of qualitative network-level behavior . For some systems—like the human red blood cell ( RBC ) —such biomarkers have already been identified . Using the concentration profiles of these biomarkers as input to a statistical model , we predict quantitative concentration profiles of other metabolites in the RBC network . These results demonstrate that if good biomarkers are available for a biological system , it is possible to use these measurements to gain insight into the quantitative state of the rest of the network . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"carbohydrate",
"metabolism",
"metabolic",
"networks",
"mathematical",
"models",
"biomarkers",
"glucose",
"metabolism",
"metabolomics",
"mathematics",
"metabolites",
"forecasting",
"statistics",
"(mathematics)",
"network",
"analysis",
"research",
"and",
"analysis",
"methods",
"random",
"walk",
"computer",
"and",
"information",
"sciences",
"behavior",
"mathematical",
"and",
"statistical",
"techniques",
"biochemistry",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"metabolism",
"statistical",
"methods"
]
| 2017 | Biomarkers are used to predict quantitative metabolite concentration profiles in human red blood cells |
The Asian tiger mosquito ( Aedes albopictus ) is an important vector for pathogens that affect human health , including the viruses that cause dengue and Chikungunya fevers . It is also one of the world's fastest-spreading invasive species . For these reasons , it is crucial to identify strategies for controlling the reproduction and spread of this mosquito . During mating , seminal fluid proteins ( Sfps ) are transferred from male mosquitoes to females , and these Sfps modulate female behavior and physiology in ways that influence reproduction . Despite the importance of Sfps on female reproductive behavior in mosquitoes and other insects , the identity of Sfps in Ae . albopictus has not previously been reported . We used transcriptomics and proteomics to identify 198 Sfps in Ae . albopictus . We discuss possible functions of these Sfps in relation to Ae . albopictus reproduction-related biology . We additionally compare the sequences of these Sfps with proteins ( including reported Sfps ) in several other species , including Ae . aegypti . While only 72 ( 36 . 4% ) of Ae . albopictus Sfps have putative orthologs in Ae . aegypti , suggesting low conservation of the complement of Sfps in these species , we find no evidence for an elevated rate of evolution or positive selection in the Sfps that are shared between the two Aedes species , suggesting high sequence conservation of those shared Sfps . Our results provide a foundation for future studies to investigate the roles of individual Sfps on feeding and reproduction in this mosquito . Functional analysis of these Sfps could inform strategies for managing the rate of pathogen transmission by Ae . albopictus .
Aedes albopictus , the Asian tiger mosquito , is an important species from both an epidemiological and an ecological perspective . Epidemiologically , it has the potential ability to transmit over 20 viruses [1] , [2] , and it plays a significant and growing role across the world as an important vector of several pathogens including those that cause dengue and Chikungunya fevers [2]–[5] . Ecologically , Ae . albopictus is considered to be one of the world's fastest-spreading invasive animal species [6] . While native to East Asia , it has recently colonized every continent except Antarctica ( most recently reviewed by [5] ) , and its range is expected to grow in the future [2] , [7] , [8] . The impact of this range expansion on disease spread is difficult to predict [9] , [10] , but it will likely pose additional threats to public health [5] . Consequently , there is an urgent need to develop effective strategies for controlling the reproduction and spread of Ae . albopictus [11] . One step toward managing the reproduction of Ae . albopictus is to investigate seminal fluid proteins ( Sfps ) , which are proteins that males transfer to females during mating . Sfps in insects are crucially important for male reproductive success , and they modulate several aspects of female post-mating behavior and physiology [12] , [13] . In Ae . albopictus , receipt of Sfps bolsters egg development under poor blood feeding conditions [14] , increases egg laying [15] , and inhibits female remating [16] , [17] . Interestingly , some of these Sfp-induced effects can last throughout the life of the female , even when she receives a only a very small dose of Sfps [17] . Since Sfps modify female behavior so drastically , their identification and functional characterization may provide promising targets for the control of insects that transmit disease-causing organisms [18]–[20] . Sfps have been identified for a number of insects , including Drosophila melanogaster [21]–[23] , medflies [24] , [25] , and species of sand flies [26] , honey bees [27] , butterflies [28] , [29] , flour beetles [30] , and crickets [31]–[34] . Sfps have recently been identified in two mosquito species , the malaria vector Anopheles gambiae [19] , [20] , [35] and the yellow fever mosquito Aedes aegypti [36] . Identifying the Sfps found in particular mosquito species can pave the way for investigations of reproduction-related roles played by individual Sfps [19] . This knowledge , together with comparisons of Sfps across species , might lead to novel or improved control strategies for these mosquitoes [20] , including Ae . albopictus . Here we use transcriptomic and proteomic approaches to identify and characterize Ae . albopictus Sfps . We used an isotope labeling technique from Findlay et al . [22] and adapted by Sirot et al . [36] for mosquitoes to identify male proteins in mated females after copulation . We identified the transferred proteins by comparing the mass spectra of proteins in our samples against the spectra from a predicted protein database . This predicted protein database was generated from sequenced transcriptomes of the male and female reproductive tract . Using this technique , we identified 198 Ae . albopictus putative Sfps .
Our methods are very similar to those of our recent study in Ae . aegypti [36] . We therefore note methodological similarities where appropriate and focus primarily on differences in our methodology . We describe the methods we used to sequence the transcriptomes of the male and female reproductive tract and generate the predicted protein database in the supporting information ( Text S1 ) . As in our study of Ae . aegypti [36] , to distinguish male-derived proteins from the female proteins in the reproductive tract of mated females , we adapted a stable-isotope labeling method originally used for D . melanogaster Sfp identification [22] . Stable-isotope labeling of proteins shifts the mass to charge ratio of the peptides such that they are unidentifiable by mass spectrometry because the observed spectra do not match predicted spectra generated from a protein database . Therefore , to identify only the male-derived proteins in mated females , we mated males reared on a standard diet to females reared on stable-isotope labeled yeast diet ( 15N-labeled yeast ) . To verify the effectiveness of the labeling , we reared as controls two groups of females: virgin females reared on the stable-isotope labeled yeast diet , and virgin females reared on an unlabeled diet . Aedes albopictus ( New Jersey strain ) were used for our study and were reared as described previously [36] . Pupae were placed into individual vials until they emerged as adults to ensure the virgin mating status of all individuals used in the experiment . Adult females were housed in 5 L bucket cages containing of up to 70 females from the same treatment ( 15N-labeled or unlabeled diet ) , and adult males were housed in 5 L bucket cages of up to 50 males . All adults were given free access to a 20% sucrose solution . For mating , each 15N-labeled female ( 4–6 days post-eclosion ) was transferred into a cage containing 40–50 unlabeled males ( 4–5 days post-eclosion ) . Matings lasted for no longer than three minutes , and when the pair began to separate at the end of mating , the female was collected and placed on ice ( for no more than 10 minutes ) until dissection . To obtain 15N-labeled mated female tissue samples , the reproductive tract below the ovaries was dissected out in 20 µl Dulbecco's PBS ( DPBS ) with protease inhibitors ( Roche Complete Protease Inhibitor Tablets , Indianapolis , IN ) . Two sample types were collected ( supernatant and pellet ) , each of which was obtained using two independent biological replicates that consisted of tissues from 17 to 20 mated females . To verify the effectiveness of our labeling technique , the reproductive tract below the ovaries was dissected from 20 15N-labeled virgin females and 24 unlabeled virgin females ( 3–5 days post-eclosion ) . Samples were prepared and stored as in [36] . To distinguish Sfps from sperm proteins , we obtained sperm-enriched samples from the testes and seminal vesicles ( where sperm are stored ) for proteomic analyses following the methods described in [36] . Testes were dissected out from 40 virgin males ( 3–8 days post-eclosion ) , and seminal vesicles were dissected from an independent set of 30 virgin males ( 3–8 days post-eclosion ) . It is important to note that our transcriptome did not include testes tissues , so the proteins we identify in the testes samples are ones that are present both in the testes as well as the seminal vesicles and/or accessory glands . Proteins from the samples were separated using gel electrophoresis on one-dimensional 4–20% polyacrylamide Mini-Protean TGX precast gels ( Bio-Rad Laboratories , Hercules , CA ) , and gels were stained using SimplyBlue SafeStain ( Invitrogen , Carlsbad , CA ) . Each gel lane was then divided into several bands ( six to eight ) in order to maximize sensitivity of protein identification using mass spectrometry , and to facilitate estimation of the molecular weights of the identified proteins . All bands ( i . e . the entire lane for each sample ) were submitted for analysis . Proteins were identified through nanoLC-MS/MS analysis followed by comparison of the observed spectra to those generated from our transcriptome-based Ae . albopictus predicted protein database ( see Text S1 for information on the methods used to develop this predicted protein database ) . These analyses were performed at the Cornell University Proteomics and Mass Spectrometry Core facility . 1D gel bands were subjected to in-gel trypsin digestion/extraction and lyophilized . Tryptic peptides were reconstituted in 2% ACN with 0 . 5% FA for nanoLC-ESI-MS/MS analysis on an LTQ-Orbitrap Velos Mass Spectrometer ( Thermo-Fisher Scientific , San Jose , CA ) equipped with a “CorConneX” nano ion source device ( CorSolutions LLC , Ithaca , NY ) and coupled to an UltiMate3000 RSLCnano chromatograph ( Thermo ) . All MS and MS/MS spectra were processed using Proteome Discoverer 1 . 3 ( Thermo ) and the raw data were exported as MGF files for subsequent database searching using Mascot Daemon ( version 2 . 3 . 02 , Matrix Science , Boston , MA ) . The acquired spectra were searched against our custom Ae . albopictus transcriptome-based database containing 29 , 503 protein sequence entries with one missed trypsin cleavage allowed . Peptide mass tolerance was set to 20 ppm and MS/MS mass tolerance was set to 0 . 8 Da . Carbamidomethylation of cysteine was set as a fixed modification , oxidation of methionine as well as deamidation of asparagine and glutamine were set as variable modifications . A false discovery rate was estimated as described previously [36] . A peptide was considered to be a high quality peptide only if it met all of the following criteria: at or above the 99% confidence threshold , peptide score ≥31 that was also at or above the identity threshold level , expectation value ≤0 . 001 , and delta mass score ≤5 ppm . The mass spectrometry results were screened against our predicted protein database to identify proteins of high confidence of being transferred from males to females during mating . For the mated female samples ( supernatant and pellet samples ) , an identified protein was considered as “high confidence” if it had hits to the predicted spectra from either two high quality peptides in the same sample or one high quality peptide in two independent biological replicates . For proteins found in male samples ( seminal vesicles and testes samples ) , a protein was considered a high confidence protein if it had a hit from one high quality peptide , since only one biological replicate was analyzed for each of these two tissue types . To verify that our labeling technique was effective , the number of proteins identified in the reproductive tracts of unlabeled virgin females was compared with the number identified in labeled virgin females [36] . Using the criteria of one high quality peptide hit ( see above ) for a high confidence protein , 573 proteins were identified in the unlabeled virgin females but only six proteins ( Aa-3848 , Aa-15006 , Aa-35743 , Aa-38093 , Aa-63600 , Aa-136683 ) in the labeled virgin females . Since our labeling technique was 99% effective in masking proteins from labeled females , proteins we identified from the labeled mated females are most likely male-derived . Any high confidence hit proteins in the mated female samples were then classified as being putative sperm proteins or Sfps based on which male samples they were also found in . Specifically , proteins identified in sperm-enriched samples from both the testes and seminal vesicle samples were classified as putative sperm proteins . Otherwise , proteins were classified as Sfps . Three proteins in the mated female samples were found in both the sperm-enriched testes and the seminal vesicle samples but were also identified in the labeled virgin female samples ( see above; Aa-35743 , Aa-38093 , Aa-136683 ) , so were not classified as putative sperm proteins . Several of the high confidence proteins had very similar sequences and therefore likely represent products of either different alleles of a single gene or conserved gene duplicates . Since our intention is only to report Sfps that are likely to be functionally unique , similar protein sequences were grouped into clusters using BLASTp . Proteins were placed into the same cluster if: a ) they were within a minimum of 50% of the length of each other , b ) they had a degree of similarity equal or larger than 91% , and c ) if the extent of the match was at least 80% of the size of the smaller sequence . Only one protein from each cluster was reported , which was the protein that was most abundant in our proteomics study had the highest number of total peptide hits across all of our samples . If two proteins within a cluster had an equal number of hits , we reported the protein whose sequence began with methionine , and secondarily the protein with the highest number of reads in our transcriptome data . In all but two cases , all proteins within the same cluster had the same classification as being a sperm protein or Sfp in our study . For the two cases where proteins within a cluster were classified as sperm proteins and Sfps , the reported proteins were classified as “unknown Sfp . ” In order to identify other candidate Ae . albopictus Sfps , we also searched our mass spectra against the predicted protein database based on the Ae . aegypti genome . The methods and results of this search are presented in the supporting information ( Text S2 ) . Functional characterization of the proteins was performed by a program taking a vocabulary of ∼250 words and comparing it to matches on several databases , as previously described [37] . Based on these matches , proteins were classified into one of the following categories: cytoskeletal , extracellular matrix , metabolism ( including oxidant and detoxification ) , immunity , hormones , proteolysis regulators ( includes proteases and protease inhibitors ) , signal transduction , transporters and protein export machinery , RNA and protein synthesis ( includes transcription factors , transcription machinery , and protein synthesis ) . Proteins that were classified in a different category were classified as “other” ( includes bacterial product , nuclear export , nuclear regulation , protein modification , proteasome machinery , transposable element , salivary , storage , viral product ) . Proteins that were not assigned to a function were classified as “unknown . ” Annotations of these categories and of protein classes were reviewed manually by J . M . C . R . Putative orthologs of the Ae . albopictus Sfps were identified by comparing sequence similarity to proteins from full predicted protein sets from seven species: Ae . aegypti ( Vectorbase AaegL1 . 4 ) , Anopheles gambiae ( Vectorbase AgamP3 . 7 ) , Culex quinquefasciatus ( Vectorbase CpipJ1 . 3 ) , Drosophila melanogaster ( NCBI ) , Apis mellifera ( NCBI ) , Mus musculus ( Ensembl ) , and Homo sapiens ( Uniprot ) . A reference predicted protein set for Ae . albopictus was created by combining several available transcriptomes for the species: the reproductive tract transcriptome ( reported in this paper ) , the sialome [38] , and the oocyte/embryo/pharate larval transcriptome [39] . To account for potentially high redundancy across transcripts with different lengths , the transcripts in this combined database were clustered at the 99% identity level using the standalone version of the program CDHit [40] , [41] . Protein sequences were defined as orthologs if they were reciprocal-best BLASTp hits ( at the CDHit cluster level for Ae . albopictus ) having an E-value<0 . 001 . To determine whether the predicted orthologs were previously reported as Sfps or sperm proteins , the orthologs were cross-referenced with published lists of Sfps and sperm proteins in their respective species: Ae . aegypti ( Sfps and sperm: [36] ) , An . gambiae ( Sfps: [19] , [20] , [35] ) , D . melanogaster ( Sfps: [21]–[23]; sperm: [42] , [43] ) , A . mellifera ( Sfps: [27]; sperm: [44] ) , M . musculus ( Sfps: [45] , [46] sperm: [42] ) , and H . sapiens ( Sfps: [47]; sperm: [48] ) . We were unable to classify any orthologs in Cx . quinquefasciatus as Sfps or sperm proteins , and any orthologs in An . gambiae as sperm proteins , since lists of those proteins have not yet been published for those species . A dN/dS analysis was performed to estimate and compare rates of evolutionary change in putative housekeeping genes and in seminal fluid proteins in Ae . albopictus and Ae . aegypti .
We identified 881 , 128 expressed-sequence tags ( ESTs ) , which were assembled into 29 , 503 contigs ( hereafter “transcripts” ) representing possible gene products ( Table 1 ) . Of the tissues we sequenced transcripts from , 13 . 5% of the transcripts are found exclusively in the male reproductive tract ( hereafter males ) , 25 . 7% are found exclusively in the female reproductive tract ( hereafter females ) , and 60 . 8% have ESTs in the reproductive tracts of both males and females ( hereafter males and females ) . The transcripts encode proteins that represent a wide array of functional categories ( Figure 1A-C ) . The largest single group of transcripts found in the male , the female , and in both males and females could not be assigned any particular category ( “unknown” ) . Other abundant functional categories included metabolism , proteolysis regulators , RNA and protein synthesis , signal transduction , and transporters and protein export machinery . Some of the less abundant functional categories included cytoskeletal , extracellular matrix , hormones , and immunity . The distributions of transcripts across the functional classes are strikingly similar between the male-specific and female-specific sequences ( Figure 1A & B ) . These distributions stand in marked contrast to that of the transcripts found in both males and females ( Figure 1C ) . This difference is primarily due to the approximately halved proportion of transcripts encoding proteins with an unknown function in both male and female tissues compared to that of male-specific and female-specific tissues . Overall , 3 , 693 ( 12 . 5% ) of the transcripts are classified as encoding secreted proteins . Interestingly , sex-specific transcripts ( those expressed only in males or only in females ) are significantly more likely to encode secreted proteins than are transcripts found in both sexes ( χ2 = 1 , 143 . 0 , df = 1 , p<0 . 001 ) . We identified a total of 314 proteins that are made in male Ae . albopictus and are transferred to females during mating . Of these , 198 are putative seminal fluid proteins ( Figure 1D; Table S1 ) , based on the criteria that they were not found in both the testes and the seminal vesicles . The remaining 116 transferred proteins are putative sperm proteins ( Table S2 ) , based on the criteria that they were found in both the testes and seminal vesicles . These putative sperm proteins likely are a very limited subset of all sperm proteins in Ae . albopictus , as our predicted protein database was derived from the male accessory gland and seminal vesicles , and did not include transcripts from the testes . We therefore focus our paper on the 198 putative seminal fluid proteins . The amino acid sequences for all proteins reported in this paper are in Table S3 , Table S4 , and Table S5 . It is interesting to note that the majority of the Ae . albopictus Sfps ( 134 , or 67 . 7% ) are derived from transcripts found in both the male and female reproductive tracts , whereas the remaining one-third ( 64 , or 32 . 3% ) are derived from transcripts found exclusively in males ( Table S1 ) . This finding highlights the benefit of using a proteomics approach to identify Sfps rather than relying on the criteria of highly male-biased or male-specific expression . It further highlights the potential existence of proteins that might function in the reproductive tract of virgin females but are additionally transferred by males as Sfps during mating ( see [52] ) . Proteolysis regulators ( proteases and protease inhibitors ) commonly comprise a large subset of the Sfps in other insects and in mammals ( reviewed by [53] ) . In Ae . albopictus , 48 of the 198 Sfps ( 24% ) are predicted proteolysis regulators , which is similar to the percentage reported in other species ( for example , 20% of D . melanogaster Sfps and 14% of Ae . aegypti Sfps are predicted proteolysis regulators [22] , [36] ) . Proteolysis regulators generally play roles in activating and regulating proteins , and potentially coordinate actions of multiple Sfps [20] , [53] . Studies of proteolysis regulators in the Sfps of other species have revealed important functions related to reproduction , including roles in egg production , semen coagulation , sperm storage and activation , fertility , and protecting the female against infections , and pathogen transmission ( reviewed in [53] , [54] ) . In the following sections we discuss the sequence similarity of the Ae . albopictus Sfps with proteins ( including reported Sfps ) from other species , highlight potential functions of Sfps relevant to Ae . albopictus reproductive biology , and compare the evolutionary rates of the Sfps with those of housekeeping genes . To identify orthologs to the Ae . albopictus Sfps , we compared each Sfp to proteins in the full proteomes of seven species: the mosquitoes Ae . aegypti , Cx . quinquefasciatus , and An . gambiae , the pomace fly D . melanogaster , the honey bee A . mellifera , the mouse M . musculus , and humans ( H . sapiens ) . The summary of orthology information is presented in Table 2 . Of the 198 Ae . albopictus Sfps , 93 ( 47 . 0% ) have a putative ortholog in the full proteome of at least one of these species ( Table S1 ) . There is some conservation in protein sequences across these seven species . Specifically , 72 Ae . albopictus Sfps ( 36 . 4% ) are conserved in Ae . aegypti , 46 ( 23 . 2% ) are conserved in all three mosquito species , 43 ( 21 . 7% ) are conserved in Dipteran species , 37 ( 18 . 7% ) are conserved in insects , and 30 ( 15 . 2% ) are conserved in all seven species examined ( Table S1 ) . To determine whether these putative orthologs are known Sfps or sperm proteins in their respective species , we cross-referenced the putative orthologs to published lists ( if available ) of Sfps and sperm proteins . It is important to note that any relative percentages of orthology involving Sfps of other species should be interpreted cautiously . This is because the studies of Sfps in other species have used a variety of methods , some of which are more exhaustive than others . With that caveat , comparisons of Ae . albopictus Sfps with those reported in the other seven species demonstrate limited conservation in the complement of Sfps . Specifically , only 34 ( 17 . 2% ) of the Sfps from Ae . albopictus have a putative ortholog to a reported Sfp in at least one other species ( Table S1 ) . Of those , 18 have a putative ortholog that was reported as a Sfp in Ae . aegypti . This was more than the number of Sfp orthologs in any of the other species in our comparison ( Table 2 ) , and was expected based on species relatedness and similarity in methodology . Notably , only four of the Ae . albopictus Sfps have a putative Sfp ortholog in more than one species ( Table S1 ) . Specifically , three Sfps ( Aa-8246 , Aa-24416 , Aa-45626; see Table S1 ) have an ortholog to a Sfp in two other species . Information about the predicted protein classes to which two of them ( Aa-8246; Aa-24416; a putative angiotensin-converting enzyme and a putative serpin ) belong is included in Table 3; the third is a predicted aspartic/aspartate aminotransferase which , to our knowledge , have not been reported to have a direct role in post-mating responses or fertility . A fourth Sfp ( Aa-14624 , a predicted heat shock protein ) has an ortholog to a Sfp in three other species , and is discussed in Table 3 and in the text below . The 198 putative Sfps together have a variety of predicted functions ( Figure 1D ) , and likely play diverse roles that are relevant to the reproductive biology of Ae . albopictus . In Table 3 , we propose several possible reproductive and post-mating related roles of selected Ae . albopictus Sfp protein classes based on the demonstrated or predicted roles of putative orthologs or protein classes . In the following sections we limit our discussion to potential intriguing roles of selected Ae . albopictus Sfps in relation to two facets of reproductive biology of this species: processes affecting fertilization ( sperm protection and function ) and processes affecting fecundity ( egg development and feeding behavior ) . In a wide range of taxa , Sfps have been reported to evolve rapidly on average [102]–[104] , and to turn over quickly at the sequence level , with substantially different complements of Sfps in different species [22] , [23] , [105]–[110] . The latter trend appears to apply to Aedes mosquitoes as well - as described above , many of the Sfps identified in this study were found only in Ae . albopictus . Only 93 ( 47% ) of the 198 Ae . albopictus Sfps share orthology with proteins from the seven other species included in our comparisons . In order to assess whether Ae . albopictus Sfps also evolve rapidly at the sequence level , we estimated dN/dS for the 72 Sfps that have predicted orthologs in Ae . aegypti , recognizing that the most rapidly-evolving genes might not appear in this dataset as they might not have recognizable orthologs in Ae . aegypti . As a control , we used a set of 3498 transcripts ( encoding putative housekeeping products ) that were also identified in the current study ( see Methods ) . The results are presented in Figure 2 ( the raw data are available in Table S6 and Table S7 ) . In contrast to results from other taxa , there was no difference in the rate of evolution between the shared Sfps and the control genes in the two species ( Wilcoxon rank sum test W = 129390 , P = 0 . 680 ) . Moreover , none of the 72 Ae . albopictus Sfps that have orthologs in Ae . aegypti appear to have experienced positive selection ( w = dN/dS >1 ) in this pairwise comparison . For both the control and Sfp genes , divergence should be sufficiently low to permit accurate estimation of dN and dS for most loci ( median dS = 0 . 71 for both sets of genes; median dN = 0 . 020 for Sfps , 0 . 024 for control genes ) . Synonymous sites may approach saturation for a subset of genes , with the upper quartile of dS approaching 1 for both gene sets ( Sfps: 0 . 91; controls: 0 . 97 ) . However , the effect of synonymous site saturation should be to inflate estimates of w , and in this case we see very few cases where dN/dS >1 . It is interesting to note that among those 72 Ae . albopictus Sfps having orthologs in Ae . aegypti , the Sfps derived from transcripts found exclusively in males had a higher average dN/dS than those derived from transcripts found in both males and females ( Mann-Whitney U test U = 88 . 000 , P<0 . 001 ) . Inference of positive selection from pairwise sequence alignments is inherently conservative . Since most sites in most proteins are likely under strong constraint , average dN/dS will be much less than 1 , even if a few sites are subject to positive selection . Therefore , we used codeml ( part of the PAML package; [51] ) to infer site-specific selection on 46 Ae . albopictus Sfps for which orthologs were found in each of the other three mosquito species ( Ae . aegypti , Cx . quinquefasciatus , An . gambiae ) . Consistent with the pairwise inferences of dN/dS , little evidence of positive selection was found using this approach ( Table S8 ) . Even before correcting for multiple testing , only one SFP showed any evidence for positive selection on a subset of codons , and even then only in the less stringent M8 vs . M7 comparison ( Aa-18562; M8 vs . M7: P = 0 . 015 , M8 vs . M8a: P = 0 . 165 ) . While these four mosquito species are more distantly related than is typically used for the inference of positive selection , analyses using distant relatives can identify loci under selection [111] , [112] . For example , positive selection has also been inferred for genes present in distantly related mammalian species [112] , indicating that rapid sequence evolution can be inferred despite the requirement for orthology between distant relatives . We note that these codeml analyses are limited to genes that are relatively conserved , since they are found in all four mosquito species . Nonetheless , the contrast with other taxa is striking . In Drosophila , for example , Sfps were initially characterized in D . melanogaster and/or D . simulans , with putative orthologs subsequently identified in additional species . As such , this set of genes was restricted to genes with orthologs in multiple species of Drosophila . Nonetheless , positive selection was inferred for six out of twenty-five Drosophila Sfps ( at a false-discovery rate of 0 . 1 ) using codeml [109] . Thus , it appears that those Sfps that are detectable across mosquitoes are also constrained at the sequence level , recognizing that these mosquito Sfps are shared across genera , whereas the Drosophila Sfps are shared among species within that genus . Both pairwise analyses within the genus Aedes and multiple-species inferences indicate that Sfp conservation differs between mosquitoes and Drosophila , but it is unclear what biological features underlie this difference . Rates of Sfp evolution may be driven in part by co-evolution of male and female proteins in response to conflict between male and female reproductive interests . This conflict is expected to be higher in polyandrous species , such as Drosophila , than in species ( including these mosquitoes ) in which females usually mate with only one male . Sequence constraint of this subset of Sfps suggests that alterations to their sequence are disadvantageous . Therefore , identifying and interfering with the pathways of these Sfps may prove beneficial for vector control . We identified in Ae . albopictus 314 male-derived proteins that are transferred to females during mating . To create a reference for identifying these proteins , we developed transcriptome sequence datasets of Ae . albopictus male reproductive tissues ( seminal vesicles and accessory glands ) and the female lower reproductive tract . The 198 seminal fluid proteins we report here represent a wide variety of functions ( Figure 1D; Table 3; Table S1 ) , and likely play important roles in aspects of Ae . albopictus reproductive biology possibly including sperm protection , sperm-egg binding , and egg production . Ninety-three ( 47% ) of the Ae . albopictus Sfps we identified have putative orthologs to proteins in the full proteomes of other insects and mammals ( Table 2; Table S1 ) . However , only 34 ( 17 . 2% ) of the Ae . albopictus Sfps have putative orthologs to Sfps in other species ( Table 2; Table S1 ) . On one hand , this suggests rapid evolution of the composition of seminal fluid in these species , although this finding should be treated with caution as identification of Sfps are limited by the sensitivity of the techniques used . On the other hand , for those Ae . albopictus Sfps for which orthologs can be detected in other species , there is little indication of positive selection on the Sfps in pairwise or multi-species comparisons ( Figure 2 ) . Further population-level studies , as well as comparative studies using more closely related species , will help to clarify the extent to which Ae . albopictus Sfps undergo positive selection . This work contributes to our growing knowledge of Sfps in a diverse array of taxa , and establishes a foundation for several important lines of future research: First , this work sets the stage for investigating the roles of individual Ae . albopictus Sfps on female post-mating changes in physiology and behavior [15]–[17] , [80] . Much work in D . melanogaster , another Dipteran species , has elucidated the roles of specific Sfps on female post-mating behavior ( reviewed by [21] ) , and recent work on An . gambiae has identified one Sfp that regulates semen coagulation and sperm storage [19] . Given the important vector status of Ae . albopictus and the potential for further disease risk due to its rapidly expanding range [5] , elucidating the phenotypic effects of the Sfps in this species may assist researchers in identifying molecular targets for control [18]–[20] . Second , this work in combination with the identification of Sfps in Ae . aegypti [18] , [36] might assist in pinpointing the molecular basis for ecological patterns of cross-mating dynamics in these two species . Studies have revealed a consistent asymmetric pattern in which the receipt of Sfps from Ae . albopictus induces typical post-mating changes in female Ae . aegypti , but the receipt of Sfps from Ae . aegypti has little to no effect on female Ae . albopictus . This pattern has been suggested with respect to several post mating behaviors including host-seeking [113] , egg development and deposition [15] , [79] , [80] , [113] , and refractoriness to mating [114] . This asymmetry in Sfp cross-reactivity might create strong selective pressure for Ae . aegypti females to avoid mating with Ae . albopictus in regions where the two species coexist [115] . More generally , this work facilitates comparisons of Sfp components and functions across species . These comparisons can lead to the identification of sequences that are conserved and that promote male and female reproductive success . This work will also aid ongoing comparisons of Sfps across mosquito species that are vectors of disease pathogens , and ideally lead to the identification of novel targets for genetic-based control strategies that can be applied to multiple mosquito species [20] . Forty-six Ae . albopictus Sfps have putative orthologs in three other mosquito species ( Ae . aegypti , Cx . quinquefasciatus , and An . gambiae ) and showed no evidence for positive selection . These Sfps may be promising targets for use in genetic-based control strategies that modify and release male mosquitoes to induce reduced fertility or vector competence in their mates [116] . In conclusion , the work presented here provides a foundation for future investigations involving the molecular basis of multiple facets of Ae . albopictus biology , including reproductive biology , invasion ecology , hybridization and evolution , disease transmission dynamics , and vector control strategies . | The highly invasive Asian tiger mosquito ( Aedes albopictus ) transmits several pathogens that cause disease in humans and other animals . Therefore , Ae . albopictus poses a large and growing threat to public health across the world . One step toward managing the reproduction and threat of this species is to determine factors that influence its reproductive biology . Seminal fluid proteins ( Sfps ) are transferred from male mosquitoes to females during mating , and receipt of Sfps changes female reproductive behavior and physiology . Here we report the identity of 198 Ae . albopictus Sfps . We discuss the potential roles and impacts of these Sfps on reproduction . In addition , we compare Ae . albopictus Sfps with proteins ( including reported Sfps ) from other species , including two other important mosquito vectors of pathogens that cause human diseases . Our results provide a foundation for future studies to investigate the roles of individual Sfps on Ae . albopictus reproduction . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"and",
"Discussion"
]
| [
"spectrometric",
"identification",
"of",
"proteins",
"organismal",
"evolution",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"genome",
"evolution",
"animals",
"developmental",
"biology",
"animal",
"behavior",
"genome",
"analysis",
"molecular",
"genetics",
"zoology",
"fertilization",
"epidemiology",
"proteomics",
"insects",
"disease",
"vectors",
"behavioral",
"ecology",
"arthropoda",
"animal",
"physiology",
"biochemistry",
"mosquitoes",
"sperm-egg",
"interactions",
"proteomic",
"databases",
"ecology",
"entomology",
"vector",
"biology",
"gene",
"identification",
"and",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"evolutionary",
"biology",
"genomic",
"databases",
"organisms"
]
| 2014 | Identification and Characterization of Seminal Fluid Proteins in the Asian Tiger Mosquito, Aedes albopictus |
Neuronal degeneration is a hallmark of many DNA repair syndromes . Yet , how DNA damage causes neuronal degeneration and whether defects in different repair systems affect the brain differently is largely unknown . Here , we performed a systematic detailed analysis of neurodegenerative changes in mouse models deficient in nucleotide excision repair ( NER ) and transcription-coupled repair ( TCR ) , two partially overlapping DNA repair systems that remove helix-distorting and transcription-blocking lesions , respectively , and that are associated with the UV-sensitive syndromes xeroderma pigmentosum ( XP ) and Cockayne syndrome ( CS ) . TCR–deficient Csa−/− and Csb−/− CS mice showed activated microglia cells surrounding oligodendrocytes in regions with myelinated axons throughout the nervous system . This white matter microglia activation was not observed in NER–deficient Xpa−/− and Xpc−/− XP mice , but also occurred in XpdXPCS mice carrying a point mutation ( G602D ) in the Xpd gene that is associated with a combined XPCS disorder and causes a partial NER and TCR defect . The white matter abnormalities in TCR–deficient mice are compatible with focal dysmyelination in CS patients . Both TCR–deficient and NER–deficient mice showed no evidence for neuronal degeneration apart from p53 activation in sporadic ( Csa−/− , Csb−/− ) or highly sporadic ( Xpa−/− , Xpc−/− ) neurons and astrocytes . To examine to what extent overlap occurs between both repair systems , we generated TCR–deficient mice with selective inactivation of NER in postnatal neurons . These mice develop dramatic age-related cumulative neuronal loss indicating DNA damage substrate overlap and synergism between TCR and NER pathways in neurons , and they uncover the occurrence of spontaneous DNA injury that may trigger neuronal degeneration . We propose that , while Csa−/− and Csb−/− TCR–deficient mice represent powerful animal models to study the mechanisms underlying myelin abnormalities in CS , neuron-specific inactivation of NER in TCR–deficient mice represents a valuable model for the role of NER in neuronal maintenance and survival .
DNA is continuously damaged by spontaneous hydrolytic decay , endogenous metabolites ( e . g . reactive oxygen species , malondialdehyde ) , and environmental genotoxins . DNA lesions can give rise to irreversible mutations and chromosomal aberrations that may trigger carcinogenesis . Alternatively , DNA damage can cause replicative senescence and cell death , which promotes the process of aging [1] . Cumulative DNA damage has also been implicated in the functional deterioration and degeneration of long-living post-mitotic cells such as neurons [2] . To counteract the harmful effects of DNA injuries , cells have a variety of DNA surveillance and repair systems . The importance of these genome maintenance pathways for human health is well illustrated by a heterogeneous set of inherited syndromes that are associated with defects in specific DNA repair pathways resulting in cancer predisposition , developmental abnormalities , accelerated aging and neurodevelopmental or neurodegenerative abnormalities [1] , [3]–[5] . Nucleotide excision repair ( NER ) is a key DNA repair pathway for removal of UV-induced DNA damage and a wide range of other helix-distorting lesions , including bulky chemical adducts and specific types of oxidative damage [1] . In NER the DNA lesion is removed as a part of a 25–30 nucleotide single-strand fragment excised via a multi-step reaction followed by resynthesis of the excised strand [4] , [6]–[8] . NER can be divided into two subpathways that differ in the damage recognition step: While global genome NER ( GG-NER ) removes distorting DNA damage throughout the genome , transcription-coupled NER ( TC-NER ) specifically targets transcription-blocking lesions in the template strand of active genes to allow recovery of transcription after damage induction [1] , [4] , [7] , [8] . Several NER proteins have functions beyond NER , which is particularly evident for the transcription/repair factor TFIIH , which is required for the local opening of the damaged DNA in NER , but in addition plays an essential role in transcription . Furthermore , several lines of evidence indicate that TC-NER components are involved in repair of transcription-blocking lesions independent of the NER core complex , putatively via recruitment of other repair mechanisms . The term transcription-coupled repair ( TCR ) has been used to designate this broader , still poorly defined repair process [7] , [9] , [10] . NER gene defects are associated with a heterogeneous set of rare clinical syndromes , whose characteristics can be explained by the type of NER pathway that is affected or by defects in additional functions of these NER components in other DNA repair pathways or transcription . Selective defects in GG-NER , resulting from mutations in the XPC and XPE ( also termed UV-DDB2 ) genes encoding GG-NER-specific damage recognition proteins , cause xeroderma pigmentosum ( XP ) , a photosensitivity syndrome characterized by UV-hypersensitivity , pigmentation abnormalities and UV-induced skin cancer predisposition [11] , [12] . Cancer-predisposition in XP-C patients is explained by bulky lesions that accumulate over the entire genome causing mutations after replication [1] . Selective defects in TC-NER result from mutations in the genes encoding the proteins CSA or CSB , both of which are selectively recruited to stalled RNA polymerase II [13] . Mutations in CSB and CSA are associated with Cockayne syndrome ( CS ) , a progeroid disorder characterized by cachectic dwarfism and progressive neurological abnormalities , in addition to skin photosensitivity [14]–[16] . CS patients do not show cancer predisposition , which is explained by the normal function of GG-NER , and indicates that TC-NER is not required for preventing cancer . On the other hand , most CS pathological features cannot be explained by the sole loss of TC-NER function as they do not occur in XP-A patients , which show a combined GG-NER/TC-NER deficiency , resulting from mutations in the gene encoding for the core NER protein XPA . Thus XP-A patients present with UV-hypersensitivity and skin cancer predisposition , like XP-C patients , usually in combination with progressive neurological abnormalities ( see below ) [17]–[19] , but they do not develop cachectic dwarfism and other progeroid features of CS patients . This has led to the notion that the CS phenotype is largely the consequence of an overall TCR defect , i . e . , the inability to rescue transcription arrested by NER- and non-NER-types of DNA damage [7] , [20] , [21] . In addition it has been suggested that CS is associated with transcriptional abnormalities independent of DNA lesions [7] , [22] , [23] . The complementarity of NER and TCR DNA repair pathways and disorders resulting from deficiencies in these processes is also illustrated by XPCS patients , which display both XP and CS symptoms . XPCS is caused by mutations in the XPB or XPD genes , both encoding helicases of the transcription/repair factor TFIIH , or in the XPG gene , encoding the endonuclease that mediates the 3′ incision of the excision step [24]–[27] . Mutations in the XPB and XPD genes may also cause pure XP or trichothiodystrophy ( TTD ) , a disorder that is characterized by sulphur-deficient hair , in association with a variable spectrum of abnormalities that usually include neurodevelopmental deficits . Mutations that cause XP preferentially afflict the NER activity of TFIIH , while TTD mutations destabilize the TFIIH complex causing exhaustion of TFIIH in specific cell types [27] . The occurrence of CS symptoms in association with specific XPB and XPD mutations point to functions beyond NER and basal transcription presumably linked to non-NER TCR activities akin to CSA and CSB [22] , [27]–[31] . XPG mutations associated with XPCS have been proposed to destabilize the interaction between XPG and TFIIH , while mutations causing XP disrupt its endonuclease activity , further pointing to a non-NER activity underlying CS symptoms [25] , [26] , [32] . The presence of progressive juvenile or adult onset neurological abnormalities in XP-A patients has provided a strong hint that the NER pathway is important for neuronal function and maintenance [14] , [19] , [33]–[36] . The neurological symptoms are characterized by progressive sensory and motor deficits , as well as cognitive deterioration and emotional abnormalities , and are associated with widespread neuronal degeneration in multiple brain areas and the spinal cord [17]–[19] . XP-C patients ( who are only deficient in GG-NER ) do not develop overt neurological symptoms , indicating that the neurodegenerative changes follow from TC-NER or a combined GG-NER and TC-NER dysfunction . A dominant role of the TC-NER pathway in the nervous system is also suggested by the occurrence of XP-A-like progressive neurological abnormalities in CS patients . However , in CS patients neuropathological changes are primarily characterized by myelin abnormalities , while neurons and their axons seem relatively unaffected [14] , [35] , [37] , [38] . An additional complicating factor is formed by patients carrying CSA or CSB mutations that develop UV-sensitive syndrome , a disorder that is characterized by the skin abnormalities of CS in the absence of other CS features . The lack of typical CS features in these patients has been linked to residual TCR activities required for repair of oxidative DNA lesions , while TC-NER of UV-induced DNA lesions was deficient [39] . In sum , the data from XP and CS patients indicate that combined deficiency of GG-NER and TC-NER as in XP-A patients predominantly afflicts neurons , while deficiencies of TCR predominantly cause myelin problems . However , the precise mechanisms underlying the differential cellular vulnerabilities in XP and CS nervous system are still poorly defined , in particular in CS where distinct degenerative mechanisms may operate in oligodendrocytes and neurons [7] , [22] . Although previous studies have shown that mouse models for XP , CS , XP-CS and TTD reliably recapitulate the repair defect ( i . e . GG-NER and/or TC-NER/TCR ) , UV-sensitivity and skin cancer predisposition associated with the corresponding NER syndromes ( Table 1 and Table 2 ) , this does not apply to the neurological features [40] . In particular , Xpa−/− mice fail to exhibit obvious neurological symptoms and neuropathological changes observed in human XP-A [41]–[44] . Likewise , Csa−/− and Csb−/− mouse models for CS , except for photoreceptor-loss , do not show overt neurological abnormalities [45] , [46] . However , a detailed systematic neuropathological analysis is still lacking . In the present study we reexamined various NER and TCR mutant mouse models for neurodegenerative abnormalities to assess and dissect the contribution of the different repair systems in preventing neurodegeneration . To permit analysis of the direct contribution of DNA repair defects to neurological functioning , in the absence of pathology elsewhere in the body due to systemic DNA repair deficiency , we have generated a Cre-lox-based conditional Xpa mouse model that enables inactivation of the Xpa gene selectively in neurons . In particular , we used this novel mouse model to examine the effect of combined NER and TCR deficiency on neuronal survival .
As a first step to study the role of NER and TCR in maintaining neuronal integrity , we have re-examined six previously reported mutant mouse models for the presence of neuropathological abnormalities . These mouse lines consisted of Xpc−/− mice , in which only GG-NER is completely inactive [43] , [47]; Xpa−/− mice , in which both GG-NER and TC-NER are fully deficient [41] , [43]; Csa−/− and Csb−/− mice , in which TC-NER and presumably the entire TCR pathway is abrogated , but which have proficient GG-NER [45] , [46] , [48]; XpdXPCS mice ( homozygous for the G602D XPCS mutation in the Xpd gene ) , which carry a partial GG-NER and a partial TCR defect [49]; and XpdTTD mice carrying Xpd alleles with the R722W TTD mutation , that also have a partial GG-NER and TCR defect , in addition to TFIIH instability causing transcriptional insufficiency in terminally differentiated cells with consequent brittle hair and nails ( Table 1 and Table 2 ) [50] , [51] . Consistent with their respective NER-deficiencies the mutant mice manifest various degrees of increased susceptibility to UV- and 7 , 12-dimethylbenz[a]anthracene ( DMBA ) -induced skin carcinogenesis ( Table 1 and Table 2 ) . TCR-deficient mice to varying extent exhibit other symptoms like reduced growth , osteoporosis , photoreceptor loss , liver and kidney aging , and reduced lifespan ( Table 2 ) [20] , [28] , [44]–[56] . Consistent with previous reports [43]–[50] analysis of thionin- and hematoxylin/eosin-stained brain sections revealed that the gross anatomy and histological organization of all central nervous system regions of aforementioned mutant mice were indistinguishable from wild type animals at 6 months of age , precluding overt neurodevelopmental deficits or neuronal degeneration . Also the central nervous system of 70–100 week old Csb−/− and Xpa−/− mice appeared normal . Therefore , to examine the possible occurrence of subtle abnormalities , we employed immunohistological approaches . First , to determine whether central nervous system cells of the NER-deficient mice experience genotoxic stress , we studied the expression of the transcription factor p53 , which is activated by multiple types of DNA damage [57] . p53-immunoreactive cells were not detected in the central nervous system of wild-type and XpdTTD mice . Instead , occasional cells with p53-immunoreactive nuclei were observed throughout the nervous system of Csa−/− , Csb−/− and XpdXPCS mice , while even more sporadic p53 induction was visible in Xpa−/− and Xpc−/− animals ( Figure 1 ) . p53-staining was associated with neurons ( NeuN-positive cells; Figure 1A ) , astrocytes ( GFAP- or S100-positive cells; Figure 1C and Figure S1 ) , and sometimes oligodendrocytes ( APC-positive cells; Figure S1 ) . The relative amount of glial versus neuronal p53-staining varied per brain region , and to some extent per mouse model: in cerebellar cortex the large majority of p53-positive cells consisted of neurons , in the neocortex p53-positive cells were neurons or non-neuronal cells in equal amounts , while in spinal cord and the brainstem reticular formation p53-immunoreactivity was predominantly or almost exclusively associated with glia cells ( Figure 1E and Figure S1 ) . p53-immunoreactive neurons generally showed normal nuclear morphologies , in contrast to most p53-immunoreactive glial cells . For instance , a subset of p53-positive astrocytes in spinal cord and the brainstem reticular formation showed nuclei with a DAPI-negative centre that was intensely p53-positive ( Figure S1 ) . Another nuclear abnormality of p53-positive glia consisted of a larger nuclear size ( Figure S1 ) . Also in Xpa−/− and Xpc−/− nervous systems p53-positive cells consisted of both neurons and astrocytes , but their frequency was too low to allow systematic analysis of the relative proportion of neuronal versus non-neuronal cells in different brain areas . Taken together the data indicate that mice with a complete ( Csa−/− , Csb−/− ) or severe partial ( XpdXPCS ) TCR defect show nuclear p53 expression in sporadic neurons and glia throughout the nervous system , pointing to the occurrence of genotoxic stress . To further investigate the presence of degenerative changes in the nervous system of NER-deficient mice we examined microglia cells which proliferate and acquire activated morphologies in conditions of neuronal and glial damage [58] , [59] . Immunostaining for Iba-1 , a marker of all microglia cells and Mac2 ( also known as galectin-3 ) , a protein selectively expressed by activated phagocytosing microglia [58] , did not , or only sporadically , reveal activated microglia cells in the nervous system of wild-type , Xpa−/− , Xpc−/− and XpdTTD mice ( Figure 2 and Figure S2 ) . In contrast , prominent levels of activated microglia were present throughout the nervous system of Csa−/− , Csb−/− mice and , to a somewhat lesser extent , XpdXPCS mice ( Figure 2 , Figures S2 and S3 ) . Typically , Mac2-positive microglia cells occurred in small clusters in areas with myelinated fibers . Thus , in the forebrain Mac2-positive microglial cells were concentrated in the corpus callosum , the anterior commissure , the capsula interna and the fornix , while in the caudal brain high levels of activated microglia occurred in the cerebellar white matter , throughout the reticular formation and in fiber tracts in the brainstem ( Figure 2 , Figure S2 and Figure S3 ) . We performed a more in-dept analysis of the time of onset and course of these features in Csb−/− mice at different ages . Prominent levels of Mac2-positive cells were already present before 10 weeks of age ( not shown ) , and the density of Mac2-positive cells did not show a distinct increase with age ( Figure 2G ) . To examine whether microglia activation is paralleled by changes in astrocytes , we also stained for glial fibrillary acidic protein ( GFAP ) , an astrocytic protein up-regulated under conditions of neuronal injury . Increased GFAP staining was observed in the nervous system of mouse mutants that also showed microglial cell activation , i . e . Csa−/− , Csb−/− and XpdXPCS mice ( Figure 2I–2O ) . Increased GFAP staining was most prominent in the brainstem reticular formation and spinal cord ( Figure S4 ) . No obvious changes in GFAP staining were noted in some white matter areas such as the corpus callosum and the capsula interna of Csa−/− , Csb−/− and XpdXPCS mice , which may be explained by relatively higher baseline GFAP-immunoreactivity in these areas in wild-type mice . To further examine astrocytic changes in white matter areas , we examined the expression of Hsp25 ( also known as Hsp27 or Hspb1 ) , a small heat shock protein that is expressed at high levels in a subset of astrocytes in conditions of injury [60] . Indeed Csa−/− , Csb−/− and XpdXPCS , but not Xpa−/− , Xpc−/− and XpdTTD nervous systems showed the appearance of intensely Hsp25-immunoreactive astrocytes in multiple regions including the brainstem reticular formation , spinal cord , the white matter of cerebellum and the corpus callosum forebrain ( Figure S4 ) . The above data indicate that Csa−/− , Csb−/− and XpdXPCS mice show microglia and astrocyte activation in multiple central nervous system areas , indicative of the occurrence of cellular degeneration or another detrimental process . As these mutant mice also displayed p53-immunoreactive cells , we performed double labeling of p53 with Mac2 to determine whether p53-immunoreactive cells are contacted by phagocytosing microglia cells . However , Mac2-positive microglia cells were never found in the vicinity of p53 cells ( Figure 3A ) . Instead , Mac2-positive microglia were frequently in close proximity of oligodendrocytes ( Olig2 and APC-positive; Figure 3B , 3C ) , that otherwise showed a healthy appearance with normal DAPI-stained nuclei . These data indicate that glial abnormalities may be associated with subtle alterations in oligodendrocytes . To examine whether the presence of activated microglia was associated with myelin abnormalities , we performed double labeling of neurofilament-H and myelin basic protein to outline axons and their myelin sheets . No differences in myelin basic protein staining were observed in spinal white matter and corpus callosum of wild-type and Csb−/− mice ( not shown ) . Finally , to determine whether Csa−/− , Csb−/− and XpdXPCS mice show increased levels of cell death of oligodendrocytes or other cells , we stained for active caspase 3 , which is a final executioner caspase associated with multiple cell death pathways [61] . Very sporadically caspase 3 immunoreactive cells were observed in the nervous system of all mutant mouse models; all positive cells showing morphologies compatible with glial cells ( Figure 3D , 3E ) . Quantitative analysis in spinal cord indicated that the number of active caspase 3-positive cells , although still very low , was higher in Csa−/− , Csb−/− and XpdXPCS mice as compared to the other genotypes ( Figure 3F ) . It was not possible to determine whether the active caspase 3 cells represented oligodendrocytes or astrocytes because they did not stain for cellular markers such as GFAP , S100 , APC ad NeuN . Consistent with active caspase 3 staining , a silver degeneration procedure , that outlines degenerating neurons and their processes , indicated that none of the NER mice showed detectable levels of neuronal degeneration at 26 weeks of age . Taken together the data show that Csa−/− , Csb−/− and XpdXPCS mice develop prominent microglia cell activation as well as astrocytic changes that may be mostly triggered by subtle oligodendrocyte abnormalities . The NER mouse models investigated above exhibited either no detectable neuronal abnormalities or evidence for only subtle neuronal dysfunction and degeneration , indicating that inactivation of NER or TCR pathways by itself is not sufficient to significantly affect long-term survival of neurons in mice . Previous studies disclosed synergistic deleterious effects of intercrossing XP ( Xpa−/− or Xpc−/− ) with CS ( Csa−/− , Csb−/− , XpdXPCS ) mice , resulting in double mutants with very short life span and dramatic progeroid features [49] , [62]–[64] . This raises the possibility that neuronal degeneration may be achieved by inactivation of multiple NER components afflicting both NER and TCR pathways . However , the very short lifespan as well as the serious systemic abnormalities of the double mutant mice precludes systematic analysis of neuronal degeneration , which could also be an indirect consequence of impaired function of other organs and systems . To address this issue , we generated a Cre-lox-based conditional Xpa mouse model that enables selective inactivation of the Xpa gene in postnatal neurons of CS mouse lines and hence to study the effect of combined TCR and NER inactivation in neurons of adult mice that do not suffer from other severe deficits . To establish a conditional Xpa knockout mouse model , we generated a targeting construct in which exon 4 of the Xpa gene is fused in frame to the mouse Xpa cDNA containing the remaining coding sequence and including a synthetic polyA sequence , followed by a PGK promoter-driven hygromycin selectable marker gene , and a LacZ-GFP fusion gene ( Figure 4A ) . The splice acceptor-Murfi cassette ensures proper splicing and translational stops in all frames respectively when the Xpa gene is knocked out ( Figure 4A ) . The functionality of this conditional genomic-cDNA fusion allele was tested in UV-hypersensitive Xpa−/− ES cells ( Figure S5 ) . These experiments showed that the Xpac conditional allele fully averted the UV-hypersensitivity of Xpa−/− ES cells ( Figure S5 ) . Next Xpac/+ ES cells , obtained by transfection of IB10 ES cells ( Figure 4B ) , were used for blastocyst injections and subsequent generation of Xpac/+ mice . To determine whether Cre recombinase was capable of excising the floxed Xpa sequence in vivo , we generated Xpac/−/Cag-Cre mice by crossing Xpac/+ mice with Xpa+/− mice carrying a Cag-promotor driven Cre transgene ( Cag-Cre ) , which drives Cre recombinase expression immediately after conception [65] . Southern blot analysis showed Cre-recombinase excision of the floxed sequence in Xpac/−/Cag-Cre embryos at ∼100% efficiency ( Figure 4C ) . Consistent with ubiquitous recombination , Xpac/−/Cag-Cre embryos stained blue upon X-gal staining due to LacZ expression , while Xpac/− embryos remained unstained ( Figure 4D ) . Xpac/−/Cag-Cre mouse embryonic fibroblasts ( MEFs ) like Xpa−/− MEFs [41] showed severe UV-hypersensitivity , while Xpac/− MEFs show wild-type UV-resistance , consistent with Cre-dependent inactivation of the conditional allele ( Figure 4E ) . Next , we crossed Csb−/−/Xpac/+ and Csb−/−/Xpa+/−/Cag-Cre mice to obtain Csb−/−/Xpac/− and Csb−/−/Xpac/−/Cag-Cre mice . In line with the phenotype of Csb−/−/Xpa−/− mice [63] , Csb−/−/Xpac/−Cag-Cre pups displayed severe postnatal growth deficits , cachexia , disturbed gait , and death before weaning , while Csb−/−/Xpac/− littermates did not develop overt pathology ( Figure 4F ) . Taken together , these data demonstrate that we have generated a valid conditional Xpa mouse model that enables us to study the Csb−/−/Xpa−/− phenotype in a cell or tissue-specific manner . To study the effect of Xpa inactivation in the absence of Csb , specifically in postnatal neurons , we crossed Csb−/−/Xpac/− mice with a calcium/calmodulin-dependent protein kinase IIα ( CamKIIα ) Cre transgenic line that expresses Cre-recombinase selectively in postnatal neurons throughout the forebrain [66] , [67] . Forebrain-specific recombination was confirmed by PCR and analysis of LacZ expression . Csb−/−/Xpac/−/CamKIIα-Cre mice grew into young adulthood without any noticeable phenotype , showed a normal body weight and appearance at the age of 6 months , but from the age of 9–12 months became smaller and exhibited reduced weight as compared to littermates with other genotypes , i . e . CamKIIα-Cre , Csb−/−CamKIIα-Cre , Xpac/−CamKIIα-Cre and Csb−/−/Xpac/− littermates ( Figure 5A , Figure S6 ) . In addition , from 9–12 months of age , Csb−/−/Xpac/−/CamKII-Cre mice started to display seizure behavior , characterized by episodes of immobility . Subsequently , Csb−/−/Xpac/−/CamKII-Cre mice became moribund , all animals dying prematurely between the age of 12 and 22 months ( Figure 5B ) , while animals from littermates with other genotypes survived up to 24 months ( the oldest age examined ) . Analysis of locomotor behavior using the accelerating rotarod assay demonstrated that Csb−/−/Xpac/−/CamKIIα-Cre mice performed within the normal range at the age of 6 months , but showed reduced performance at 12 months ( Figure S6 ) . For further analysis of behavioral abnormalities , we used an open-field exploratory test . This test revealed that Csb−/−/Xpac/−/CamKIIα-Cre mice avoided exploration of the central part of the open field , which is considered a measure of anxiety-related behavior [68] . Total movement time and distance were the same as for the other groups ruling out impaired mobility as explanation for the difference in the test . The ratio of the ambulatory activity in the center and the total walking distance was already reduced at 3 months of age , and further declined at 6 and 12 months of age ( Figure 5C ) . Macroscopic examination of the brain of Csb−/−/Xpac/−/CamKIIα-Cre mice revealed no obvious changes at 3 months , mild atrophy of the cortex at 6 months , and severe atrophy of the cortex at older age ( Figure S6 ) . The sizes of olfactory bulbs , cerebellum and spinal cord were the same as in other groups . Analysis of coronal sections of 12–16 month-old Csb−/−/Xpac/−/CamKIIα-Cre brains showed a large reduction in cortical thickness , atrophy of other telencephalic areas ( i . e . hippocampus , caudatus-putamen and septum ) , and dramatically enlarged lateral ventricles ( Figure 5D , 5E ) . No abnormalities were observed in non-telencephalic areas , consistent with specific inactivation of the conditional Xpa allele in forebrain neurons . Atrophy of telencephalic areas was paralleled by a marked increase in GFAP immunoreactivity , while GFAP staining in other brain areas was the same as in Csb−/−/Xpac/− and Csb−/− mice ( Figure 5F ) . Atrophied brain areas also showed loss of the neuronal somato-dendritic marker microtubule-associated protein 2 ( MAP2 ) , in particular in the hippocampal CA1 region , indicative of neuronal degeneration ( Figure S7 ) . Staining for p53 revealed a prominent increase in the number of p53 immunoreactive neurons in the forebrain of Csb−/−/Xpac/−/CamKIIα-Cre mice as compared to Csb−/− mice and other genotypes that showed essentially no p53 immunoreactive cells ( Figure 6A , 6D ) . In addition , Csb−/−/Xpac/−/CamKIIα-Cre forebrain exhibited a strong increase in neurons expressing ATF3 ( Figure 6B , 6D ) , a stress-inducible transcription factor that is induced following genotoxic stress via p53-dependent and -independent pathways [69] , [70] . Finally , direct evidence for neuronal degeneration was obtained by staining for active caspase 3 and by using a silver staining procedure: active caspase 3 staining revealed intensely stained neuronal profiles ( Figure 6C , 6D ) . Similarly , the silver degeneration staining method outlined infrequent argyrophylic neuronal profiles , reflecting neurons that are in the process of dying . In addition , the silver staining uncovered high levels of argyrophilic axonal degeneration in the corpus callosum , the fimbria-fornix , the anterior commissure , and the cortifugal fiber bundles coursing in the capsula interna , the cerebral peduncle and the pyramidal tract ( Figure 6E ) , which is consistent with the selective occurrence of neuronal degeneration in forebrain neurons . Together these data indicate that Csb−/−/Xpac/−/CamKII-Cre mice display chronic neuronal degeneration that in the long term has resulted in severe neuronal loss and atrophy of the forebrain regions . The distribution of degenerative changes in Csb−/−/Xpac/−/CamKII-Cre mice is consistent with the specific inactivation of Xpa in forebrain neurons induced by CamKII-promotor driven Cre-recombinase expression [66] , [67] , and highlights the vulnerability of Csb-deficient forebrain neurons to loss of Xpa function . To determine the effect of Xpa inactivation in other neuronal populations of the Csb−/− brain , we crossed Csb−/−Xpac/− mice with a postnatal Purkinje cell specific Cre ( L7-Cre ) transgenic line [71] to obtain Csb−/−/Xpac/−/L7-Cre mice . Analysis of motor behavior with accelerating rotarod revealed no or very mild motor abnormalities in Csb−/−/Xpac/−/L7-Cre mice at the age of 3 and 6 months ( the oldest age examined ) . However , neuropathological analysis disclosed multiple signs of selective Purkinje cell degeneration resembling pathological changes in forebrain neurons of Csb−/−/Xpac/−/CamKII-Cre mice ( Figure 6 ) . Abnormalities included the presence of argyrophilic axonal degeneration , specifically in the cerebellar white matter and cerebellar nuclei , i . e . the areas that contain Purkinje cell axons ( Figure 6F ) , and sporadic argyrophilic debris in the molecular and Purkinje cell layer , while no argyrophilic changes occurred in other brain areas . In addition , the Purkinje and molecular layers also showed a strong increase in GFAP-immunoreactivity ( Figure 6G ) , while staining for calbindin , a protein that in the cerebellum is selectively expressed in Purkinje cells , revealed calbindin-negative regions in the molecular layer , indicative of loss of Purkinje cells ( Figure 6H ) . Furthermore , a subset of Purkinje cells ( with morphologies varying from relatively normal to severely atrophic cells ) displayed strong nuclear ATF3 staining ( Figure 6I ) , which was distinct from the non-specific cytoplasmic staining of Purkinje cells produced by the ATF3 antibody . Nuclear ATF3 staining was not observed in Purkinje cells ( nor other cerebellar neurons ) of wild-type , Csb−/− , Csb−/−/Xpac/+/L7Cre , Csb−/−/Xpac/− and Xpac/−/L7-Cre , as well as forebrain-specific Csb−/−/Xpac/−/CamKII-Cre mice . Quantification of ATF3-immunoreactive Purkinje cells in mid-sagittal sections of 6 month-old Csb−/−/Xpac/−/L7-Cre mice ( n = 3 ) indicated that 1 . 3±0 . 6% ( Mean ± SE ) of Purkinje cells were ATF3-positive . Finally , staining for active caspase 3 revealed infrequent ( <1 in 5000 ) positive Purkinje cells ( Figure 6J ) . In conclusion , the data obtained with Csb−/−/Xpac/−/L7-Cre mice further demonstrate that the addition of an Xpa defect to Csb-deficient neurons results in pronounced neuronal degeneration .
The abnormalities identified in the CS ( Csa−/− and Csb−/− ) , and XPCS ( XpdXPCS ) mice consisted of 1 ) the presence of activated phagocytosing microglia cells in regions containing myelinated axons such as the corpus callosum , the brainstem reticular formation and the spinal cord; and 2 ) sporadic cells with intense p53-immunoreactive nuclei . Microglia activation was frequently accompanied by signs of astrocytosis indicative of a central nervous tissue injury response . We did not find an association between microglia activation and p53-positive cells , and neither was microglia activation associated with detectable axonal degeneration . However , activated microglia cells were often in close contact with oligodendrocytes . These data indicate that microglia activation follows from oligodendrocyte or myelin abnormalities . Previous electron microscopic analysis did not reveal abnormalities in the morphology and thickness of myelin sheets in Csb−/− mice [45] , and in the current study , apart from evidence suggesting a minor increase in apoptosis of oligodendrocytes , we did not identify other overt oligodendrocytic abnormalities . Hence , the precise cellular abnormality that triggers microglia activation in myelinated regions of Csa−/− , Csb−/− , and XpdXPCS mice remains to be determined . Importantly , however , the presence of activated microglia is consistent with the notion that irregular patchy myelination with minimal axonal degeneration is a dominant neuropathological hallmark of CS [14] , [22] , [35] , [37] , [72] . Hence , our findings together with human neuropathological data strongly indicate that oligodendrocyte abnormalities are a prime defect in CS . We also show that XpdTTD mice , unlike XpdXPCS mice , do not develop microglia activation in myelinated areas . This is in line with the notion that myelin abnormalities in TTD patients and XpdTTD mice result from developmental deficits and arise via different mechanisms than in CS patients [14] , [22] . Our data further illustrate that specific point mutations in the Xpd gene result in different cellular deficits and associated pathologies in the mouse , mimicking the different pathologies in patients [27] , [31] , [49] . The second abnormality that we identified in the CS ( Csa−/− and Csb−/− ) and the XPCS ( XpdXPCS ) mouse nervous systems consisted of sporadically distributed p53-immunoreactive neurons and astrocytes , and , albeit very infrequent , oligodendrocytes . p53-immunoreactive cells occurred in all brain areas , but the proportion of neuronal versus glial p53-immunoreactive cells varied among brain areas . Thus , in cortex and cerebellum a large proportion of p53-positive cells are neurons while in the brain stem and spinal cord the far majority , if not all , p53-positive cells are glial cells . The expression of p53 , which is known to be activated by multiple types of DNA damage and which mediates neuronal degeneration [57] , provides indirect evidence for the occurrence of genotoxic stress , which can be explained by cumulative DNA damage resulting from compromised DNA repair . Interestingly , a subset of p53-positive astrocytes showed abnormal nuclear morphologies , which is compatible with reports of astrocytic nuclear abnormalities in CS patients [38] , [72] , [73] , and further indicate that astrocytes are vulnerable to loss of TCR function . We did not observe abnormal nuclear morphology in p53-immunoreactive neurons , nor did we obtain direct evidence for ongoing neuronal death using two neuropathological markers for dying neurons , i . e . active caspase 3 immunoreactivity , and silver degeneration staining . However , the process of death and removal of individual neurons may occur within a few hours , making the in vivo detection of asynchronous sporadically distributed cell death challenging [74] , [75] . Hence , our methods do not exclude the possibility of a low frequency of ongoing neuronal degeneration . The lack of an obvious neurodegenerative phenotype in the CS mouse models is compatible with the neuropathology of CS patients indicating relatively modest neuronal degeneration in most brain areas [35] , [38] , [72] , [73] . Interestingly , cerebellar granule cells , which are among the most severely affected populations of neurons in CS patients [76] , most frequently showed p53 immunoreactivity in the CS mice , indicative of a differential vulnerability of cerebellar granule cells to loss of TCR function in both CS patients and mouse models . Furthermore , p53-immunoreactive granule cells have been demonstrated in autopsy cases of CS [76] . Together our data indicate that Csa−/− , Csb−/− , and XpdXPCS mice reproduce the major aspects of CS neuropathology , albeit in a mild form , which may explain the absence of macroscopic neuropathological and obvious neurological deficits associated with CS . In this context it would be interesting to know whether patients with UV-sensitivity syndrome ( UVSS ) , also carrying mutations in CSA and CSB genes , develop the same mild abnormalities . The presence of activated microglia in UVSS patients would support the notion of a continuum of CS phenotypes ranging from CS type II to UVSS [16] as also suggested by a CS patient with a CSB null mutation displaying adult-onset neurological symptoms [37] . As the pathologies of Csa−/− , Csb−/− , and XpdXPCS mice are relatively similar , our data also indicate that the CS neurodegenerative features can not be explained by molecular mechanisms that do not include all three proteins . Furthermore , the data indicate that the CS neurodegenerative changes result from deficits in a shared non-NER activity of these proteins as Xpa−/− mice with complete loss of GG-NER and TC-NER function did not reproduce the neuropathological features that we observed in Csa−/− , Csb−/− , and XpdXPCS mice ( see below ) . This is consistent with a broader TCR process , which encompasses transcription-coupled repair of non-NER/non-distorting transcription-blocking lesions involving CS and TFIIH proteins . The GG-NER-defective Xpc−/− and total NER-defective Xpa−/− mutant mice at 26 weeks of age showed very low levels of p53-immunoreactive neurons and astrocytes , which nevertheless was higher than in XpdTTD and wild-type mice of the same age , in which we did not detect any cells with nuclear p53 immunoreactivity throughout the nervous system . These data suggest that Xpa−/− and Xpc−/− mice have a central nervous system phenotype , albeit marginal . In case of Xpc−/− mice the phenotype is compatible with that of XP-C patients that , although neurologically and cognitively asymptomatic , may develop mild neurodegenerative changes [19] . However , in Xpa−/− mice the phenotype is very different from the severe progressive neurodegenerative changes of many XP-A patients , which develop juvenile or adult progressive neuronal degeneration throughout the central and peripheral nervous system depending on the severity of NER dysfunction [17] , [19] , [33] , [34] , [36] . Neurons from Xpa−/− mice display considerably increased sensitivity to UV radiation [77] and the cross-linking agent cisplatin [78] , consistent with loss of NER function and excluding redundancy of NER activity by other proteins at least for the lesions induced by these agents . The discrepancies between human and rodents may follow from differences in the rate of production and type of DNA lesions caused by endogenous metabolites , and from the shorter lifespan of mice . To investigate the effect of combined NER and TCR-deficiency on neuronal survival , we generated a Cre-lox-based conditional Xpa mouse model to inactivate Xpa selectively in postnatal neurons in Csb−/− mice . The use of a conditional Xpa mouse model was required in view of our previous findings that global Csb−/−/Xpa−/− double mutant animals show degenerative changes in multiple organs as well as a very short life span [62] , [63] , precluding prolonged analysis of neurodegeneration and separation from direct and indirect consequences . Our data show that Csb−/− mice with neuron-specific inactivation of Xpa develop progressive neuronal degeneration , indicating that the XPA protein ( and the NER pathway as a whole ) is essential for the survival of mouse neurons in the absence of the CSB protein . The time course and distribution of neurodegenerative changes indicate that the affected neurons degenerate asynchronously over a prolonged time window . When Xpa is inactivated in forebrain neurons of Csb-deficient animals , mild behavioral abnormalities were observed at 3 months of age , while death , associated with severe atrophy of forebrain areas , occurred between 12–21 months of age . Analysis of the distribution of dying neurons , as identified by active caspase 3 or ATF3 staining , showed that the level of ongoing neuronal degeneration at a given time point was low . Similarly , in Csb-deficient mice with selective inactivation of Xpa in Purkinje cells which were analyzed at a single time point , a subset of Purkinje cells had disappeared ( identified as loss of calbindin staining ) , a very small subset was in the process degenerating or dying ( ATF3 and caspase 3 staining ) , while a subset showed a normal appearance consistent with asynchronous degeneration . Such an asynchronous neuronal degeneration is consistent with cell death resulting from the accumulation of stochastic DNA damage [79] , [80] , and strongly resembles the pattern of neuronal degeneration in Ercc1Δ/− mice that are impaired in several DNA repair systems , i . e . nucleotide excision repair , interstrand crosslink repair and double strand break repair [81] . Together , the data with conditional Xpa/Csb-deficient mice indicate that adult neurons in rodents are vulnerable to endogenous DNA lesions when deficient in both NER and TCR , but are able to cope with these lesions when either the TCR or NER pathway are defective . While the NER and TCR pathway share the TC-NER activity , they have non-overlapping activities consisting of GG-NER and the still poorly defined non-NER TCR activities . In neurons , factors of the GG-NER machinery , in particular XPC , have been shown to operate in a specialized type of transcription-associated repair , termed domain-associated repair ( DAR ) . DAR operates on both strands in active genes , including regions of a gene that RNA polymerase II does not reach and has been proposed to complement TCR [80] , and it may possibly mask or compensate for the loss of TCR [80] . This is supported by the demonstration that XPC-deficient mice that are selectively deficient in GG-NER when crossed with CSB-deficient mice have a similar phenotype as Csb−/−Xpa−/− mice [64] . Our data indicate that Xpc−/− and Xpa−/− animals develop similar marginal central nervous system phenotypes consisting of highly sporadic p53-positive cells . Together these data suggest that in mice Xpc is equally important as Xpa for the central nervous system . Also in man , XPC-deficiency may result in subtle neurodegenerative changes [19] , although XPA-deficiencies results in much more severe neurodegenerative phenotypes [14] , [17] , [19] . Non-NER TCR has been proposed to operate in conditions of specific transcription-blocking oxidative DNA lesions , putatively via recruitment of alternative DNA repair pathways [7] , [9] , [10] . This explains why cells from CS mice and CS patients , unlike XPA-deficient cells , show increased vulnerability to some types of oxidative stress , and may more readily accumulate oxidative DNA lesions [9] , [20] , [39] , [82] . In addition , increased levels of oxidative DNA lesions have been reported in brain tissue of Csb−/− mice [10] . Thus , the inability to cope with oxidative lesions may explain the pathological phenotype of CS mice , as well as the severe degenerative phenotype in the conditional Xpa-deficient Csb−/− mice . However , the precise identity of DNA lesions and the question whether the CS phenotype truly results from a repair deficiency remains to be further explored . In summary , our data indicate that the GG-NER , TC-NER , and non-NER TCR mechanisms operate together in maintaining the integrity of neurons , and that the absence of one pathway aggravates the risk for deficiencies in other pathways , explaining the severe neurodegenerative phenotype in double mutants . The extent to which a combined deficiency of NER and TCR is detrimental to non-neuronal nervous systems cells remains to be determined in future studies by selectively inactivating Xpa in these cells in TCR deficient mice . We propose that neuron-specific inactivation of Xpa- in Csb-deficient mice represents a powerful model for studying XP neurological disease and the role of NER in neurons . As neurologic symptoms seen in XP are hallmark features of age-related neurodegenerative diseases these mice may also reproduce aspects of accelerated aging .
Experiments were performed in accordance with the “Principles of laboratory animal care” ( NIH publication no . 86-23 ) and the guidelines approved by the Erasmus University animal care committee . Animals used were Xpc−/− , knock-out for the Xpc gene [47] , Xpa−/− , knock-out for the Xpa gene [41] , Csa−/− , knock-out for the Csa gene [46] , Csb−/− , in which the CS1AN patient mutation is mimicked resulting in a null mouse [45] , XpdXPCS , homozygous for the G602D XPCS point mutation in the Xpd gene [49] and XpdTTD , carrying Xpd alleles with the R722W TTD mutation [50] bred in a pure C57BL/6J background . To obtain a conditional Xpa knockout mouse model , we generated a targeting construct in which exon 4 was fused in frame to the mouse Xpa cDNA ( containing the remaining coding sequence and including a synthetic polyA sequence ) , followed by a PGK promoter-driven hygromycin selectable marker gene ( Figure 4A ) . A genomic clone containing 10 kb of the 129ola mouse Xpa locus ( pMMXP3-6#13; [41] ) , was used to re-clone an approximately 10 kb size BamHI fragment , containing exon 3 to 6 , into the psp72 vector . Following XbaI digestion , part of exon 4 and intron 4 was replaced by a cassette containing the mouse Xpa cDNA including the natural 3′ UTR and polyadenylation signal followed by a PGK promoter-driven hygromycin selectable marker and a LoxP site respectively . Next , the SmaI site downstream of the LoxP site was used to introduce a cassette containing a splice acceptor sequence ( SA ) , an ochre stopcodon multiple reading frame insertion ( Murfi ) linker , a ribosomal entry site ( IRES ) , and a LacZ/GFP fusion reporter gene ( as a blunted SalI fragment ) . The SmaI site in intron 3 was used to insert a blunted XhoI-SalI loxP fragment from pGEM30 ( kindly provided by Dr . W . Gu , University of Cologne ) . This targeting construct , which was designated pIP-Xpa-con , contains homologous arms of 4 kb at the 5′ end and 5 kb at the 3′ end . The 129Ola-derived ES cell line IB10 was electroporated with NotI linearized pIP-Xpa-con DNA and cultured in gelatin-coated dishes as described before [45] . Hygromycin ( Roche , 843555 ) was added 24 hr after electroporation to a final concentration of 100 µg/ml . Cells were maintained under selection for 7–8 days , after which clones were isolated and expanded in 24-well plates . Genomic DNA from individual hygromycin-resistant clones was digested with EcoRI and analyzed by Southern blotting using a 500 bp DraI fragment ( “intron 5/exon 6” probe; obtained from a 7 . 5 kb PCR fragment spanning exon 5 and 6 ) . EcoRI digested DNA from targeted ES clones was subsequently screened with the hygromycin ( cDNA ) probe to confirm proper homologous recombination at the 5′ end of the targeting construct . For the generation of Xpac/− ES cells , we followed the same procedure as described above , except that Xpa−/− ES cells [54] were used . To test the functionality of the loxP sites , Xpac/+ ES cells were electroporated with a purCre plasmid ( kindly provided by Dr . M . Jaegle , Erasmus MC ) and cultured on gelatin dishes as described . Puromycin ( Sigma , P7255 ) was added 24 hr after electroporation to a final concentration of 100 µg/ml . Cells were maintained under selection for 3 days . Genomic DNA from individual puromycin-resistant clones was digested with EcoRV and analyzed by Southern blotting using a 500 bp PCR fragment of the LacZ gene . Properly targeted IB10 ES clones were karyotyped and cells from two independent clones ( selected for the presence of 40 chromosomes ) were injected into 3 . 5-day-old C57BL/6J blastocysts . Male chimeric mice were mated with C57BL/6J females to obtain heterozygote offspring . Heterozygous males and females were bred to Xpa+/− as well as Csb−/+ animals to ultimately obtain Xpac/− , Xpac/+ and Csb−/−/Xpac/− animals . Genotyping was initially performed by Southern blot analysis of genomic DNA obtained from tail biopsies of 8–10-day-old born pups . A description of PCR-based genotyping methods is given below . Xpac/− and Csb−/−/Xpac/− animals were also interbred with Cag-Cre [65] , CamKII-Cre ( line L7ag#13 ) [66] , [67] , and L7-Cre ( line L7Cre-2 , [71] ) Cre-recombinase transgenic mice , which were kindly provided by A . de Wit ( ErasmusMC ) , S . Zeitlin ( Columbia University ) , and J . J . Barski ( Max-Planck-Institute of Neurobiology , Martinsried , Germany ) , respectively . Primary mouse embryonic fibroblasts from the various single and double mutant mouse models ( three independent lines per genotype ) were isolated from day 13 . 5 embryos and cultured as described before [83] . Mice and cells were genotyped by PCR for the wild-type and ( conditional ) mutant Xpa or Csb alleles using a primer mix that ( per genotype ) amplifies both the wild-type and targeted alleles in a single reaction [41] . The presence or absence of the conditional Xpa allele was detected by PCR using primers XpaFex3 ( 5′-TTT GAT CTG CCA ACG TGT G-3′ ) and XpaRex4 ( 5′-GCT TCG CTT CTG TCT TGG T-3′ ) . The presence or absence of the Cre transgene was detected by PCR using primers 5′-GCA CGT TCA CCG GCA TCA AC-3′ and 5′-CGA TGC AAC GAG TGA TGA GGT TC-3′ . Both products were amplified with the same PCR program: 5 min . 93°C , 1 min . 93°C , 1 min . 58°C , 2 . 5 min . 72°C ( 35 cycles of the latter three steps ) , 5 min . 72°C . Cells or embryos were fixed for 30 minutes at 4°C in a buffer containing 1% paraformaldehyde , and subsequently washed 3×15 minutes with PBS/0 . 01% NP40 . Cells or embryos were stained overnight at 37°C in dark in a staining solution containing 3 . 1 mM K3Fe ( CN ) 6 , 3 . 1 mM K4Fe ( CN ) 6 , 0 . 15 M NaCl , 1 mM MgCl2 and 1 mg/ml X-gal ( Roche Applied Sciences , USA , IN ) . For tissues , the same procedure was used , except that the fixation time was extended to 3 hours . Seeded cultures at a density of 1000 spontaneously immortalized MEFs on a 6 cm dish were exposed to different doses of UV-C ( 254 nm , Philips TUV lamp ) . The cells were allowed to grow for another 7 days after which the resulting clones were fixed , stained and counted . For each independent cell line , the amount of surviving clones at each dose of UV , 3 dishes per dose , was calculated as the percentage of clones on the plate without UV . For the open field test , animals were placed for 30 min in a square ( 26×26×26 cm ) open field box , equipped with photobeam sensors ( TruScan E63 10–12 , Coulbourn Instruments ) , and attached to a computer to record the following ambulatory parameters: total distance , center distance , total move time , center time and corner time . Each test session lasted 30 minutes , and data were collected in 5 minute intervals . The anxiety ratio was calculated by dividing center distance by total distance . Rotarod analyses were performed as described previously [84] . Mice were anesthetized with pentobarbital and perfused transcardially with 4% paraformaldehyde , and brains were dissected out , weighed , and postfixed overnight in 4% paraformaldehyde at 4°C . For standard histological analyses brains were paraffin-embedded , sectioned at 4 µm and stained with haematoxylin/eosin solution . For other staining procedures brain specimen were embedded in gelatin blocks [81] and sectioned at 40 µm with a freezing microtome . Sections were processed , free floating , using immunofluorescence or a standard avidin-biotin–immunoperoxidase complex method ( ABC; Vector Laboratories ) with diaminobenzidine ( 0 . 05% ) as the chromogen . In addition , a selected number of frozen sections were processed for a silver staining procedure that selectively labels dying neurons and their processes [81] . Immunoperoxidase-stained sections were analyzed and photographed using a Leica ( Nussloch , Germany ) DM-RB microscope and a Leica DC300 digital camera . Sections stained for immunofluorescence were analyzed with a Zeiss ( Oberkochen , Germany ) LSM 510 confocal laser scanning microscope using 40x/1 . 3 and 63x/1 . 4 oil-immersion objectives . Primary antibodies reported in this study are as follows: mouse anti-APC ( Calbiochem , clone CC-1 , 1∶2000 ) ; rabbit anti-activating transcription factor 3 ( ATF3; Santa Cruz Biotechnology , Santa Cruz , 1∶1000 ) ; rabbit anti-cleaved caspase 3 ( Asp175; Cell Signaling Technology , 1∶200 ) ; mouse anti-calbindin ( Sigma , clone CB-955 , 1∶10000 ) ; rabbit anti-GFAP ( DAKO , 1∶5000 ) ; mouse anti-GFAP ( Sigma , clone G-A-5 , 1∶20000 ) ; rabbit anti-HSP25 ( Stressgen , 1∶7000 ) ; rabbit anti-Iba1 ( WAKO Chemicals , 1∶ 2000 ) ; rat anti-Mac2 ( Cedarlane , 1∶2000 ) ; mouse anti-MAP2 ( Millipore , clone AP20 , 1∶200 ) ; rat anti-myelin basic protein ( Millipore , MAB386 , 1∶500 ) ; mouse anti-NeuN ( Millipore MAB377 , 1∶2000 ) ; rabbit anti-neurofilament-H ( Millipore , 1∶2000 ) ; rabbit anti-olig2 ( IBL , 1∶2000 ) ; rabbit anti-p53 ( Leica , 1∶2000 ) ; mouse anti-S100B ( Sigma , clone 1B2 , 1∶2000 ) ; and guinea pig anti-VGLUT1 ( Millipore , 1∶2000 ) . For avidin-biotin–peroxidase immunocytochemistry biotinylated secondary antibodies from Vector Laboratories ( Burlingame , CA ) diluted 1∶200 were used . FITC- , cyanine 3 ( Cy3 ) - , and Cy5-conjugated secondary antibodies raised in donkey ( Jackson ImmunoResearch , West Grove , PA ) diluted at 1∶200 were used for immunofluorescence . Immunoperoxidase-stained sections were analyzed and photographed using a Leica DM-RB microscope and a Leica DC300 digital camera . To determine the relative staining intensity of GFAP staining , sections were photographed using a 5× objective , and optical densities were determined from TIFF files using MetaMorph 4 . 6 image analysis software . Optical densities determined in rectangular areas of 200×250 µm . To minimize variability resulting from the staining procedure this analysis was performed with sections stained in a single immunostaining session . Statistical analyses were done with GraphPad Prism software ( San Diego , USA ) . Means from different age groups , and different transgenic mouse lines were compared using one-way-ANOVA with Tukey's post tests . | Metabolism produces reactive oxygen species that damage our DNA and other cellular components , and as such it contributes to the aging process , including neuronal degeneration . Accordingly , genetic disorders associated with impaired DNA damage repair are frequently associated with premature onset of aging pathology in a variety of tissues , including the brain . This is well-illustrated by the progeroid DNA repair syndromes xeroderma pigmentosum ( XP ) and Cockayne syndrome ( CS ) , in which patients suffer from defects in nucleotide excision repair ( NER ) and transcription-coupled repair ( TCR ) , two partially overlapping DNA repair systems that remove helix-distorting and transcription-blocking lesions , respectively . We have used a panel of XP and CS mice ( including conditional double-mutant animals ) to systematically investigate the impact of NER and TCR defects on neuronal degeneration . We have shown that , whereas a TCR defect causes white matter pathology , a NER defect can result in age related cumulative loss of neurons . These findings well match the neuropathology observed in CS and XP patients , underscoring the impact of spontaneous DNA damage in the onset of neuronal aging . Therefore , the XP and CS mouse models serve as valuable tools to delineate intervention strategies that combat age-associated pathology of the brain . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"neurological",
"disorders",
"neurology",
"biology",
"neuroscience"
]
| 2011 | Age-Related Neuronal Degeneration: Complementary Roles of Nucleotide Excision Repair and Transcription-Coupled Repair in Preventing Neuropathology |
Chagas Disease , caused by the protozoan Trypanosoma cruzi , is a major health and economic problem in Latin America for which no vaccine or appropriate drugs for large-scale public health interventions are yet available . Accurate diagnosis is essential for the early identification and follow up of vector-borne cases and to prevent transmission of the disease by way of blood transfusions and organ transplantation . Diagnosis is routinely performed using serological methods , some of which require the production of parasite lysates , parasite antigenic fractions or purified recombinant antigens . Although available serological tests give satisfactory results , the production of reliable reagents remains laborious and expensive . Short peptides spanning linear B-cell epitopes have proven ideal serodiagnostic reagents in a wide range of diseases . Recently , we have conducted a large-scale screening of T . cruzi linear B-cell epitopes using high-density peptide chips , leading to the identification of several hundred novel sequence signatures associated to chronic Chagas Disease . Here , we performed a serological assessment of 27 selected epitopes and of their use in a novel multipeptide-based diagnostic method . A combination of 7 of these peptides were finally evaluated in ELISA format against a panel of 199 sera samples ( Chagas-positive and negative , including sera from Leishmaniasis-positive subjects ) . The multipeptide formulation displayed a high diagnostic performance , with a sensitivity of 96 . 3% and a specificity of 99 . 15% . Therefore , the use of synthetic peptides as diagnostic tools are an attractive alternative in Chagas’ disease diagnosis .
Chagas disease is a major health and economic problem in Latin America , for which no vaccine or appropriate drugs for large-scale public health interventions are yet available [1] . It is caused by the protozoan parasite Trypanosoma cruzi , found throughout the Americas in a variety of wild and domestic mammalian reservoirs , and it is usually transmitted by infected blood-sucking triatomine bugs . It is estimated that ~5 . 7 million people are currently infected with T . cruzi and that up to 120 million individuals living in endemic areas in Latin America are at risk of infection [2] . Chagas Disease remains the most important parasitic disease in the Western Hemisphere , with an estimated disease burden , as measured by disability-adjusted life-years , that is 7 . 5 times as great as that of malaria [2] . Increasing travel and immigration have also brought the risk of T . cruzi infection into non endemic countries [3] . Several efforts have successfully been undertaken to control transmission in Latin America , with a concomitant decrease in the number of acute vector-borne infections [4] . However , humans can also become infected with T . cruzi through the ingestion of tainted food and fluids , receipt of contaminated blood transfusion or organ transplantation , laboratory accidents , and from mother-to-child during pregnancy/delivery [1 , 4] . The diagnosis of Chagas disease is challenging because it is often asymptomatic in its acute phase and evolves into a chronic stage in which the disease presents diverse clinical forms [1] . In addition , and due to a major decline in parasitemia during the chronic phase , the detection of T . cruzi in blood samples by direct examination , hemoculture , or xenodiagnosis is difficult and time-consuming . Several PCR-based procedures have been reported that , although highly specific , present suboptimal sensitivity and require technological expertise and specialized expensive laboratory equipment [5] . In this framework , detection of anti-T . cruzi antibodies remains the most effective method for demonstrating direct exposure to the parasite [6] . At present , the most widely used serologic methods are indirect hemagglutination assay ( IHA ) , indirect immuno-fluorescence assay ( IIF ) , and enzyme-linked immunosorbent assay ( ELISA ) using total parasite homogenates or semipurified antigenic fractions [7] . Despite their satisfactory performance , these tests show variations in their reproducibility and reliability that can be attributed to poor standardization of the reagents or intrinsic variability of immune responses in patient populations [8–10] . In the absence of a single reference test showing 100% specificity and sensitivity , the current guidelines developed by the World Health Organization advise the use of two serologic tests for reaching a conclusive diagnosis . In the case of ambiguous or discordant results , diagnosis using a third technique should be conducted . In addition , there are other still unmet needs and gaps such as access to diagnostics in point-of-care sites for neglected populations [11 , 12] , as well as development of much needed tests for early identification of congenital transmission; rapid assessment of drug treatment efficacy or prognostics tests for disease progression [10 , 13] . Recombinant DNA and peptide synthesis technologies historically allowed the production and one-step purification of large amounts of T . cruzi immunodominant antigens [14] . However , several studies showed that the use of single antigens in an assay did not confer the sensitivity required for a diagnostic test [14 , 15] , which prompted the development of tests based on combinations of antigens[16 , 17] , some of which were evaluated in multicenter trials and are commercially available [18–20] . Synthetic peptides are advantageous for diagnostic applications because they are: i ) well defined ( ease of quality control ) , ii ) easily produced in large amounts , ii ) highly pure and often cost-saving if compared to the production of natural or recombinant antigens in vitro [21]; and iv ) also chemically stable ( can be stored lyophilized or dessicated and tend to be stable for several years ) . Short synthetic peptides spanning linear B-cell epitopes can also be used in serodiagnostic applications to increase specificity ( that is , decrease the number of false positives ) by replacing the use of whole protein antigens , therefore avoiding the display of unnecessary sequences that may lead to ‘false positive’ results . Specificity is a critical issue in serodiagnosis of Chagas Disease , where most reagents present cross-reactivity against other co-endemic parasites such as Leishmania spp . [18 , 21] . Peptide sensitivity , on the other hand can be increased using more densely presented immunoreactive epitopes ( i . e . by creating a synthetic poly-epitopic molecule ) or by combining multiple antigenic peptides in a single multiplex-assay [21–23] . A number of studies described the use of short peptides , containing either one or several epitopes for diagnosis of Chagas disease and other infectious diseases [23–34] . Recently , we have prioritized a number of candidate diagnostic targets from the genome of T . cruzi [35] and conducted a large-scale screening of parasite B-cell linear epitopes using high-density peptide microarrays [36] . This approach led to the identification of several hundred novel epitopes associated to chronic Chagas Disease , from which we selected 30 for further characterization . In this paper , we describe their diagnostic evaluation in ELISA format using a large panel of serum samples . In addition , and following an in silico-guided antigen combination strategy , we developed a proof-of-principle diagnostic kit based on these reactive peptides .
More than 2 , 000 candidate serodiagnostic peptides were previously identified by our group using a T . cruzi/Chagas HD peptide microarray [36] . To guide the selection of a subset of peptides for further serological characterization , a filtering strategy was conducted , as follows . First , peptides with serodiagnostic potential ( high signal-to-noise ratio in the microarray experiments ) were mapped to 187 distinct antigenic protein regions ( stretches of adjacent peptides in a protein sequence ) . These antigenic regions may contain either a single B-cell linear epitope or , in some cases , a limited number of partially overlapping epitopes [37] . Next , antigenic regions were grouped into clusters of sequence-related peptides , in such a way that peptide sequences sharing stretches of 7 or more identical amino acids were put into the same cluster . We reasoned that peptides within a cluster may be both sequence and also likely antigenically related , whereas peptides from different clusters may likely represent the targets of different antibody specificities . From each cluster only a single antigenic region was kept ( the one with highest microarray average seroreactivity ) . After this filter 95 unique antigenic regions were obtained ( non-redundant , non-similar ) . From this set we selected 30 peptides from the top of the ranking for further characterization ( the most reactive 15-mer from each antigenic region was selected ) . To minimize possible bias in our selection , the number of selected peptides from overrepresented sequences such as those from the mucin-associated surface protein ( MASP ) family [38] and from previously known antigens with mapped epitopes [24 , 39–43] was limited to 3 and 4 , respectively . Sequence and features of our final set of synthetic peptides is summarized in Table 1 . Peptides in Table 1 were synthesized and used in ELISA assays as described below ( see also Results ) to screen for reactivity against Chagas positive and negative ( control ) samples . Once we obtained a first matrix of reactivity of peptides vs individual serum samples , we applied the EpiSelect algorithm to guide the selection of sets of peptides for the formulation of multiepitope assays . Implementation of the algorithm has been described [47] , but briefly the algorithm aims to find the smallest selection of peptides ( epitopes ) that in concert maximizes the coverage ( reactivity ) against a given set of subjects . The input to the algorithm was the matrix of peptide reactivity values determined by ELISA , encoded as z-scores defined as the number of standard deviations above background . Positive peptides were defined using a z-score threshold of 3 . Synthetic peptides were purchased from Schafer-N ( Copenhagen , Denmark ) . Peptides were synthesized using standard FMOC chemistry , purified by HPLC ( > 90% purity ) and characterized by mass spectroscopy . A C-terminal cysteine residue was included in all peptides for conjugation to maleimide-activated BSA . An additional amino acid residue ( leucine ) was added at the N-terminus of peptide p1 , to avoid the partial deamination associated with an N-terminal glutamine [48] . Lyophilized peptides were resuspended in sterile-filtered water ( Sigma Product w3500 ) , and conjugated to maleimide-activated BSA ( mBSA , Sigma-Aldrich Product B7542 ) according to the manufacturer’s protocol , using a molar ratio of 35:1 peptide to mBSA [49] . Peptide-mBSA conjugates were stored in 50% glycerol at -20°C until use . Peptides that failed to solubilize under these conditions were discarded for the analysis . Human serum samples from T . cruzi-infected patients used in this study were obtained from the Laboratorio de Enfermedad de Chagas , Hospital de Niños "Dr . Ricardo Gutierrez" ( HNRG , Buenos Aires , Argentina ) ( n = 80 ) . Human serum samples from patients with American Tegumentary Leishmaniasis ( ATL ) used in this study were obtained from the Instituto de Patología Experimental , Universidad Nacional de Salta ( IPE , Salta , Argentina ) ( n = 19 ) . All procedures were approved by the research and teaching committee and the bioethics committee of both institutions , and followed the Declaration of Helsinki Principles . Written informed consent was obtained from all individuals ( or from their legal representatives ) , and all samples were decoded and de-identified before they were provided for research purposes . Chagasic patients were in the asympomatic chronic stage of the disease without cardiac or gastrointestinal compromise ( age range: 11 to 51 years old , median age: 20 ) . Serum samples were collected from clotted blood obtained by venipuncture and analyzed for T . cruzi-specific antibodies with the following commercially available kits: ELISA using total parasite homogenate ( Wiener lab , Argentina ) and IHA ( Polychaco , Buenos Aires , Argentina ) . ATL patients were diagnosed using a combination of techniques: direct observation of parasites ( amastigotes ) on smears of dermal scrapings; a test of delayed-type hypersensitivity ( Montenegro skin test ) ; and a clinical assessment ( see [50] ) . The negative panel was composed of samples from healthy , non-infected individuals that gave negative results in the aforementioned tests , and were obtained either from the blood bank “Fundación Hemocentro Buenos Aires” ( FHBA Buenos Aires , Argentina ) ( n = 82 ) or from IPE ( n = 18 ) . Samples from FHBA were also negative for HIV , Hepatitis B , Hepatitis C , HTLV I and II , Treponema pallidum ( syphilis ) and for Brucelosis ( Huddlesson test ) . To calculate the minimum sample size required to estimate sensitivity or specificity for a specified interval of confidence and precision under a normal approximation , we used the following formula: n=Z2P^ ( 1−P^ ) d2 Where Z is the z-score from a standard normal distribution ( e . g . 1 . 96 for a 95% confidence interval ) , P^ is the pre-determined ( guess ) value of sensitivity ( or specificity ) based on previous experience/judgment , and d is the required precision [51] . Therefore , for Z = 1 , 96 ( 95% CI ) , P^ = 0 . 99 , and d = 0 . 05 ( 5% error ) , the estimated sample size is 73 . Therefore 73 is the minimum number of Chagas positive samples ( to estimate sensitivity ) and Chagas negative samples ( to estimate specificity ) . Microplates containing 96 or 384-wells ( Thermo Scientific ImmunoPlates , MaxiSorp ) were coated overnight at 4°C with 100 ng/well of peptide-mBSA or with different peptide mixtures ( 80 ng/well of each one ) in PBS pH 7 . 4 . Blank signal was determined using mBSA-coated wells . After 4 washings with TBS-T ( 50 mM Tris-HCl ( pH 7 . 6 ) , 150 mM NaCl , 0 . 05% ( v/v ) Tween20 ) , the plates were blocked for 1 h at room temperature with 100 μl/well of assay buffer ( 3% ( w/v ) skimmed milk in TBS-T ) . The plates were washed and incubated for 1 h with human sera diluted as indicated ( 1:100 or 1:10 ) in assay buffer at room temperature . Optimization of the assay conditions was performed by a checkerboard titration analysis using 10 ng or 80 ng of peptide-mBSA , and different dilutions of secondary antibody ( peroxidase-conjugated goat anti-human IgG antibodies ( Sigma-Aldrich , St Louis , MO ) ( 1:5 , 000; 1:10 , 000; 1:20 , 000 and 1:80 , 000 ) . After washings , 100 μl of secondary antibody diluted as indicated ( 1:10 , 000 for assays using a single peptide per well , or 1:80 , 000 for multiepitope assays ) in assay buffer were added to each well and incubated for 1 h at room temperature . Following additional washings with TBS-T , the reaction was developed with tetramethylbenzidine for 15 min ( TMB , Sigma-Aldrich , St Louis , MO ) and stopped by addition of 0 . 2 M sulphuric acid . Absorbance values were measured at 450 nm in a microplate absorbance reader ( FilterMax F5 Multimode , Molecular Devices , Sunnyvale , CA , USA ) . All serum samples were tested in duplicate . Values were averaged and blank-corrected . The same 16 serum samples from healthy blood donors were tested in each ELISA plate . The cut-off value was determined for each peptide and for each plate using the mean of the control blood donor samples plus 3 SD ( the cut-off was set accounting for multiple-hypothesis testing ) . For each peptide or peptide mixture , standardized reactivity scores ( z-scores ) and the diagnostic analytical characteristics of sensitivity , specificity and AUC ( Area under the ROC–Receiver Operating Characteristic–curve , as a performance metric ) were calculated . Reagent sensitivity was calculated as the number of positive subjects ( i . e . infected patients samples that were reactive against a particular peptide ) over the total number of infected subjects tested; specificity was calculated as the number of negative subjects ( non-infected control subjects that were seronegative against a particular peptide ) over the total number of non-infected control subjects tested and AUC was calculated using the from the z-scores of infected subjects and non-infected subjects . For receiver operating characteristic ( ROC ) analyses [52] , the results were expressed as the percentage of reactivity of the mean absorbance at 450 nm of the positive reference control serum included in each assay run . The Mann-Whitney test and ROC analysis were performed using the GraphPad Prism software ( version 6 for OSX; San Diego , CA , USA ) or ROCR R package [53] .
Based on our previous screening of serodiagnostic peptides for Chagas Disease using HD peptide microarrays [36] , 30 peptides were selected for further serological characterization and downstream validation . The strategy for selection of these peptides is outlined in Fig 1 ( see also Methods ) , and essentially was guided to select a non-redundant set of peptides showing the highest antibody-binding signal in any array . After removing 3 peptides that showed solubility problems , the remaining 27 peptides were coupled to a carrier protein ( mBSA ) and assayed in ELISA format against a sera panel of 62 chronically infected Chagasic patients and 16 healthy controls . Initially , all human sera were tested at 1:100 dilutions . The panel of peptides included 16 peptides corresponding to previously uncharacterized T . cruzi proteins ( novel antigens ) that emerged during our screening [36] , 7 peptides representing novel epitopes in previously characterized B-cell antigens and 4 peptides corresponding to previously known linear B-cell epitopes , which were used as positive controls ( see Table 1 and S1 Fig ) . We also included in our panel an additional peptide ( p17 ) as an internal negative control . Although belonging to a validated T . cruzi antigen [54] , this peptide was derived from a protein region that showed consistently very low signal in all microarray replicates . Diagnostic sensitivity , specificity and AUC values for each peptide are shown in Table 2 ( complete data available in S1 Table ) . The diversity of reactivities in the collection of sera samples when assayed against individual peptides is also evident when visualizing the data in the form of a heatmap plot ( available in S2 Fig ) . As shown , promising diagnostic performances were observed for most of the assayed peptides . Sensitivity values ranged from 30–92% ( >50% in 22 out of 27 ) , and specificity values were extremely high , which is consistent with our screening strategy [36] . In this context , it is worth noting that sensitivity values of all individual T . cruzi antigens described so far and proposed and/or included in serodiagnostic tests ranged from 80–99% [14] . Overall , and as previously reported for the TSSA antigen [37] , a strong correlation between assays in the standard ELISA format and in microarray format was observed for each peptide ( Table 1 ) , thus providing additional validation and support for the use of HD-peptide arrays for discovery of new serology-based biomarkers . We further evaluated the diagnostic specificity of the 16 best performing peptides ( see Table 1 ) by using an extended panel of 61 control sera obtained from healthy subjects ( Chagas-negative samples ) . As before , individual peptides coupled to mBSA were assayed in ELISA format . Diagnostic specificities and ROC-AUC were re-calculated for each peptide ( top entries in Table 2 ) . The average specificity was 97 . 23% and in all cases specificities > 95% were observed . Notably , most of the positive responses observed in this expanded set of Chagas-negative samples correspond to only 3 of the 61 sera samples tested . These samples ( also negative for the highly-sensitive trans-sialidase inhibition assay [55] ) were highly reactive against more than half of the peptides ( 12 , 11 and 9 peptides each , see S1 Table in the ‘Additional negative sera’ section ) , suggesting a broad and yet-to-be explained cross-recognition towards T . cruzi-derived sequences . If these Chagas-negative serum samples were removed , specificity values of our peptides would increase up to an average 98 . 5% . Based on the results described above , we undertook an in silico-guided approach to design a multiplex assay with improved diagnostic performance . Using ELISA data from individual peptides , we applied the EpiSelect algorithm [47] ( see Methods ) to identify several optimal ( minimal ) virtual peptide sets that in concert provided maximal coverage of the analyzed subjects . This analysis was performed after removing data from the 9 serum samples that were previously used in microarray experiments , to avoid optimistically biased results . The analysis performed on the tested peptides and 53 Chagas-positive subjects showed that 3 peptides were enough to reach a theoretical sensitivity of 100% ( Fig 2 ) . Data used for this analysis is available in S1 Table . The optimal set was composed by peptides {pc1 , pc2 , and p6} , resulting in an average of 2 . 51 reactive peptides per subject , closely followed by the peptide set {pc2 , p11 , and p6} with an average of 2 . 43 reactive peptides per subject . The reactivity patterns for these sets are shown in Fig 2 and S1 Table . Interestingly , at least 1 of the 3 novel peptides p6 ( as in Fig 2 ) , p2 or p8 ( alternatives ) would be required to achieve a sensitivity of 100% with a 100% specificity ( see also S1 Table ) . Other peptides such as p5 , p7 , p11 , p12 , p16 , p19 and p24 also displayed excellent diagnostic characteristics , with individual high sensitivity ( > 70% ) and specificity ( up to 95% ) . Hence , these peptides can be eventually incorporated into the multiplex design to increase its robustness ( for example , to increase the number of reactive peptides per subject ) . Based on these analyses , we prepared and tested a number of multi-epitope peptide combinations in ELISA format against an extended panel of sera from chagasic ( positive ) and healthy ( negative ) subjects . One such combination {pc1 , pc2 , pc3 , p6 , p13} , was tested against 22 positive and 24 negative serum samples and gave a diagnostic sensitivity of 72 . 7% and a specificity of 91 . 7% . Following the same methodology ( S1 Table ) , we tested a slightly different formulation of peptides ( pc1 , pc2 , p6 , p7 and p24 ) against an increased number of sera samples ( 53 Chagas-positive and 31 Chagas-negative ) obtaining an improved performance , with a sensitivity of 92 . 45% and a specificity of 93 . 55% . Finally , with the aim of obtaining a peptide combination with enhanced robustness , we re-analyzed the reactivity profile of each individual serum sample ( S1 Table ) against our panel of peptides , and identified a few Chagas positive subjects that gave low or even negative reactivity to many peptides . From this analysis , we identified peptides that would theoretically maximize the sensitivity of the multiplex assay , despite not showing the best possible coverage of our subject ( sera ) collection . Thus , we arrived at a high performance multi-epitope formulation of seven peptides {pc1 , pc2 , pc3 , p6 , p7 , p13 , and p24} . To validate this final formulation , we increased the amount of coated peptide to 80 ng of each peptide per well and the serum concentration to 1:10 . After these modifications , the performance of this formulation , when tested against 82 Chagas-positive and 80 Chagas-negative sera samples gave a sensitivity of 96 . 34% and a specificity of 100% , with an AUC value of 0 . 9974 ( Fig 3 ) . We have also assessed the performance of this multiepitope formulation against a panel of 19 sera from subjects with positive diagnosis for American Tegumentary Leishmaniasis ( see Methods ) , and another 18 negative ( control sera ) from the same endemic region . Only a single ( negative ) subject gave a positive response in our multiepitope assay ( Fig 3C ) . Except for this case , the observed absorbance in the ELISA assays was nil . The specificity of the multiepitope formulation for this panel was 97 . 30% , with an overall specificity ( considering all negative samples from all panels ) of 99 . 15% . Table 3 summarizes the performance of this combination of peptides . This therefore represents a highly promising novel multiepitope formulation for the diagnosis of Chagas Disease .
Serological diagnostics methods for infectious diseases have usually evolved from first-generation lysate-based reagents . Through time , more defined formulations of diagnostic reagents have followed . Second-generation diagnostic kits based on purified antigenic fractions or third-generation kits based on recombinant proteins are now in widespread use . To develop new diagnostic tools that are simple and have few manipulation steps , one of the central aspects that currently limits the suitability of diagnostic kits is the need to produce , prepare and purify the antigens , along with the corresponding quality control . Short synthetic peptides can be produced cheaply in large quantities , and are chemically stable and amenable for long-term storage . Synthetic peptides have been already tested in a wide range of diagnostic applications and proved valuable for diagnosis of viral , bacterial , parasitic and autoimmune diseases [21 , 30–34] . Therefore , fourth-generation diagnostic kits based on well-defined peptidic antigens are now within reach . Here we present a next-generation diagnostic formulation for Chagas Disease based on short peptides . Significant efforts have been invested by various groups over time to identify and test antigenic peptides for serodiagnosis of Chagas Disease , some of which displayed promising analytical characteristics . For example , peptides Ag2/B13/Pep2 , TcD/Ag13 , TcE and TcLo1 . 2 , have been combined to create a multi-epitope recombinant neo-protein of excellent performance [24] , and peptides from the cytoplasmic repetitive antigen ( CRA ) /Ag30 and flagellar repetitive antigen ( FRA ) /Ag1 [54] have been recently shown to present good specificity and sensitivity [56] . The advent of novel high-throughput approaches spawned by the post-genomic era is starting to impact on the discovery of new biomarkers and the development of diagnostic tools for a number of important pathogens [10] . We have recently showed the utility of a fast approach to screen for new T . cruzi antigens that is based on high-density peptide microarrays [36] . The advantage of this platform is that it allows to identify antigens and at the same time obtain a fine mapping of their linear epitopes . Using this strategy we have identified and mapped the epitopes of >90 novel T . Cruzi antigens [36] . As a followup of this first screening for peptidic antigens , we provide here an extensive serological characterization of 27 peptides , 18 of which represent novel epitopes that were mapped using our strategy , or represent recently discovered antigens but for which no fine epitope mapping was yet available ( see Table 1 ) . For example , even though the trans-sialidase/SAPA antigen ( accession number X57235 , TcCLB . 509495 . 30 is the most similar genome locus tag ) has been known for quite some time , peptide p13 ( also annotated as ‘trans-sialidase’ ) is not derived from the originally described antigen , but from another member of the superfamily ( TcCLB . 506961 . 25 ) with only 29% identity to the original trans-sialidase/SAPA . Therefore , p13 is a new/novel antigen and epitope that bear no resemblance to any of the mapped epitopes already described [43 , 57] . Similarly , even though the proteins encoded by the genes TcCLB . 511633 . 79 ( microtubule-associated protein ) , or TcCLB . 506391 . 30 ( EF-hand protein 5 ) were already described and used as antigens [20 , 46] , this is the first time that their fine mapped epitopes are tested for diagnostic purposes . Other peptides such as p16 , p7 , p11 and p19 are part of proteins that have been identified as potential antigens [35] but with no other serological evidence before our microarray experiments . Peptide p1 , on the other hand , was derived from a member of the Mucin-Associated Surface Protein ( MASP ) family [38] , which is a large family of genes which were shown recently to be the target of the adaptive immune response in an animal model of infection [58] . The MASP protein encoded by gene TcCLB . 507071 . 20 was selected from the genome , as part of an effort to obtain a detailed characterization of the antigenicity and epitopes of this gene family in human infections [59] . Peptide p6 contains a slightly different version of the sequence TTRAPSRLREID , which has been identified as the major and conserved linear B-cell epitope included within the otherwise highly polymorphic TcMUCII family of T . cruzi proteins [44 , 60] . Whereas peptide p2 is a novel epitope from a putative 60S ribosomal protein L7a , that we have also previously identified as a potential antigen [35] . Using a panel of Chagas-positive and negative ( control ) samples , we performed a thorough serological characterization of the selected peptides . This allowed us to obtain a relatively large matrix of ELISA responses for all peptides against individual serum samples . This led us to identify a number of peptides with promising diagnostic potential , such as peptides p1 , p7 , p11 , p16 and p19 , which presented sensitivities above 80% , with no false positive responses in the first evaluation using a small panel of 16 sera , and only a few false positive responses ( with specificities from 96 . 5% to 100% ) in a second evaluation using a larger panel of sera . These sensitivities are similar to those originally reported in the first characterizations of validated serodiagnostic antigens such as TcD ( 95% for chronic subjects [61] ) and SAPA ( 10% for chronic subjects , 90% for acute infection [62] ) , which were later improved when developed into a multiantigen diagnostic reagent ( e . g . the Chagatest kit of Wiener Labs that includes these antigens claims a sensitivity of 98 . 8%[63] ) . Hence , even if some peptides displayed sensitivities that were not very high when assessed singly , they were high enough as to keep them under consideration for development of an assay based on combinations of peptides . The matrix of ELISA responses was then used to guide the rational formulation of a multiepitope diagnostic reagent using a well-defined algorithm for the inclusion of peptides . The first combinations tested did not achieve a significantly high performance , even if the theoretical prediction ( Fig 2 ) would suggest otherwise . One reason for this is that even though the input to the EpiSelect algorithm included the level of response of each subject against each peptide ( represented as the number of standard deviations above negative controls ) , the effect of combining peptides produced a higher background signal that was not predicted by the algorithm . Another reason was the inclusion in our panel of Chagas-positive sera of several samples with moderately low antibody titers overall ( see for example the 9 sera grouped in the bottom branch in S2 Fig ) . Despite these pitfalls , the detailed data present in this matrix was pivotal in identifying peptides for inclusion in the final multiepitope formulation . The rationale for inclusion of peptides was the ability of a given peptide ( as observed in the matrix ) to potentially overcome a negative response for a given serum sample . For example , peptides p6 and p2 , followed by p11 represented an optimal complement of the two best performing peptides , pc1 ( from the antigenic repeat of the CA-2/B13 antigen Ag2 ) and pc2 ( the serodiagnostic epitope TcE ) for diagnosis . Also , peptide p13 when combined with peptides pc1 and pc2 was one of the few peptides that provided relatively high signal in the ELISA assay against the group of sera with relatively low overall responses . The fact that we could consistently increase the performance of each combination upon following this rationale shows the usefulness of this approach . Interestingly , all peptides in the final multiepitope formulation are highly conserved ( see S1 Text ) . A sequence similarity search across available complete genomes ( e . g . those from the CL-Brener [64] and Sylvio X10 [65] strains using BLASTP ) or from draft assemblies ( Tula cl2 , Esmeraldo cl3 , Dm28c or JRcl4 in the TriTrypDB resource [66] , release 30 from February 2017 , using TBLASTN ) shows that all peptides are highly conserved across strains representing different evolutionary lineages of the parasite ( TcI , TcII , TcV , TcVI ) . The observed diagnostic performances for all peptides and peptide combinations tested were very promising , particularly considering that all assays were based on short synthetic peptides . Our final best performing multi-epitope combination was based on a combination of seven antigenic peptides . With an equimolar mixing of peptides , we attained a very high ( >96% ) level of sensitivity and specificity . These are highly promising values for a first optimization attempt; the final ELISA assay/formulation could be indeed further improved using different blocking reagents , coupled detection system and , most importantly , by adjusting the relative concentration of different peptides in the final mixture . Analysis of potential cross-reactivity with other co-endemic diseases and pathogens is essential to validate any diagnostic reagent . In the case of Chagas Disease , cross-reactivity against infections with Leishmania species is a particular concern [67] . We have included a panel of serum samples from confirmed cases of tegumentary leishmaniasis from the northern province of Salta , Argentina to assess the performance of our formulation . This also gave us the opportunity to improve the assessment of specificity by analizing a paired set of negative ( control ) samples ( chagas-negative and leishmaniasis-negative ) from the same endemic region . From a set of 37 of these samples which were negative for Chagas Disease , only one gave a positive cross-reactive response ( Fig 3 ) . Although this highlights the need to perform a more extensive characterization of this cross-reactive sample ( e . g . against our complete panel of peptides ) , and eventually revise the combination of peptides in our formulation , the current multiepitope assay has a sufficiently high specificity at this stage ( 99 . 15% ) , comparable to other commercially available kits [63] that can certainly be improved by optimization of the assay or by replacing of cross-reactive peptides . Besides the obvious attention to the diagnostic performance of the identified peptides , these results serve to validate the use of high-density peptide microarrays as a fast screening platform . The fact that all selected peptides gave positive responses against several Chagas-positive subjects show that this technology can be trusted to rapidly identify and map epitopes of complex pathogens . It is also worth mentioning here that there are about a hundred additional antigenic regions within the signal range observed in the peptide microarray screening from which these peptides were identified [36] and that await further serological characterization . This observation , together with the fact that the microarray screening only covered ~3% of the parasite proteome , show that there is still a large repertoire of Chagas-specific antibody specificities that remain serologically unexplored . The results presented herein hence provide a novel , robust multi-epitope formulation as a basis for the development of improved peptide-based serodiagnostics for Chagas Disease . In contrast with chimeric DNA constructs that encode multiepitope recombinant proteins , the fact that this diagnostic reagent is based on the combination of short peptides that can be synthesized separately and easily formulated in a mix-and-match approach , means that it can be improved successively over time with only a reasonable effort . | Chagas disease , caused by the parasite Trypanosoma cruzi , is a life-long and debilitating illness of major significance throughout Latin America , and an emergent threat to global public health . Diagnostic tests are key tools to support disease surveillance , and to ultimately help stop transmission of the parasite . However currently available diagnostic methods have several limitations . Identification of novel biomarkers with improved diagnostic characteristics is a main priority . Recently , we conducted a large-scale screening looking for new T . cruzi antigens using short peptides displayed on a solid support at high-density . This led to the identification of several hundred novel antigenic epitopes . In this work we validated the serodiagnostic performance of 27 of these against an extended panel of human serum samples . Based on this analysis , we developed a proof-of-principle multiplex diagnostic kit by combining different validated reactive peptides . Overall , our data support the applicability of high-density peptide microarrays for the rapid identification and mapping epitopes that could be readily translated into novel and useful tools for diagnosis of Chagas disease . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"biotechnology",
"synthetic",
"biotechnology",
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"immune",
"physiology",
"engineering",
"and",
"technology",
"synthetic",
"biology",
"immunology",
"tropical",
"diseases",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"protozoans",
"neglected",
"tropical",
"diseases",
"bioassays",
"and",
"physiological",
"analysis",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"proteins",
"antigens",
"immunoassays",
"proteomics",
"protozoan",
"infections",
"microarrays",
"trypanosoma",
"cruzi",
"biochemistry",
"trypanosoma",
"chagas",
"disease",
"eukaryota",
"diagnostic",
"medicine",
"peptides",
"synthetic",
"peptides",
"physiology",
"biology",
"and",
"life",
"sciences",
"organisms"
]
| 2017 | Next-generation ELISA diagnostic assay for Chagas Disease based on the combination of short peptidic epitopes |
Elucidation of new biomarkers and potential drug targets from high-throughput profiling data is a challenging task due to a limited number of available biological samples and questionable reproducibility of differential changes in cross-dataset comparisons . In this paper we propose a novel computational approach for drug and biomarkers discovery using comprehensive analysis of multiple expression profiling datasets . The new method relies on aggregation of individual profiling experiments combined with leave-one-dataset-out validation approach . Aggregated datasets were studied using Sub-Network Enrichment Analysis algorithm ( SNEA ) to find consistent statistically significant key regulators within the global literature-extracted expression regulation network . These regulators were linked to the consistent differentially expressed genes . We have applied our approach to several publicly available human muscle gene expression profiling datasets related to Duchenne muscular dystrophy ( DMD ) . In order to detect both enhanced and repressed processes we considered up- and down-regulated genes separately . Applying the proposed approach to the regulators search we discovered the disturbance in the activity of several muscle-related transcription factors ( e . g . MYOG and MYOD1 ) , regulators of inflammation , regeneration , and fibrosis . Almost all SNEA-derived regulators of down-regulated genes ( e . g . AMPK , TORC2 , PPARGC1A ) correspond to a single common pathway important for fast-to-slow twitch fiber type transition . We hypothesize that this process can affect the severity of DMD symptoms , making corresponding regulators and downstream genes valuable candidates for being potential drug targets and exploratory biomarkers .
Microarray-based expression profiling is a widely used , quick and inexpensive method to obtain information about the specific diseases . A traditional approach when searching for drug targets or candidate biomarkers for a specific disease is to look for genes differentially expressed between the disease and appropriate “control” samples . Various techniques have been applied to find statistically significant differentially expressed genes , including classical statistical tests ( e . g . t-test ) and those specifically developed for microarray data analysis ( Limma [1] , SAM [2] , shrinkage T-statistic [3] and other ) . To get the deeper understanding of the disease mechanisms , the functional analysis of differential genes can be performed using a number of different methods [4] . Typically they rely on Gene Ontology ( GO ) – based annotation of genes . Common approach is to pre-select differentially expressed genes based on differential fold-change and/or p-value threshold , and find the statistically enriched GO groups using Fisher's exact test . More sensitive approaches are based on gene set enrichment analysis ( GSEA [5] , [6] ) to avoid differential cut-off selection issue . In addition to Gene Ontology , the protein-protein functional associations , regulatory or biochemical networks can also be used as a source of functional protein annotation in enrichment analysis [6] , [7] , [8] . More elaborated classification and functional annotation methods [9] , [10] are usually applied to protein-protein networks only . The potential drawback of this kind of networks for the analysis of expression data is that they eventually skip the important transcriptional factors if they are not differentially expressed themselves . In this paper we used a proprietary literature-derived gene expression regulation network as a source of functional protein annotation . This global expression network consists of direct or indirect effects of a network node ( protein ) on expression of other genes [11] . Unlike conventional GSEA [5] , [6] , which uses predefined collection of gene sets , Sub-Network Enrichment Analysis ( SNEA ) algorithm , implemented in Pathway Studio® software [11] , constructs comprehensive collection of gene sets from ResNet , a global literature-extracted protein-protein regulation network . The gene sets are constructed for each individual network node ( “seed” ) and consist of all its downstream expression targets only ( star-like subnetworks ) . The central idea of SNEA approach is that if the downstream expression targets of a “seed” are enriched with differentially expressed genes , then the “seed” is likely to be one of the key regulators of the differential expression changes , e . g . a transcription factor responsible for the observed changes in expression or an upstream member of signaling pathway [12] . This literature-driven approach connects differentially expressed genes to major implicated pathways and key expression regulators . In contrast to other methods that utilize the same idea of finding upstream network regulators using expression data [13] , [14] , SNEA allows identification of any potentially important protein ( not obligatory a transcriptional factor ) leading to the observed expression changes , even if its own expression doesn't change . It becomes possible because of the usage of ResNet database where all relations are taken from the literature only . Hence , there is no restriction on the protein type that can be considered as potential “seed” , provided that it is reported to influence each individual downstream gene expression . We have applied this approach to study Duchenne muscular dystrophy ( DMD ) using publicly available gene expression profile datasets and identified a set of potential regulators and downstream biomarkers of DMD progression and severity . Duchenne muscular dystrophy is an X-linked recessive muscular disorder , caused by mutations in the dystrophin gene ( DMD ) [15]–[17] . Affecting about 1∶3500 newborn males , it is the most common form of muscular dystrophies and the most common sex linked disease in males [18] . The underlying genetic cause of DMD is the presence of a variety of DMD gene mutations that result in dystrophin reduction/absence in skeletal muscle [17] . Lack of dystrophin has multiple unfavorable consequences to a muscle fiber ( reviewed in [19] ) , leading to apoptosis or necrosis with subsequent inflammation and fibrosis at the site of damage . The process of muscle regeneration is also activated , but , in humans , with the course of the disease the repair capacity declines and becomes insufficient [20] . Muscle tissue is replaced with adipose and fibrous connective tissue [21] . The average life expectancy of DMD patients varies from late teens to early thirties , and can be improved by respiratory support [22] , [23] and drug therapy [24] . Currently , there is no cure for DMD , but some treatments targeting the secondary consequences of dystrophin deficiency , such as muscle damage , necrosis , apoptosis and failure of regeneration , are already available for patients . Glucocorticoids , such as prednisone and deflazacort , are widely used to alleviate some of the disease's symptoms [25] . Several tests are used in diagnostics of DMD , including measurement of physical parameters , serum level of creatine kinase , genetic testing for DMD mutations and muscle biopsy to confirm the reduction in dystrophin content . More accurate , preferably non-invasive and biologically explainable markers are needed to predict prognosis , estimate disease's severity and progression . Also new biomarkers are required in treatment and clinical trials for DMD , where they can be used to monitor drug efficiency and choose optimal drug dose . In order to identify potential drug targets along with corresponding biomarkers , we have searched for the consistent SNEA regulators and their downstream expression targets using publicly available differential gene expression profiles and literature-extracted expression regulation network from muscle biopsies of patients with DMD . Suggested workflow implies aggregation of the data from multiple datasets and elucidation of common mechanisms that underlie differential expression . Studying these mechanisms from the prospective of searching for new drug targets can provide valuable insights in both biological and medical research .
The overall analysis workflow is presented in Figure 1 . Five NCBI GEO DMD-related microarray expression profiles from muscle biopsies were aggregated according to the procedure described in Methods . To ensure robustness of our analysis we constructed five leave-one-out datasets each time aggregating four distinct experiments and omitting one out of total five available experiments . We also constructed single large dataset ( referred to as “aggregated dataset” ) , where all five available microarray experiments were aggregated . Additional dataset ( referred to as “reference dataset” ) was constructed on the base of published meta-analysis [26] , see Methods . We performed SNEA with default parameters for each of the six datasets ( five leave-one-out datasets plus aggregated dataset ) and obtained six lists of 100 significant regulators . Regulators common for all six datasets were combined with regulators obtained by SNEA of reference dataset . This resulted in the list of 76 unique regulators , which can be viewed as potential drug targets . We also performed permutation test to ensure that this overlap is significant . Next , we turned to selection of differentially expressed genes . For each of the 6 datasets ( five leave-one-out datasets plus aggregated dataset ) we performed gene ranking using combination of different methods ( see Methods section ) . Then we identified genes which were present in top-500 lists for all six datasets . Out of all these consistently differentially changed genes , we have selected only those which were expression targets of selected consistent significant regulators . This produced a list of 140 candidate genes ( 105 over- and 35 under-expressed ) . These genes ( potential biomarkers ) have been sorted using the combination of expression rank in the aggregated dataset and the number of significant regulators as a score ( see Methods section ) . We also manually evaluated top-20 up-regulated genes and top-10 down-regulated genes in respect to the supporting evidences from the available literature . All analytical procedures were applied separately to over-expressed genes and under-expressed genes to look individually at processes and pathways activated and repressed in DMD . The significant regulators of up- and down- regulated differentially expressed genes from six datasets were cross-validated and only those identified in all datasets were selected for further analysis . They were combined with regulators obtained from the SNEA of the reference dataset to produce the final list of 76 unique significant regulators shown in Table 1 below . More information about these regulators can be found in Table S1 . We have selected genes , which were consistently differentially expressed in six datasets ( one aggregated dataset and five leave-one-out datasets ) . The fold-change threshold was established by analyzing fraction of genes present in all six top-k rankings for varying k , Figure 3 . As can be seen , fraction of common genes in top-k rankings for different types of gene expression reaches a plateau for k roughly equal to 500 . This means , that adding more genes will not increase percentage of overlap between different gene rankings . Hence we limited our analysis to top-500 differentially expressed genes for different types of regulation . The percentage of consistent genes in top-k of all datasets is about 40% ( Figure 3 ) . It means that analysis of differentially expressed genes from a single dataset can potentially lead to 60% of false positives . To increase reproducibility of obtained results we focused on the genes , presented in all six top-500 rankings . From the top 500 up-regulated genes in aggregated dataset we have selected 240 genes also present among top 500 up-regulated in all 5 leave-one-out datasets . Similarly , from the top 500 down-regulated genes in aggregated dataset we have selected 191 genes also present among top 500 down-regulated in all 5 leave-one-out datasets . These two lists were combined into a single list of 431 consistently up/down regulated differential genes . We performed Fisher exact test to find significantly enriched categories from Gene Ontology , corresponding to biological processes . Results , presented in Table 2 , in general reflect known changes that take place in affected muscles: up-regulated genes are commonly associated with inflammation and immune response , apoptosis and wound healing; down-regulated genes – with metabolic processes and muscle contraction . Genes were further analyzed in order to evaluate their quality as biomarkers . A promising biomarker should be easily detected and correspond to a DMD-related process ( e . g . muscle biology , fibrosis , inflammation ) or DMD-related condition ( e . g . dilated cardiomyopathy ) . We used a proprietary Ariadne DiseasesFX Database , which contains literature-extracted information about various types of relations between genes and diseases as well as data on presence of gene products in biofluids and among secreted proteins . We also made use of Ariadne ResNet 7 and Muscle Biology Gene Ontology , see Methods . Associations between 431 consistently up/down regulated genes and DMD-related processes and conditions are depicted in Table S2 . Out of 431 consistently changed genes , we have selected only those which are expression targets of significant regulators , selected using the above procedure . This produced a list of 140 candidate genes ( 35 down-regulated , 105 up-regulated ) that have been finally sorted using combination of rank in aggregated dataset and number of significant regulators ( see Methods ) . Most of them correspond to the processes of development and regeneration , immune response , response to glucocorticoids , hypoxia and extracellular matrix organization . Top-ranked 20 positive and 10 negative genes have been individually analyzed using biological information available from scientific literature ( PubMed ) . Mainly they are connected to fibrosis , inflammation , energy metabolism and other processes known to be affected in DMD . It was found that 12 out of these 30 were previously reported as related to muscle processes/disorders , the fact that can be considered as a proof of concept , providing the possibility to suggest new possible biomarker candidates on the base of suggested procedure . In summary , this study demonstrates the possibility to decipher regulatory mechanisms of the specific disease ( Duchenne dystrophy here ) along with corresponding exploratory biomarkers on the base of multiple microarray data meta-analysis only . A lot of predicted expressional regulators are known to be involved in DMD , suggesting that others will also be verified hereafter . This means that all of the proposed regulators can be considered for further drug discovery , whereas their consistently differentially expressed downstream genes can serve as exploratory biomarkers with implicated mechanistic models .
All available microarray datasets of human DMD with more than 10 samples ( total 5 datasets , see Table 3 ) were downloaded from NCBI GEO database [http://www . ncbi . nlm . nih . gov/geo/] . For each probeset intensity values were log-transformed and normalized to zero mean and unit variance . Missing data were imputed using K-nearest neighbor method with k = 10 . We have also utilized data presented in [26] , where the lists of up- and down-regulated genes were extracted from research papers , related to skeletal muscle development and pathologies . We limited this dataset to studies of DMD or mdx mice resulting in total 2227 genes which were reported to be differentially expressed in at least in one paper prior to December 2005 . For these genes we generated a pseudo-expression dataset for further analysis similar to the standard microarray experiment . If gene was reported to be up-regulated , the gene was assigned a positive value equal to corresponding number of supporting studies; if gene was reported to be down-regulated , the assigned value was negative . To combine the data from different datasets , we performed the following aggregation procedure . For each probeset we calculated within-dataset log-ratio , two-sample Welch's t-test , Wilcoxon rank sum test and area under ROC curve . If gene on a chip was represented by two or more probesets , we selected the probeset with the least p-value for Wilcoxon rank sum test . We also calculated several other statistics , using popular methods designed specifically for microarray data: limma , SAM and shrinkage T-statistic . Limma , Linear Models for Microarrays [1] , [84] , is based on a Bayesian hierarchical model for posterior odds of differential expression . SAM , Significance Analysis of Microarrays , was proposed in [2] . Shrinkage T-statistic stabilizes the variances in the denominator via a James-Stein approach [3] . Finally , we have combined the results from different experiments to generate the single “differential” rank for each gene . Separate gene rankings were obtained for nine measures: log-ratio , Welch's t-statistic and corresponding p-value , Wilcoxon's W-statistic and corresponding p-value , AUC , limma , SAM and shrinkage T-statistic . We used Fisher's method to combine p-values of the same type [85]; values of other statistics were averaged for each gene . The final gene rank R was calculated as mean of the ranks from all methods . Each gene was also assigned a single differential log ratio value calculated as an average differential log-ratio from 5 original gene expression datasets . In order to ensure reproducibility of obtained results , we performed a procedure , analogous to leave-one-out cross-validation: we constructed additional datasets each time aggregating 4 distinct microarray experiments out of total 5 available experiments . Thus we obtained 5 leave-one-out datasets where each microarray experiment was omitted . We also built one large dataset , where all 5 available microarray experiments were aggregated . All subsequent analyses were performed for resultant 6 datasets and the results were cross-validated as further described . For functional analysis of high-throughput data on the level of potential regulators we used Sub-Network Enrichment Analysis ( SNEA ) algorithm , implemented in Pathway Studio software [11] . SNEA is a variation of gene set enrichment analysis algorithm , but unlike GSEA [5] , [6] that uses predefined gene sets , SNEA utilized sub-networks to construct gene sets on the go . Here , each subnetwork consists of a node ( mainly protein or class of proteins – “functional class” ) in ResNet and all its expression downstream targets which are automatically derived from the literature . Global expression network includes direct ( i . e . transcriptional factor A1 is reported in the literature to regulate specific gene B1 ) and indirect ( i . e . growth factor A2 , that can activate specific signaling pathway results to the change of downstream gene B2 expression ) relations Ai->Bi . For each subnetwork seed SNEA considers all its expression targets as a gene set that is used for the classical GSEA ( Mann-Whitney or Kolmogorov-Smirnov statistical tests ) . Thus , SNEA determines the activity of expression regulators based on the differential expression of its targets and favors ( assigns lower p-value ) those of them which have more significant expression changes downstream . We performed the SNEA in Pathway Studio with the default parameters: Sub-Network type: gene expression , Mann-Whitney test , p-value<0 . 05 , number of regulators <100 for all log-ratio values ( DMD vs . control ) from the 6 aggregated datasets . The consistency of default parameters has been tested using 10 permutation tests . It has been shown , that the rate of significant SNEA seeds accidentally found in SNEA results applied to randomized experiment is less than 5% , which is in agreement with default p-value cutoff 0 . 05 . For the reference dataset we ran SNEA with the same parameters using number of studies which reported gene to be differentially expressed . All enrichment algorithms were applied separately to over-expressed and under-expressed genes . The final sorting of the differentially expressed genes have been done using the following scorewhere N – number of significant regulators upstream of the i-th gene and R –gene rank in aggregated dataset resulted from expression data analysis only . Most computations were done using R [http://www . r-project . org/] and BioConductor [http://www . bioconductor . org/] . Values of limma , SAM and shrinkage T-statistic were computed using GeneSelector package [86] . Sub-Network Enrichment Analysis was performed using Pathway Studio 7 . 1 from Ariadne Genomics along with ResNet 7 , database storing literature-derived network of biological relations [http://www . ariadnegenomics . com/] . Proprietary Ariadne DiseasesFX database was used for evaluation of gene quality as disease biomarker [Table S2] , and ChemEffect [12] was used for studying drugs , related to the regulators of interest . Muscle Biology Gene Ontology [http://wiki . geneontology . org/index . php/Genes_Involved_in_Muscle_Biology] was used to select genes associated with muscle-related processes . | Comparison of gene expression in diseased and normal tissue is a powerful tool of studying processes involved in pathogenesis and searching for potential drug targets and biomarkers of the disease's progression and treatment outcome . We have developed a novel approach for systematic knowledge-driven analysis of gene expression profiling data , which can suggest the underlying cause of the observed differential expression by identifying which expression regulators might be involved . These regulators can not only be the promising subjects of further investigation , but also potential drug targets , as normalization of their activity might alleviate some of the disease's symptoms . The targets downstream of suggested regulators can be proposed as exploratory biomarkers in disease treatment and prognosis . We used our approach to analyze public gene expression datasets of Duchenne muscular dystrophy – a progressive inherited disease in males . Some of the regulators and biomarkers that we found were already investigated in the context of DMD , while some of them were not yet studied and may be of interest for biological and clinical studies . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
]
| [
"medicine",
"pathology",
"drugs",
"and",
"devices",
"sports",
"and",
"exercise",
"medicine",
"pediatrics",
"pharmacology",
"pediatrics",
"and",
"child",
"health",
"biology",
"systems",
"biology",
"diagnostic",
"medicine",
"drug",
"research",
"and",
"development",
"neurological",
"disorders",
"neurology",
"general",
"pathology",
"computational",
"biology",
"genetics",
"and",
"genomics"
]
| 2012 | Novel Approach to Meta-Analysis of Microarray Datasets Reveals Muscle Remodeling-related Drug Targets and Biomarkers in Duchenne Muscular Dystrophy |
Chronic inflammation promotes oncogenic transformation and tumor progression . Many inflammatory agents also generate a toxic microenvironment , implying that adaptive mechanisms must be deployed for cells to survive and undergo transformation in such unfavorable contexts . A paradigmatic case is represented by cancers occurring in pediatric patients with genetic defects of hepatocyte phosphatidylcholine transporters and in the corresponding mouse model ( Mdr2-/- mice ) , in which impaired bile salt emulsification leads to chronic hepatocyte damage and inflammation , eventually resulting in oncogenic transformation . By combining genomics and metabolomics , we found that the transition from inflammation to cancer in Mdr2-/- mice was linked to the sustained transcriptional activation of metabolic detoxification systems and transporters by the Constitutive Androstane Receptor ( CAR ) , a hepatocyte-specific nuclear receptor . Activation of CAR-dependent gene expression programs coincided with reduced content of toxic bile acids in cancer nodules relative to inflamed livers . Treatment of Mdr2-/- mice with a CAR inhibitor blocked cancer progression and caused a partial regression of existing tumors . These results indicate that the acquisition of resistance to endo- or xeno-biotic toxicity is critical for cancers that develop in toxic microenvironments .
The microenvironment of chronically inflamed tissues is a source of multiple mediators that trigger and sustain cellular transformation and tumorigenesis [1–4] . Within the broad and heterogeneous group of inflammation-associated cancers , a distinct class is represented by those tumors that develop within a microenvironment containing high concentrations of toxic substances causing chronic cellular damage and compensatory tissue regeneration . A straightforward logical assumption is that for cells to emerge , thrive and eventually develop cancers in such contexts , they must acquire early in tumorigenesis the ability either to efficiently cope with the damage exerted by toxic agents or to promote their detoxification . To directly test this hypothesis , we used a well-characterized model of liver cancer , in which the absence of ABCB4 , a transporter for phosphatidylcholine expressed selectively in hepatocytes and encoded by the Mdr2 gene , results in defective emulsification of bile acids and their precipitation on the bile canalicular surface of hepatocytes , thus leading to membrane damage , cell death and chronic inflammation . In the absence of any exogenous mutagen , Mdr2-/- animals develop liver cancers with 100% penetrance at 12–15 months of age [5–9] . These cancers are etiologically and genetically similar to those occurring in pediatric patients with type 2 Progressive Familial Intrahepatic Cholestasis ( PFIC ) , in which mutations in the same family of hepatocyte transporters results in liver cancer by the age of five [10 , 11] . To understand the molecular bases of cancer development in this specific context , in which cellular toxicity is caused by chronic exposure to non-neutralized endogenous compounds ( namely , non-micellar hydrophobic bile acids ) , we performed gene expression , epigenomic and metabolomic profiling in hepatocytes to identify changes in gene expression programs and regulatory networks associated first with inflammation and then with cancer development . We found that while chronic liver inflammation was associated with the induction of a stress-response characterized by the induction of metalloproteinases and collagen genes among the others , Hepatocellular Carcinoma ( HCC ) development was characterized by the downregulation of these inflammatory programs and instead a robust transcriptional activation of genes encoding enzymes involved in the two phases of metabolic transformations and detoxification , namely Phase I ( oxidation , reduction and hydrolysis ) and Phase II transformations ( conjugation , e . g . to glutathione ) , as well as efflux transporters involved in the extrusion of transformed metabolites from cells . The induction of such response was associated with reduced content of toxic bile acids in cancer nodules relative to inflamed livers . Computational mining of the genomic data indicated that this gene expression program was driven by the increased expression and activation of the hepatocyte-specific Constitutive Androstane Receptor ( CAR , encoded by the Nr1i3 gene ) , a transcription factor of the nuclear receptor superfamily known to control xenobiotic detoxification genes . Consistent with these data , CAR inhibition with a specific antagonistic ligand reduced tumor burden and resulted in the regression of cancer nodules . Altogether , our data suggest that by mounting an appropriate detoxification response , hepatocytes became able to cope with the accumulation of toxic bile acids during liver inflammation , thus acquiring the capacity to thrive and undergo neoplastic transformation in an otherwise toxic environment .
We first performed RNA sequencing ( RNA-seq ) to analyze the changes in transcriptome of hepatocytes during Mdr2-/- liver disease progression . Mdr2 is expressed selectively in hepatocytes , thus justifying the use of a full knockout for these experiments . Hepatocytes represent more than 75% of the cell populations of a normal liver . However , immunohistochemistry ( IHC ) staining of liver sections showed a massive infiltration of IBA-1 positive macrophage cells in inflamed livers of 8-months old mice and even more so in HCC nodules ( 15 to 17 months old mice ) ( S1A and S1B Fig ) . Moreover , attempts to isolate pure hepatocytes after collagenase perfusion via the portal vein of Mdr2-/- mice were not successful because of the extensive co-purification of macrophages . Therefore , to obtain hepatocyte-enriched liver samples we treated mice with liposomes loaded with clodronate to deplete liver macrophages [12] . The 48 hours clodronate treatment resulted in a significant macrophage depletion from both inflamed livers and cancer nodules of Mdr2-/- mice , as shown in S1A and S1B Fig . An RNA-seq analysis carried out in livers of untreated and clodronate treated mice revealed that genes differentially expressed by clodronate treatment were significantly enriched for ontology terms associated to macrophage and lymphocyte function ( S1C Fig , S1 Table ) . In addition to macrophage depletion , clodronate treatment reduced B and T lymphocytes content in the normal liver but it had no significant effect on the inflamed and neoplastic livers of Mdr2-/- mice ( S2 Fig ) . Therefore , we used this approach to generate RNA-seq data sets in clodronate-treated and macrophage-depleted livers , including: i ) inflamed livers of 8 months old Mdr2-/- mice; ii ) isolated nodules from 15-to-17-months old Mdr2-/- mice , and iii ) age matched FVB/NJ wild type mice . All samples were sequenced to an average depth of ~30 million paired-end reads , using five animals per experimental group ( S3 Fig ) . Of the 1279 differentially expressed genes ( DEGs ) identified in the comparison of inflamed vs . wild type livers ( fold change ≥ ∣2∣ , q-value ≤ 0 . 05 , FPKM ≥ 2 in at least one time point ) , 1100 ( 86% ) were upregulated in the inflamed livers , consistent with a widespread increase in the expression of inflammatory genes ( Fig 1A , left panel ) . In the comparison between inflamed livers and HCC nodules , gene expression changes were of comparatively lower amplitude ( 579 DEGs ) and they occurred similarly in both directions: 265 genes ( 45 . 8% ) were up regulated and 314 ( 54 . 2% ) were down-regulated ( Fig 1A , right panel; the complete list of genes is reported in S2 Table and a Q-PCR validation of selected DEGs is reported in S4 Fig ) . A gene ontology ( GO ) analysis of the genes differentially expressed in inflamed vs . WT livers and in HCC vs . inflamed livers is shown in Fig 1B . The full list of enrichments using two different approaches is reported in S3 Table ( David GO analysis ) and S4 Table ( Revigo clusters ) . The top-ranking categories of up-regulated genes in inflamed livers were related to cell adhesion , migration , organization of the extracellular matrix and actin cytoskeleton . Some notable genes related to cell adhesion and cytoskeleton organization included many collagen genes ( Col3a1 and Col4a5 among many others ) , cadherin-1 ( Cdh1 ) and matrix metalloproteinase-2 ( Mmp2 ) ( Fig 1C and 1D ) . In addition , nearly all enzymes involved in glutathione metabolism were among the most up-regulated genes in inflamed livers , including glutamate-cysteine ligase ( Gclc ) , the first and rate-limiting enzyme of glutathione synthesis , several glutathione S-transferases ( Gstt2 and Gsta2 among the others ) , glutathione peroxidases ( such as Gpx4 ) and the main transcriptional regulators of antioxidant genes , Nfe2l2 ( encoding for NRF2 ) and its dimerization partner Mafk ( Fig 1C and 1D and S2 Table ) . These observations are consistent with the notion that chronic portal inflammation in Mdr2-/- livers results in increased oxidative stress as well as connective tissue deposition , leading to progression to fibrosis [9 , 13] . More interestingly , while all genes involved in fibrosis were selectively down regulated in the transition from inflammation to cancer , the antioxidant response was further upregulated in liver cancers ( Fig 1B and 1D ) . GO categories associated to genes down-regulated in inflamed livers were mainly related to oxidation-reduction processes ( Fig 1B ) . Specifically , genes encoding hydroxylases involved in bile acid biosynthesis ( such as the critical enzymes Cyp7b1 and Cyp8b1 ) were downregulated as part of the negative feedback exerted by the excess of non-neutralized bile acids ( S2 Table ) . Several Phase I enzymes belonging to the cytochrome p450 ( CYP450 ) superfamily ( Cyp2a5 , Cyp2c39 , and Cyp4a32 among many others ) and their main transcriptional regulator , the hepatocyte-specific nuclear receptor Constitutive Androstane Receptor ( CAR , encoded by Nr1i3 ) [14 , 15] were also downregulated in inflamed livers ( Fig 1D and 1C ) . Phase I enzymes act by hydroxylating hydrophobic xeno- and endobiotics ( including hydrophobic bile acids ) , thus increasing their solubility and reducing their toxicity . Interestingly , the same group of genes was strongly up-regulated in cancer nodules together with classical liver cancer markers ( Afp , encoding alpha-fetoprotein among many others ) ( Fig 1C and 1D ) [16 , 17] . Overall , transcriptomic analyses indicate that the massive inflammatory and fibrotic response occurring in inflamed Mdr2-/- livers is reduced in tumor nodules , concurrently with the upregulation of hydroxylases of the CYP450 family . In order to better discriminate groups of genes with distinct behaviors during tumorigenesis , we divided differentially expressed genes into 8 clusters ( S5 Table ) based on their transcriptional profiles in the two disease stages considered ( Fig 2 , left panels ) . Next , to identify the transcription factors ( TF ) that selectively control gene expression in each cluster , we determined the TF consensus DNA binding sites ( described by position weight matrixes , PWMs ) that were statistically overrepresented in the promoters of the differentially expressed genes relative to a background including the promoters of all Ensembl-annotated genes ( ± 1000 bp relative to their transcription start sites ) [18] . An initial list of candidate regulatory TFs was generated based on the statistical over-representation of the cognate DNA recognition motif and then filtered based on the expression of each TF ( Fig 2 , right panels ) . We also performed GO analysis of genes differentially expressed within these 8 clusters and identified the associated enriched GO terms ( Fig 2 , bottom panels ) . The full list of GO categories and PWMs associated to each cluster is available in S6 and S7 Tables . Overall , cluster 1 , 2 and 3 included genes whose expression was increased in inflamed livers . While the expression of genes associated with fibrosis and inflammation ( cluster 1 and 2 ) decreased in tumor nodules , genes involved in oxidative stress responses , angiogenesis and cell proliferation ( cluster 3 ) were all further up-regulated in the transition from inflammation to cancer . To identify the TFs selectively involved in the transition between inflamed and tumor livers , we focused our attention on the genes that were selectively up-regulated at the tumor stage ( cluster 4 , 5 and 6 ) . Interestingly , this subset of genes was homogeneously related to oxidation-reduction processes , notably those catalyzed by CYP450 family hydroxylases and involved in mono-oxygenation and detoxification of hydrophobic substances . Within the same clusters , the most over-represented DNA binding motifs included those for CAR ( Constitutive Androstane Receptor , encoded by Nr1i3 ) , AHR ( Aryl Hydrocarbon Receptor ) and its dimerization partner ARNT ( AHR Nuclear Translocator ) , three TFs significantly overexpressed at the HCC stage ( S2 Table ) . Overall , these data indicate that the transition from inflammation to cancer is associated with the increased expression of genes involved in detoxification of xeno- and endobiotics such as bile acids [14] . To gain insight into the functional consequences of transcriptional changes occurring in the transition from inflammation to tumors , we used Ingenuity Pathway Analysis ( IPA ) . IPA was used to assemble a network based on genes differentially expressed between inflamed and tumor samples ( Fig 3 ) . This analysis revealed that tumor development was characterized by a global up regulation of members of the endobiotics detoxification pathways . Specifically , toxic bile in Mdr2-/- livers induced the expression of CAR ( Nr1i3 ) , which regulates Phase I enzymes ( Cyp1a2 , Cyp2a5 , Cyp2c and Cyp4a families , POR , NQO1 ) , Phase II enzymes involved in conjugation of toxic agents ( such as several glutathione S-transferases ) and Phase III transporters ( Abcc4/Mrp4 , which mediates the cotransport of reduced glutathione with bile acids , thus enhancing their detoxification ) [14 , 19] . It is important to note that also PXR ( Nr1i2 ) , whose expression was increased in inflamed liver ( cluster 3 in Fig 2 ) , is activated by bile acids , in particular by the highly toxic lithocholic acid ( LCA ) [20] and that it also regulates Phase I and II enzymes and Phase III transporters . In the transition from inflammation to tumors NRF2 ( Nfe2l2 ) was downregulated but the antioxidant pathway was maintained upregulated probably by the xenobiotic receptors CAR , PXR and AHR . We next reanalyzed previous whole exome sequencing ( WES ) and whole genome sequencing ( WGS ) data we generated using Mdr2-/- cancers [11] in order to determine whether genes encoding enzymes involved in bile acid detoxification undergo amplification events . We found that a consistent number of genes ( n = 44 ) involved in endobiotics detoxification pathway were amplified in 8 out of 10 sequenced samples ( S5 Fig ) . Overall , these data suggest the existence of a selective pressure favoring the overexpression of genes encoding enzymes that control detoxification of bile acids and that in some cases this can be achieved by gene amplification . To obtain a more detailed view of the gene regulatory networks underlying adaptive changes in gene expression during tumorigenesis in the Mdr2-/- livers , we performed H3K27Ac ChIP-seq on the same fifteen samples ( five per experimental group ) used for RNA-seq profiling . Histone H3 Lysine 27 ( H3K27 ) acetylation is a histone modification deposited at gene promoters and at enhancers when they are bound by activating TFs , and is thus informative of the activity state of these cis-regulatory elements [21–23] . Based on the H3K27Ac profiles , WT liver , inflamed Mdr2-/- livers and cancers clustered separately ( Fig 4A ) , indicating a differential usage of the genomic cis-regulatory information . To gain insight into the functional responses controlled by cis-regulatory regions selectively acetylated in inflamed livers and then in cancers , we used the GREAT tool [24] . GREAT links sets of genomic regions to putative biological functions based on the functional annotations of the nearby genes , with a score that takes into account the distance between regions and genes and therefore the likelihood of correct assignment . Genomic regions specifically activated in inflammation were enriched in functional terms related to extracellular matrix organization , as well as with leukocyte migration , a result consistent with the inflammatory and fibrotic response identified by gene expression profiling at the same stage ( S8 Table and Fig 4B ) . Coherently with transcriptomic data , the genomic regions associated to these fibrotic functional terms were deacetylated in the transition from chronic inflammation to cancer , together with regions associated to developmental functions . The transition to inflammation was also characterized by a repression of acetylated regions associated with liver metabolic function ( oxidation-reduction process ) . Interestingly , the oxidation-reduction processes terms were retrieved on regulatory regions upregulated in the comparison between nodules and inflammation ( Fig 4B ) . Snapshots of three representative samples ( Fig 4C ) show tumor-specific increased acetylation and RNA expression of Cyp2c37 and Cyp4a14 , encoding Phase I detoxification enzymes , and Abcc4 , encoding a bile acid transporter . Conversely , the H3K27Ac signal on the promoter of the Epcam gene , encoding the cell adhesion molecule , was induced in inflamed livers and subsequently repressed in tumors . Next , we determined the TF consensus DNA binding sites that were statistically overrepresented in the differentially acetylated cis-regulatory regions [18] . In line with the GREAT analysis , motifs for TFs involved in the inflammatory and fibrotic response ( including NF-kB and AP-1 ) were enriched in regions that gained acetylation in inflamed livers relative to normal controls , ( Fig 4D and S9 Table ) . More importantly , when analyzing motifs over-represented in cancers relative to inflamed livers , we identified the DNA binding site for CAR ( Nr1i3 , belonging to the NR1 family of Thyroid hormone receptor-related factors ) , which is also overexpressed in Mdr2-/- liver cancers . Overall , these results are in line with the transcriptional profiling datasets discussed above and suggest that the increased activity of a CAR-activated detoxification pathway occurs ( and might be functionally involved ) in the transition from inflammation to cancer . The upregulation of CAR expression in HCC as well as the over-representation of CAR motifs in the cis-regulatory elements that were selectively hyper-acetylated in cancers as compared to inflamed livers , suggest the hypothesis that transformed hepatocytes have acquired the ability to efficiently detoxify bile acids , which in turn would explain their ability to survive and proliferate in the context milieu of Mdr2-/- livers . To directly address this possibility , we used HPLC separation coupled to tandem mass spectrometry ( HPLC-MS/MS ) to quantify the concentrations of free bile acids and their taurine and glycine conjugates in the same liver samples used for expression and epigenetic analyses . 15 different species of bile acids were unambiguously identified and quantified ( Table 1 ) . PCA analysis showed that wild type , inflamed and tumor samples could be clearly separated from each other and that tumor samples were more similar to wild type controls than to inflamed livers ( Fig 5A ) . Total bile acid levels were significantly increased in inflamed livers compared to age-matched controls ( Fig 5B ) . More interestingly , tumor samples were characterized by a significant decrease ( p ≤ 0 . 05 ) of total bile acids levels , which however did not return to the same values as those measured in the normal tissue . The observed decrease is not due to an impaired bile acids synthesis by tumor cells , as the expression of the rate-limiting enzymes Cyp7a1 and Cyp27a1 was not affected in the transition from inflammation to tumors . When only hydrophobic bile acids ( namely , those with the highest cytotoxic potential ) were considered , tumor livers showed a robust decrease in their levels , that returned to those measured in the matched wild type controls ( Fig 5C and Table 1 ) . In particular , the most hydrophobic and toxic bile acid , lithocolic acid ( LCA ) decreased to median levels that were even lower than those measured in normal livers ( Fig 5D ) . Overall , these data demonstrate that the total pool of bile acids , and even more so the hydrophobic and toxic ones , were substantially reduced in the transformed hepatocytes of Mdr2-/- tumors , indicating that the transition between the inflammatory stage and tumors involves an increased ability to detoxify and dispose of these endobiotics . Finally , we set out to investigate whether CAR inhibition might impact viability of tumors in Mdr2-/- mice . To this aim , we randomized 15 months-old Mdr2-/- mice ( 5 animals per group ) to receive either 5α-androstan-3α-ol , a selective CAR inhibitor [25] or vehicle only via intraperitoneal injection . After 2 weeks of CAR inhibitor treatment and 48 hours of clodronate treatment to deplete macrophages , mice were sacrificed and tumors from the two cohorts were compared in terms of nodule number , size , histology and tumor content ( S10 Table ) . Alanine aminotransferase ( ALT ) and aspartate aminotransferase ( AST ) levels were determined in plasma samples collected from vehicle and CAR inhibitor-treated mice to determine the possible occurrence of general hepatotoxicity . ALT and AST plasma levels were constitutively elevated in Mdr2-/- mice but not further increased following treatment ( S6 Fig ) . Expression of representative CAR targets measured by quantitative RT-PCR on nodules from treated and untreated mice ( such as Cyp3a11 , Cyp2b10 , Cyp2c37 and Cyp1a2 ) [15] , was repressed by the CAR inhibitor , thus indicating efficient CAR inhibition in treated mice ( Fig 6A ) . Treated animals showed a significantly lower number of neoplastic lesions ( adenomas and carcinomas ) per mouse when compared with the untreated cohort ( Fig 6B and S7 Fig ) . When only carcinomas were considered , treated mice showed a significant depletion in HCC nodules ( Fig 6C ) , while adenomatous nodules ( containing no HCC foci ) were over-represented ( Fig 6D ) . In addition , no mouse treated with CAR inhibitor had nodules bigger than 20 mm ( Fig 6E ) . Altogether , inhibition of CAR activity had an overall negative impact on tumor progression and on existing cancer nodules .
A large body of evidence indicates that chronic inflammation is associated with many cancer types but it is still unclear how cancer cells become able to survive within the cytotoxic microenvironment associated with high local concentrations of some inflammatory agents . In this study , we set out to dissect the epigenetic and transcriptional changes occurring in the transition from chronic inflammation to tumors in the specific context of the liver carcinogenesis in the Mdr2-/- mouse . Although this mouse is traditionally considered a paradigmatic model of inflammation-induced cancer , it is genetically and etiologically different from common types of human HCC and it is more appropriate to deem it representative of those tumors that develop in the context of a highly toxic microenvironment that determines a vicious circle of cell death and regeneration . The main finding of our study is that for tumors to develop in such a context , it is necessary that transcriptional and genetic changes occur that equip parenchymal cells with the ability to resist toxicity exerted by micro-environmental components . By gene expression and epigenomic profiling , we found that HCC livers are characterized by CAR-mediated activation of Phase I and II detoxification pathways and transporters , which promote the detoxification and excretion of toxic bile acids that accumulate in livers of Mdr2-/- mice and cause the extensive inflammation typical of the precancerous stage . Elimination of bile acids includes Phase I reactions ( mainly consisting in the hydroxylation of bile acids and in a consequent reduction of their hydrophobicity ) and Phase II reactions consisting in the conjugation of bile acids with molecules that further increase their hydrophilicity , thus reducing toxicity and enabling urinary excretion [26] . In cholestatic diseases , Pregnane X Receptor ( PXR , Nr1i2 ) and the Constitutive Androstane Receptor ( CAR , Nr1i3 ) represent the two key nuclear receptors controlling the expression of enzymes and transporters involved in bile detoxification and excretion in hepatocytes [27] . We found that PXR expression is induced already in inflamed livers and remains high in HCC . This effect might be the result of PXR activation by LCA , a well know direct ligand of PXR [20] , which we found increased in inflamed livers . Interestingly , the appearance of cancer nodules in the inflamed livers is characterized by a transcriptional upregulation of the two other main regulators of xenobiotic metabolism , namely AHR and CAR . As a result , the majority of genes involved in Phase I and Phase II detoxification pathways are strongly upregulated in cancers . Consistently , the increased expression of xenobiotic receptors is paralleled by the upregulation of genes with overrepresented xenobiotic receptors ( XR ) motifs in their promoters . It is also interesting to note that tumors developing in Mdr2-/- mice show a very low burden of DNA mutations but massive gene amplification and rearrangements at late cancer stages [8 , 11] . Genes encoding components of the xenobiotic detoxification pathways are indeed amplified themselves , which is consistent with a selective advantage provided by their increased expression . Overall , these results indicate that the accumulation of toxic hydrophobic bile acids , and in particular LCA , during chronic inflammation activates a detoxification program that is initially insufficient to prevent cellular damage and the vicious cycle of death and regeneration that characterizes the long pre-tumoral stage of liver disease . It is only a further increase in the activity of the detoxification program controlled by xenobiotic receptors that enables hepatocytes to undergo uncontrolled proliferation , likely stimulated by the chronic inflammatory environment . In this regard , CAR has been shown to be activated by LCA in vivo and other bile acids were found to activate the ligand binding domain of CAR in vitro [15 , 28] . The secondary bile acid LCA is the most potent cholestatic agent and causes liver damage unless it is efficiently eliminated [29] . Notably , CAR has been shown to have a fundamental protective role in the response to LCA in vivo since CAR-KO mice have more severe defects in LCA detoxification compared to PXR-KO mice [15] and CAR activation in transgenic mice confers resistance to the hepatotoxicity of LCA [30] . Importantly , our results indicate that CAR inhibition arrests tumor progression in that it reduces the number of bigger lesions with high HCC content . Although it was not feasible to test CAR inhibitor during the transition from inflammation stage to early adenoma ( a six months-long process ) , it is tempting to speculate that pharmacological inhibition of CAR may be useful also to block HCC onset . Consistently with our findings , chronic CAR activation has been shown to result in liver carcinogenesis , as CAR-KO mice are completely resistant to tumorigenic effects of chronic xenobiotic stress [31] and long-term activation of CAR and β-catenin induces liver tumorigenesis [32] . Overall , our study points to a general framework for tumorigenesis occurring in the context of toxic micro-environments that may extend to other cases such as tumors associated with chronic exposure to noxious chemicals . Specifically , conditions that induce a stress response program also increase the selective pressure on pre-neoplastic cells to develop powerful mechanisms to cope with the same stress , as also indicated by the frequent amplification of genes encoding components of the endo/xenobiotic detoxification pathways in Mdr2-/- HCC . These data also suggest the possibility to use prophylactic or therapeutic approaches targeting xenobiotic receptors in such contexts .
Experiments involving mice have been carried out in accordance with the Italian Laws ( D . L . vo 116/92 ) which enforces the EU 86/609 directive . The Ministry of Health was notified of this project in March 2014 ( Project number: 02/2014 ) . Founders of the FVB . 129P2-Abcb4tm1Bor ( Mdr2-/- ) and FVB/NJ ( Mdr2-WT ) mice were purchased from The Jackson Laboratory . Colonies of both strains were maintained under specific pathogen-free conditions . Mice ( both males and females ) were treated with liposomes loaded with 5 μg clodronate or with PBS ( www . clodronateliposomes . com ) via tail vein injection , 48h prior to sacrifice . Each nodule or liver tissue sample was partly being snap frozen for DNA/RNA/protein extraction . Furthermore , a portion of each specimen was histologically assessed after overnight fixation in 4% formaldehyde and paraffin inclusion . Anti-IBA1 immuno-stains were performed on 4 μm sections . After de-waxing and re-hydration in ethanol , antigen de-masking was done in sodium citrate buffer in a water bath at 95°C for 45 minutes . Endogenous peroxidases were quenched with a 5 min treatment in 3% H2O2 . Slides were incubated with rabbit IBA1 antibody ( Wako , 019–19741 ) diluted 1:500 , and developed with HRP polymer ( DAKO ) . Slides were finally counterstained with hematoxylin and mounted with Eukitt . The histological classification of hepatocellular proliferative lesions was performed according to Thoolen et al . [33] . For each mouse , either the composition of the tumor ( in terms of percentage of adenoma and/or carcinoma ) or the number of hepatocellular adenomas , early carcinomas ( defined as adenomas containing focus of arising carcinoma ) , carcinomas , and the total number of neoplastic lesions were evaluated by a mouse pathologist . Samples were coded without reference to experimental group and examined blindly . Blood samples were incubated on ice for 30 min to coagulate and were centrifuged for 10 min at 5000 rpm to separate the serum . Colorimetric determination of ALT levels was performed using TECO Diagnostics assay kits ( Teco Diagnostics , Anaheim , CA ) . Procedures were performed as described by the manufacturer , except for a proportional decrease in volume to minimize the use of serum per assay . Colorimetric determination of AST levels was performed in the diagnostic laboratory of Humanitas Clinical and Research Center . RNA-seq was carried out using previously described protocols [34] on an Illumina HiSeq2000 platform . Frozen tissue samples were homogenized with a dounce homogenizer or with gentleMACS Dissociator ( Miltenyi Biotec ) , depending on the tissue volume . Total RNA was extracted using Maxwell 16 LEV SimplyRNA cells kit ( Promega ) and run on Agilent Bioanalyzer 2100 to assess sample integrity . mRNA-seq library preparation from 4 μg of total RNA was performed with TruSeq RNA Sample Prep Kit V2 ( Illumina ) according to the manufacturer’s instructions . ChIP was carried out as previously described [34] . Briefly , 350 mg of liver/tumoral fixed tissue have been used for each ChIP . Chopped tissue samples were further homogenized with gentleMACS Dissociator ( Miltenyi Biotec ) prior to lysis . Homogenized tissues were processed with a two-step lysis protocol for cellular and nuclear membranes disruption , followed by chromatin shearing by sonication . Each lysate was then immunoprecipitated overnight with 5 μg of anti H3K27Ac antibody ( Abcam , ab4729 , [23] ) prebound to 100 μl of G protein-coupled paramagnetic beads ( Dynabeads ) . After beads washing , DNA was eluted and crosslink was reversed by overnight incubation at 65°C . DNA was then purified by Qiaquick columns ( Qiagen ) and quantified with PicoGreen ( Invitrogen ) . ChIP validation by Q-PCR has been done on an Applied Biosystems 7500 Fast Real-time PCR system ( SYBR Green , Applied Biosystems ) . ChIP DNA libraries were prepared as previously described [34] , and sequenced on an HiSeq2000 with a 36bp single end setting ( Supplementary materials ) . 15 months-old Mdr2-/- mice were treated with 5α-androstan-3α-ol ( Steraloids , Newport ) , a selective CAR inhibitor , as previously described [25] . The inhibitor was dissolved in a DMSO/corn oil solution and administered at 50 mg/kg by intra-peritoneal injection . Each mouse was treated every 48 h , and received a total of 6 treatments . Animals were finally sacrificed 48 h after the last inhibitor administration and after macrophage ablation by clodronate liposomes , and all detectable nodules were collected for histological analysis . Grossly detectable hepatic nodules were counted and measured with a caliper . Bile acid content was evaluated from normal , inflamed and HCC livers . Liver bile acids were extracted by Folch method in presence of 5-alpha-cholestane as internal standard and subjected to HPLC-MS/MS analysis . The analyses were performed on an API-4000 triple quadrupole mass spectrometer ( AB Sciex ) coupled with a HPLC system ( Agilent ) and CTC PAL HTS autosampler ( PAL System ) . A detailed description of the sample preparation and the subsequent MS analysis is provided in the Supplemental materials file . Short reads obtained from Illumina HiSeq2000 runs were analyzed as described [34] . Detailed computational methods are described in the Supplemental materials file . Accession numbers . Raw datasets are available in the Gene Expression Omnibus ( GEO ) database ( http://www . ncbi . nlm . nih . gov/geo ) under the accession GSE80777 , which comprises ChIP-seq data ( GSE80775 ) and expression data ( GSE80776 ) . | Chronic inflammation associated to the sustained exposure to toxic chemicals can lead to cancer , but how transforming cells acquire the ability to oppose chemo-toxicity and eventually thrive is unclear . In this study , we set out to profile the molecular changes occurring in a mouse model of liver disease caused by defective emulsification of bile acids , which leads to membrane damage , cell death , chronic inflammation and eventually cancer . We found that hepatocytes acquire early in tumorigenesis the ability to efficiently cope with the accumulation of toxic bile acids by increasing the expression of detoxification pathway components via both genetic and epigenetic mechanisms . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"liver",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"cancer",
"treatment",
"immunology",
"bile",
"cancers",
"and",
"neoplasms",
"carcinomas",
"gastrointestinal",
"tumors",
"liver",
"diseases",
"toxicology",
"oncology",
"signs",
"and",
"symptoms",
"gastroenterology",
"and",
"hepatology",
"genome",
"analysis",
"inflammation",
"genomics",
"animal",
"cells",
"gene",
"expression",
"hepatocytes",
"gene",
"ontologies",
"immune",
"response",
"hepatocellular",
"carcinoma",
"diagnostic",
"medicine",
"anatomy",
"detoxification",
"cell",
"biology",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"computational",
"biology"
]
| 2018 | Sustained activation of detoxification pathways promotes liver carcinogenesis in response to chronic bile acid-mediated damage |
The folding of proteins with a complex knot is still an unresolved question . Based on representative members of Ubiquitin C-terminal Hydrolases ( UCHs ) that contain the 52 knot in the native state , we explain how UCHs are able to unfold and refold in vitro reversibly within the structure-based model . In particular , we identify two , topologically different folding/unfolding pathways and corroborate our results with experiment , recreating the chevron plot . We show that confinement effect of chaperonin or weak crowding greatly facilitates folding , simultaneously slowing down the unfolding process of UCHs , compared with bulk conditions . Finally , we analyze the existence of knots in the denaturated state of UCHs . The results of the work show that the crowded environment of the cell should have a positive effect on the kinetics of complex knotted proteins , especially when proteins with deeper knots are found in this family .
The role of knots in protein structures is still not fully understood . The topological complexity induces stability to the structure [1 , 2] and enforces local motifs favorable for active sites of enzymes [3] . The latter fact may explain , why over 80% of known knotted proteins are enzymes with the active site located at the entangled region [4] . Nevertheless , folding process of knotted proteins is a fundamental and still not solved problem . One of the families of knotted proteins is Ubiquitin C-terminal Hydrolase ( UCH ) of which the characteristic feature is the presence of a complex topological fingerprint 523131 [4] as shown in the Fig 1 . This means that the entire protein forms a 52 knot as a whole , but some of its subchains form two trefoil knots ( see Table 1 ) . Each entry in the matrix indicates the knot type , formed by one continuous subchain , by one particular color; e . g . the unknot is denoted in white . Each such subchain starts with the N-terminal amino acid at position x and ends with the C-terminal amino acid at position y , and the corresponding colored entry in the matrix is shown in position ( x , y ) ( along respectively horizontal and vertical axes ) . Specifically , one can trace what is the topology of the subchain with one end in N-terminus . Pictorially , this is represented as the traveling down of the left-most vertical line in the matrix in Fig 1 ) . In the beginning , successive subchains are unknotted , however reaching at least Ile163 the subchain becomes trefoil knotted ( first green patch in the matrix ) . The subchain Met1-Tyr173 is still knotted , however then the chain winds back forming a slipknot loop and when the end of the subchain is in-between Glu174 and Pro180 ( parts of C-terminal β-strands ) , such subchain is unknotted ( the break between the green patches ) . Next , the subchain starting in N-terminus and ending in-between residues Pro182 and Ala216 is again trefoil knotted ( bottom green patch ) and finally the whole chain is 52 knotted ( blue patch ) . The 523131 fingerprint is unique and conserved in all UCH members , which are separated by billion years of evolution and exhibit a very low sequence similarity ( below 30% ) [5] . Notably , the formation of the larger trefoil results in the formation of the inner-most ( dipper ) trefoil knot . Therefore , in subsequent analysis by “formation of 31 knot” , we mean the formation of the larger ( and hence both ) trefoil knot . The UCH superfamily is a group of deubiquitinating enzymes ( DUBs ) . Their exact substrates have not yet been determined , however it seems that the role of UCHs is to detach the ubiquitin from small nucleophiles . Four of the UCH family members exist in humans: UCH-L1 , UCH-L3 , human UCH-L5 ( UCH37 ) and BAP-1 . They share a high degree of homology in their catalytic domains [6] , surrounded by the deepest 31 knot . Moreover , UCHs have a tissue-specific expression in complex organisms such as humans and their activity is crucial from the therapeutical point of view . For example , UCH-L3 has been shown to be upregulated in breast cancer tissues [7] , and a high expression of UCH-L5 is significantly associated with poor prognosis in human epithelial ovarian cancer [8] . On the other hand , UCH-L1 is one of the most common proteins in human brain ( composing up to 1-2% of the brain the total protein content [9] ) , and it is highly expressed in pancreatic [10] , esophageal [11] , prostate [12] , medullary thyroid [13] , colorectal carcinomas [14] and HPV16-transformed cells [15] . Its misfolded forms were connected with neuronal disorders such as Parkinson’s , Huntington’s and Alzheimer’s diseases [16] , which justifies the importance of studying the UCH folding process . In general , it is expected that folding of knotted proteins is governed mainly by the depth of the knot and the complexity of the topological fingerprint [17 , 18] . Self-tying was observed theoretically for the smallest knotted proteins with DNA binding motif and a rather shallow 31 knot [19 , 20] . They mainly fold by a slipknot conformation [17 , 19] . Similarly , it was shown that proteins with a deep trefoil knot , such as YibK and YbeA , can self-tie [21] . The theoretical results obtained for these proteins with a structure based model additionally revealed that a knotting event is a rate-limiting step [22] and the folding efficiency can be controlled by non-native contacts [23] or consideration of cotranslational on-ribosome folding [24] . For the protein with a 61 knot , DehI [25] , there were only a few successful folding pathways observed theoretically . Surprisingly , this protein folds via a simple mechanism: a large twisted loop formed on the backbone flips over another protein fragment previously arranged in a twisted loop , and in consequence , the six-fold knot is created in a single movement . Even though this protein is prone to aggregate , the experimental data support this mechanism [26] . These results suggest that bulk structure-based models can be used to investigate knotted proteins . On the other hand , experimental data show that knotting process of trefoil-knotted YibK and YbeA bacterial proteins can be specifically and significantly accelerated by the GroEL-GroES chaperonin complex [21] encapsulating the folding protein . This agrees with the theoretical investigation showing that knotting probability of polymers increases in confinement [27] . Only due to the encapsulation ( following [28] ) , successful reversible folding was observed for members of knotted proteins with DNA binding motif VirC and DndE [29] . It is then natural to expect , that chaperonins encapsulating proteins may also facilitate folding and self-knotting of eucaryotic UCHs , although no experimental result in this topic is available yet . Nevertheless , one has to bear in mind , that encapsulation is the simplest possible model of chaperonin , lacking many “biological features”—specific binding to the cage , chaperonin conformational changes , etc . However , any theoretical results have to be confronted with the experimental data concerning UCHs folding . It has been shown already that UCHs can fold and refold reversibly in two parallel pathways , each consisting of one slow and one fast phase , as determined from chevron plot [30 , 31] . Despite a common mechanism , folding processes of different UCHs is characterized by different kinetic parameters . Such differences can stem from a various depth of knots in UCH family [4] , ranging from rather shallow ( from the N-terminus ) to the deep knot , which we just found , as shown in Table 1 . The most attention-drawing are the S18Y and I93M mutations , which were found to modify ( either decrease or increase ) the risk of Parkinson’s disease [32] , and the intermediates on folding pathway , as these are especially prone to oligomerization [33] . However , the results concerning these mutations are variable and differ in different studies . Despite the successful assignment of the majority of signals in NMR spectrum of the UCH-L1 [34] , and in spite of studies of its tryptophan variants [33] , the exact conformation of intermediates is still unclear . This may be due to a broad structural plasticity around the intermediate states [30] . The self-tying was postulated via the direct knotting event in accordance with the theoretical study of the on-lattice model of a designed by hand heteropolymer chain with 52 knot [35] . However , the optical tweezers stretching experiment showed that the threading significantly decelerates the folding [36] . Still , because the topology cannot be detected in the in vitro experiment , the mechanism of knot tying remains unresolved . In this study , we asked following questions: What is the difference between the two experimentally observed , parallel folding pathways ? What is the influence of a chaperonin cage ( confinement ) on the folding and self-tying of UCHs ? And more generally , what are the dynamical properties of UCH in a bulk and in a confinement ? To answer these questions , we performed a comprehensive study of representative UCH members in a model of chaperonin cage ( mimicked by repulsive cavity ) and in the bulk , using structure-based model simulations . To ensure robustness of the results , we investigated proteins from different organisms , with low sequence similarity , and different depth of the knot . The results show that the structure-based model was sufficient to knot and unknot each of the studied proteins with and without the presence of confinement . However , only in the confinement , the simulations in transition temperature were accessible . We performed a comprehensive analysis of knot occurrence during simulations , resulting in the identification of two topologically distinct pathways . To relate our results to the experiment , we reproduce the chevron plot for a representative protein member of UCH family , revealing the existence of fast and slow phases . Next , we studied short-lived knots on the folding and the unfolding pathway and revealed for the first time the existence of random knots in the unfolded protein chain . To our knowledge , this is the first theoretical study with a direct investigation of the influence of the excluded volume on proteins containing complex knots .
To obtain robust results , three sequentially different members of UCH family denoted with their PDB codes 3IRT ( UCH-L1 ) , 2LEN ( UCH-L1 ) , and 4I6N ( UCH-L5 ) were studied . Additionally , to check an influence of the length of knot tails , we constructed in silico mutant of 4I6N , denoted as 4I6N-m , obtained by removing 7 residues from the N-terminus . All investigated structures feature a left-handed 52 knot and complex topological fingerprint 523131 . Alignment of the 3-dimensional structures of the studies proteins is shown in the Fig 1 and their most important topological and structural information is summarized in Table 1 . Further structural and sequential comparison of chosen structures is presented in S1 Appendix , part 1 . Not determined region ( amino acids 142–152 ) in the structure of 4I6N was repaired using Modeller software [37] , where the model with the lowest DOPE potential ( Discrete Optimized Protein Energy ) was chosen . The DOPE potential is one of the quantities assessing the structure correctness [38] . The dynamics of investigated structures was studied in structure based Cα model [39 , 40] with standard parameters as proposed by the SMOG server [41] . The model included bonded interactions ( bonds , planar and dihedral angles ) , bead excluded volume ( Lennard-Jones repelling part ) and non-bonded interactions described with a 10-12 Lennard-Jones potential . The non-bonded attraction was applied between residues forming contacts in the native structure , as defined in [42] . The number of the native contacts for all considered proteins is presented in Table 1 . The folding/unfolding transitions were studied through constant temperature molecular dynamics simulations with the Nose-Hoover thermostat ( coupling constant eq . 0 . 025 ) using Gromacs v4 . 5 . 4 package [43] . Temperature T ˜ in Gromacs is defined by the equation T ˜ = ( k b T ) / ( ϵ k˜b ) , where k˜ b= 0 . 00831451 . Through the text , the temperature is denoted simply as T and the Boltzman constant k ˜b as kb . There were performed 200 simulations for each structure and temperature . The number of steps was in the range 107 − 1 . 6⋅109 steps depending on the condition . The confinement is represented by a cylinder ( Fig 1 ) with a diameter equal to its height equal 6 . 0 nm [44] , introduced into the system as in [29] . The interactions between the inner wall of the cylinder and protein are purely repulsive ( only a confinement effect ) . Such model was previously used to study confinement or crowding effect on protein folding [44–46] . The data concerning the position of the knotted core and the length of the knot tails ( Table 1 ) are taken from KnotProt server [4] . The knot type of each of the subchains of the protein is determined using the implementation of the HOMFLY-PT polynomial [47–49] and the chain closing method as in [5 , 50] . The same algorithm was used to detect the entanglement along the protein backbone during simulations . The knot was regarded as present in the simulation if it was detected for at least 5 consecutive frames . The similarity to the native state was measured by the fraction of native contacts , Q . At given conformation , each native contact was regarded as present , if the distance between a pair of Cα atoms was less than 1 . 2 times their native distance . The untied structure was regarded as unfolded , if Q < 0 . 2 . By unfolding pathway , we mean the shortest part of trajectory connecting knotted structure with Q > 0 . 9 and unfolded structure . By folding pathway , we mean the shortest part of the trajectory connecting unknotted structure with Q < 0 . 4 and knotted structure with Q > 0 . 9 . Folding trajectories start from one of the previously generated 100 unknotted conformations with Q < 0 . 2 . All initial structures belong to separate clusters with 0 . 1 nm cutoff , to remove any possible bias . The structures were visualized using UCSF Chimera [51] .
The UCHs are known to fold and refold along two parallel pathways , each featuring one slow and one fast phase [30 , 31 , 52] . Therefore , to correlate our model with experimental results we recreated the chevron plot for UCH-L1 , with the temperature as a denaturant . In conventional chevron plots , the folding constants are calculated based on the time dependence of e . g . fluorescence . The fluorescence of a protein’s tryptophan depends on its neighborhood . Hence , the fluorescence trace can be understood as a measure of similarity of the tryptophan neighborhood to the native structure . In our simulations , such a measure is given by the fraction of the native contacts—Q . Therefore , we calculated the average Q as a function of time ( representative trace in Fig 5A ) . The average was taken over all simulations in a given T and in each condition ( bulk/confinement ) . Next , we fitted the smoothed Qaver ( t ) with the sum of exponential functions . In particular , we fitted the trace with the sum of the highest number of exponents , for which the fitting errors were lower than 5% of the value . The details of the plot along with the values of obtained constants and errors are presented in S1 Appendix , part 6 . Although Qaver ( t ) is only a rough equivalent for the fluorescence , in almost all cases we were able to decompose the trace as a sum of 2-4 exponentials , i . e . to find up to 2 fast and 2 slow phases . These data are shown in Fig 5B ( bulk ) and 5C ( confinement ) in the form of chevron plot with the inverse of temperature ( precisely −ϵ/kbT ) mimicking the denaturant concentration [53] . For the consistency with conventional chevron plot , we plot the logarithm of k = 1/τ where τ is a characteristic time of a given phase . The obtained values create the trends characteristic for chevron plots , therefore they were connected by dashed lines . Note that in some cases the connection of values is arbitrary . For most cases , the fitting error was an order of magnitude smaller than the value obtained ( Fig 5 ) , and those , for which the error was higher ( e . g . the point for confinement , −ϵ/kbT = −1 . 10 ) , still corresponds to reasonable values . The presence of the fast and slow phase during folding/unfolding process shows that our results are clearly consistent with previous experimental observations [30 , 52] . However , in most cases , we were not able to determine four individual phases ( as in the experiment ) . This may be due to similar characteristic times of separate phases , the model imperfection , or because of a much more complicated folding pathway . Indeed , the curvature of the limbs of the chevron plot indicates more complicated mechanism in each phase , again consistently with the experiment [33] . The detailed analysis of folding/unfolding pathway should be the next step in investigating of these proteins . To determine the influence of the confinement , we compared the “most complete” kinetic trace for the slow and the fast phases for bulk and confinement ( Fig 5D ) . The slowest phases can be fitted to an equation describing chevron plot , which yields an approximate Tf equal 114 ( −ϵ/kbT = −1 . 05 ) for bulk and 120 ( −ϵ/kbT = −0 . 99 ) for the confinement ( for details see S1 Appendix , part 6 ) . This indicates that confinement stabilizes UCH as it was observed for proteins with trivial topology [44 , 54] . Moreover , the chevron plot indicates that the confinement significantly accelerates the folding process , especially the slow phase . In particular , the simulations in the Tf for bulk were not accessible computationally due to very slow rates , while they were accessible in Tf for the confinement . This enables us to calculate near-equilibrium F ( Q ) dependence for confinement , which in principle could give additional information on UCHs folding . However in this case , due to the complexity of the folding landscape , the standard ways of its representation do not reveal any new information ( S1 Appendix , part 7 ) . Regardless of the conditions , a collapse of the protein ( the first phase of folding ) occurs relatively fast , which stays in accordance with the experimental results , that the knotting ( occurring in our model for Q > 0 . 7 ) should be the rate-limiting step [36] . Therefore in our case , the fast phase corresponds to arriving at collapsed , non-knotted form ( first part of folding ) and the slow phase should correspond to knotting and subsequential reaching of the native structure . The impact of the confinement on the slow phase indicates that the confinement facilitates knotting by restricting the conformational space of the termini . On the other hand , the confinement slows down the unfolding process by slowing down the unknotting—note the change of order in the curves in Fig 5D . The lower unfolding rates in the confinement may be also a result of the retying during unfolding ( discussed in the next section ) . As it turns out that the once unfolded knot has a higher probability to retie in the confinement which results in higher knot stability and slower unfolding . Slow unfolding is again in agreement with intuition and experimental observation made for proteins with trivial topology [55] . To additionally investigate the influence of the confinement on both unfolded and folded state , we determined the average asphericity [56] parameter for bulk and confinement in both states . Again , the asphericity of folded state was comparable in both conditions , indicating , that the confinement does not influence the near-folded structures significantly . On the other hand , the asphericity of unfolded state was different in the confinement than in bulk , showing the influence of confinement on the unfolded basin ( S1 Appendix , part 8 ) . The probability of knot presence in a polymer chain increases rapidly with its length . As a result , it is highly probable that the sufficiently long polymer will spontaneously form a knot . However , the fraction of knotted proteins is far lower than in the case of equally long polymers [57] . Moreover , the spontaneous self-tying of protein chains in the denaturated state was not reported so far even in the natively deeply-knotted structures [22 , 58 , 59] or in the case of small knotted proteins in confinement [29] . However , in the case of UCHs we observe a significant fraction of ( in most cases short-lived ) knots , appearing during folding/unfolding pathway , or in the denaturated state ( with Q < 0 . 2 ) . We found that confinement leads to faster and more efficient folding of UCH proteins for two reasons . First , encapsulation provides the possibility to fold via an alternative pathway . More precisely , the confinement facilitates folding via the trefoil knot ( the FC pathway ) for entropic reasons , while it does not affect folding via the FN pathway ( direct tying , the N-terminus folds the last ) . This surprising behavior is supported by the experimental observation of uneven influence of chaperonin [63] on the substrate protein rhodanese , which decelerates the folding of the C-terminal domain , but leaves the folding rate of the N-terminal domain unaffected [64] . Second , at the same confinement stabilizes native interactions and destabilizes non-native ones in comparison to bulk , and thus it reduces the height of the free energy barrier and accelerates the folding rate as it was observed for a protein with trivial topology [55 , 65 , 66] . The two pathways clearly distinguishable in our analysis are in accordance with two pathways identified in experiment [30] . However , there is still no technique , which could determine the topology of the protein during folding in vitro , which prevents from direct validation of our results . Some insights can however be given by the study of the tryptophan variants of UCH [33] . In particular , it was shown that the pathways differ in the structure of intermediates for which highly stable central β-sheet core and flanking α-helices and loop regions are formed differently . This is in accordance with our results , however detailed analysis of folding pathway , with special emphasis put on the location of tryptophan mutations , is required to precisely compare the experimental and theoretical results . On the other hand , our results show that the confinement introduced by the chaperone-like cage decelerates the unfolding of UCH proteins . Firstly due to a decrease in the effective mobility of the protein backbone upon encapsulation and under topological constrains , which reduce the rate at which new configuration can be explored ( especially in twisted loops ) . This argument without topological contribution is used to explain lower folding rate for a protein with the trivial topology [64] . Secondly , the decelerated unfolding in the confinement is caused by the retying phenomena . It is worth pointing out that the retying phenomena can be used by other knotted proteins with a rather shallow knot , e . g . carbonic anhydrase , to stabilize the structure in a crowded environment ( moreover carbonic anhydrase structures with deeper knots also start to be crystallized ) . Under the confinement , a significant number of short living knots is observed in the denatured state and in folding and unfolding routes of UCH proteins , what has not been reported for other knotted proteins . More and more complicated knots are more common to occur upon encapsulation , which is in the agreement with the polymer theory [61 , 62] . These knots seem to have only positive effect , i . e . their formation accelerates folding . In principle , deeply knotted structures could lead to misfolding , but contrary to the situation in the bulk , they are not formed due to the constrained configurational space . In summary , we took advantages of structure based model and knot theory , and made the step forward in characterizing folding/unfolding routes for UCH proteins identified experimentally in [30] . We identified possible oligomerization-prone forms of UCHs , which may cause neurodegenerative diseases . We found that weak confinement smooths the rough and not continuous free energy landscape of UCH proteins in a subtle way , e . g . enhancing an indirect tying route . However , at low temperature or strong confinement slower folding should be again observed due to restriction on indirect tying . The deceleration under strong confinement was suggested for a protein with the trivial topology in [63] . | Self-tying of knotted proteins remains a challenge both for theoreticians and experimentalist . In this work , we study the proteins with complex , the 52 knot , in a bulk and confined within a chaperonin box . We show that in our model we recreate the experimental results , identify two topologically distinct folding pathways and explain the beneficial role of confinement for complex knotted proteins . Encapsulation provides a possibility to fold via alternative pathway—folding via trefoil intermediate knot ( N-terminal pathway ) from entropic reason while folding via the C-terminal ( direct tying ) appears with the same probability . The results of this work show , how crowded environment in the real cell may enhance self-tying of proteins . The results are also the first step to the identification of possible oligomerization-prone forms of UCHs , which may cause neurodegenerative diseases . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results"
]
| [
"chemical",
"compounds",
"dna-binding",
"proteins",
"organic",
"compounds",
"materials",
"science",
"protein",
"structure",
"amino",
"acids",
"macromolecules",
"materials",
"by",
"structure",
"research",
"and",
"analysis",
"methods",
"polymers",
"polymer",
"chemistry",
"aromatic",
"amino",
"acids",
"protein",
"structure",
"determination",
"proteins",
"biological",
"databases",
"chemistry",
"proteomics",
"molecular",
"biology",
"protein",
"structure",
"comparison",
"biochemistry",
"biochemical",
"simulations",
"proteomic",
"databases",
"organic",
"chemistry",
"database",
"and",
"informatics",
"methods",
"tryptophan",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"macromolecular",
"structure",
"analysis"
]
| 2018 | The exclusive effects of chaperonin on the behavior of proteins with 52 knot |
The establishment and maintenance of Epstein-Barr Virus ( EBV ) latent infection requires distinct viral gene expression programs . These gene expression programs , termed latency types , are determined largely by promoter selection , and controlled through the interplay between cell-type specific transcription factors , chromatin structure , and epigenetic modifications . We used a genome-wide chromatin-immunoprecipitation ( ChIP ) assay to identify epigenetic modifications that correlate with different latency types . We found that the chromatin insulator protein CTCF binds at several key regulatory nodes in the EBV genome and may compartmentalize epigenetic modifications across the viral genome . Highly enriched CTCF binding sites were identified at the promoter regions upstream of Cp , Wp , EBERs , and Qp . Since Qp is essential for long-term maintenance of viral genomes in type I latency and epithelial cell infections , we focused on the role of CTCF in regulating Qp . Purified CTCF bound ∼40 bp upstream of the EBNA1 binding sites located at +10 bp relative to the transcriptional initiation site at Qp . Mutagenesis of the CTCF binding site in EBV bacmids resulted in a decrease in the recovery of stable hygromycin-resistant episomes in 293 cells . EBV lacking the Qp CTCF site showed a decrease in Qp transcription initiation and a corresponding increase in Cp and Fp promoter utilization at 8 weeks post-transfection . However , by 16 weeks post-transfection , bacmids lacking CTCF sites had no detectable Qp transcription and showed high levels of histone H3 K9 methylation and CpG DNA methylation at the Qp initiation site . These findings provide direct genetic evidence that CTCF functions as a chromatin insulator that prevents the promiscuous transcription of surrounding genes and blocks the epigenetic silencing of an essential promoter , Qp , during EBV latent infection .
Epstein-Barr Virus ( EBV ) is a human gamma herpesvirus that establishes latent infection in more than 90% of the adult population world-wide [1] , [2] . The ∼170 kb genome encodes ∼90 viral genes but only a few of these are expressed during latent infection . The latent infection is a cofactor in several human malignancies and may play an essential causative role in the endemic forms of Burkitt's lymphoma ( BL ) and nasopharyngeal carcinoma ( NPC ) , as well as diffuse B-cell lymphomas in HIV-AIDS and iatrogenic immunosuppressed individuals [3] . Remarkably , the viral gene expression patterns vary in each tumor type suggesting that EBV can establish multiple forms of latency [4] . These different gene expression programs have been referred to as latency types and may also correlate with the changes in host-cell differentiation state and tissue origin [4] , [5] . Changes in EBV latency type may also be important for evasion of host-immune recognition [6] . EBV latency gene expression programs have been categorized into four different types based primarily on the differential expression of the EBNA and LMP gene transcripts [4] . Type 0 latency is defined as the absence of expression of any viral genes , and is thought to exist in quiescent memory B-cells [5] , [7] . Type I latency is characterized by the expression of the EBNA1 gene only , and is observed in proliferating memory B-cells in normal hosts , and found predominantly in Burkitt lymphoma tissue and derived cell lines [8] , [9] , [10] . Type II latency is characterized by the expression of EBNA1 and LMP2 expression , with some variable expression of LMP1 . This pattern of gene expression is observed in epithelial cell derived tumors including NPC and gastric carcinomas [11] , [12] , [13] . Type III latency is characterized by the expression of EBNA-1 , -2 , -3A , -3B , -3C , -LP , LMP1 , LMP2 . This more permissive gene expression program is observed upon primary infection of naïve B-cells and is associated with B-cell proliferation and immortalization [14] . Type III latency is observed in immortalized B-cells in culture and diffuse B-cell lymphomas in immunosuppressed individuals . The natural history of EBV infection suggests that type III latency progresses to type I latency during B-cell maturation , and that viral lytic occurs in terminally differentiating plasma B-cells [5] , [15] . Latency type gene expression is regulated largely through differential promoter utilization [16] , [17] , [18] , [19] , [20] . The promoters for type III infection are activated upon primary infection by B-cell specific factors [21] . Initial transcription from Wp promoter allows the expression of EBNA2 , which functions as a transcriptional activator or the Cp promoter which drives expression of EBNA-LP , EBNA-2 , EBNA-3A , -3B , -3C and EBNA1 [22] , [23] , [24] , [25] . EBNA2 also activates LMP1 and LMP2 transcription to maintain type III gene expression [25] . Type II and Type I gene expression arise through mechanisms that are not completely understood , but involve the epigenetic silencing of the Cp and LMP1 promoter by DNA methylation and histone deacetylation [26] , [27] , [28] , [29] , [30] , [31] , [32] , [33] . EBNA1 expression is required for the replication and maintenance of the viral latent genome [34] , [35] , [36] , [37] . EBNA1 mRNA expression is maintained in type III latency by Cp promoter utilization and mRNA processing , while in type I latency EBNA1 expression is driven largely through the Q promoter ( Qp ) [17] , [31] , [38] , [39] , [40] . EBNA1 protein binds to two sites located at the +10–+57 position relative to the Qp transcription initiation site and restrict its usage in type III cells where EBNA1 proteins levels are elevated [41] , [42] . Thus , EBNA1 can autoregulate its own expression levels through promoter selection , and help to coordinate the switch between latency types . The epigenetic control of EBV promoter utilization and latency type is evident in the differential pattern of DNA methylation between latency types , and by the ability of DNA methylation inhibitors to stimulate type III gene transcription from type I latently infected cells [26] , [43] . It is also apparent that histone acetylation occurs at the Cp promoter during type III latency where they are transcribed , but not in type I latency where Cp promoter is silenced [44] . However , little else is known about the epigenetic controls the determine promoter utilization and gene expression during the different latency types . LMP1 and Cp activation depend on the enhancer functions of EBNA2 and EBNA1 . EBNA1 binds to the EBV origin of plasmid ( OriP ) replication and is essential for both viral replication and plasmid maintenance , as well as for transcriptional enhancement of EBNA2 and LMP1 [45] , [46] , [47] . The mechanism through which EBNA1 activates Cp and LMP1 from OriP , which is located over 2 kb from each promoter is not clear . It is also not known whether EBNA1 binding at Qp may also regulate transcription of type III promoters . Cellular factors that regulate communication between promoters and enhancers , have also been implicated in the organization of chromatin structure [48] , [49] , [50] , [51] . The chromatin insulator protein CTCF has been implicated in segregating active from inactive chromatin domains , as well as in mediating long-distance interactions between transcriptional regulatory regions [52] , [53] . At least one CTCF site has been mapped to a region between OriP and Cp , and its binding was found to correlate with the inhibition of Cp transcription in type I latency [54] , [55] . Other CTCF sites in EBV chromosome are known to exist , but their function has not been explored in detail [54] . In this work , we use a genome-wide ChIP assays to explore the epigenetic landscape of the EBV genome in type I and type III infected cells . We found that CTCF binding sites are positioned in key regulatory locations throughout the viral genome . We investigate in detail the function of a high affinity CTCF site positioned immediately upstream of the EBNA1 binding sites in Qp . We find that CTCF binding is required to maintain the transcriptional activity and prevent the epigenetic silencing of Qp in proliferating cells .
An EBV genome-wide real-time PCR array was used to compare the patterns of several epigenetic marks between type I and type III latent virus genomes using the chromatin immunoprecipitation ( ChIP ) assay ( Fig . 1 ) . For these experiments we used a 384-well array that covers the entire EBV genome at a density of ∼400 bp between primer sets . In a previous study , we used a similar approach to analyze the first 60 kb of the EBV genome for various histone modifications and protein factor binding sites [54] . In the present study , we compared a type I latently infected Burkitt lymphoma cell line , Mutu I , with a lymphoblastoid cell line derived from Mutu I viral DNA ( Mutu-LCL ) , ensuring that these two cell types were isogenic with respect to EBV genomes . We compared the pattern of CTCF binding sites to those of histone H3me3K9 ( H3mK9 ) and H3me2K4 ( H3mK4 ) modifications and also that of cytosine methylation ( mCpG ) using methyl DNA immunoprecipitation ( MeDIP ) . Several patterns were noteable . Peaks of histone H3mK9 and H3mK4 were complementary and non-overlapping . Major peaks of H3mK4 were detected at the regions surrounding the RNA polymerase III transcribed EBERS and the Bam A microRNA cluster in both cell types . H3mK4 peaks regions surrounding the LMP1 promoter and Cp promoter were elevated in type III relative to type I cells . H3mK9 methylation was elevated over the Cp and W repeats in type I latency , but not type III , correlating with transcription repression in type I . CTCF binding sites were located at multiple positions , with only a few differences in type I and type III . CTCF sites tended to exist between clusters of H3mK4 and H3mK9 , as can be seen at a newly discovered peak at the region 3′ of the W repeats . All the CTCF binding sites found in our assay are listed in Table S1 . mCpG patterns were also different between type I and III latency . High levels of mCpG were detected across the Cp and LMP2 regions in type I , but not in type III , correlating with transcription silencing of Cp/Wp in type I cells . Interestingly , CTCF sites tended to demarcate boundaries of mCpG , which also correlated well with H3mK9 in both cell types . A more detailed examination of the regions surrounding the major latency promoter elements reveals other features relevant to epigenetic regulation ( Fig . 2 ) . While most CTCF sites are bound similarly in each cell type , the CTCF site upstream of the EBERS was significantly reduced in Mutu I cells ( Fig . 2A ) . Interestingly , this region is enriched in mCpG in MutuI , relative to type III cells ( Fig . 2B ) . One possible explanation for the loss of CTCF binding at this region in type I cells is that enriched DNA methylation replaces and blocks CTCF binding . In contrast , this same region is elevated in H3mK9 in type III latency , where CTCF occupies a 3′ boundary upstream of the EBERS ( Fig . 2D and 2J ) . Another striking feature is the elevated H3mK4 across the EBNA2 transcript and BHRF1 miRNA cluster in type III latency , but not in type I ( Fig . 2C ) . This correlates well with the difference in RNA polymerase II transcription across this region in these two cell types . The CTCF site at this position just 3′ of W repeats provides a 5′ border for the high H3mK4 in type III latency ( Fig . 2I ) , and a 3′ border for the high H3mK9 ( Fig . 2E ) and mCpG ( Fig . 2G ) in type I latency . CTCF binds consistently at Qp in both cell types ( Fig . 2A ) . In type I cells , the Qp CTCF site separates a 3′ H3mK4 peak from a 5′ region enriched in H3mK9 and mCpG ( Fig . 2E–G ) . In type III cells , the CTCF site appears to spare Qp from surrounding regions of elevated mCpG and H3 mK9 ( Fig . 2H and J ) . The region is also reduced in H3mK4 , corresponding to a reduction in EBNA1 binding and transcription from Qp ( Fig . 2I ) . These marks are largely consistent with known transcription properties of Qp in which it is active in type I , and repressed in type III ( Figure S1 ) . To identify the specific sequence element bound by CTCF near Qp , we first examined the region for candidate CTCF binding sites using a prediction algorithm ( http://insulatordb . uthsc . edu ) . At least two candidate CTCF sites were identified at positions 43739 and 50082 ( Fig . 3A ) . These sites were synthesized as DNA oligonucleotide probes and tested by EMSA for binding to purified recombinant CTCF protein ( Fig . 3B and C ) . We found that CTCF bound efficiently to the 50082 binding site , but not to the 49739 sequence or to a control oligonucleotide ( from EBV 49901 ) that lacked any candidate CTCF binding site ( Fig . 3C ) . The precise nucleotide binding site of CTCF was mapped by DNase I footprinting using the entire Qp control region as a probe and purified recombinant CTCF ( Fig . 3D ) . We found that CTCF protected a ∼20 bp regions between 50082–50102 . In the same DNase footprinting reaction we included recombinant EBNA1 protein . EBNA1 bound to two sites in Qp covering EBV nucleotides 50142 to 50189 . The DNase I footprinting assays demonstrate that CTCF and EBNA1 can bind simultaneously to Qp , and that potential interactions between these proteins may regulate Qp . However , detailed biochemical analysis of these interactions were limited by the different salt sensitivities of the purified proteins in the DNA binding assays ( Fig . 3D ) . The physiological significance of this differential salt sensitivity is not clear . To investigate the functional significance of CTCF binding at Qp in cell-based assays , we engineered a substitution mutation in the CTCF binding site at Qp in EBV bacmids using recombineering with GALK gene insertion and gene replacement [56] ( http://recombineering . ncifcrf . gov ) ( Fig . 4A ) . GALK insertion , CTCF substitution mutation ( ΔCTCF ) , and Wild-type ( Wt ) rescue mutants in EBV bacmids were validated by restriction enzyme digestion ( Fig . 4B ) , PCR across the junctions ( Fig . 4C ) and sequencing of the insertions ( data not shown ) . Bacmid DNA for ΔCTCF and Wt rescue control was introduced into 293 cells and stable transformants were selected for hygromycin and GFP expression . After 8 weeks of selection , stable cell pools were assayed by ChIP assay to validate that the substitution mutation disrupted CTCF binding in living cells ( Fig . 4D ) . As expected , CTCF failed to bind to Qp in the ΔCTCF mutant ( Fig . 4D , top panel ) . We also found that EBNA1 binding to Qp was reduced in the ΔCTCF mutant ( Fig . 4D , lower panel ) , suggesting that CTCF facilitates EBNA1 binding at Qp in living cells . Stable cell pools were also assayed at 8 weeks for their relative expression GFP , EBNA1 , CTCF and PCNA ( Fig . 4E ) . We found no apparent differences in the expression of these proteins after 8 weeks of selection in 293 cell pools . While early passage cell pools showed little difference in GFP and EBNA1 expression between ΔCTCF and Wt rescue genomes , we observed a marked loss of GFP expression in the ΔCTCF relative to Wt rescue genomes after longer passages in culture ( Fig . 5A and B ) . At 4 and 8 weeks after transfection , ΔCTCF and Wt rescue had nearly identical percentage of GFP positive cells . In contrast , at 16 weeks ΔCTCF pools were ∼9% positive , while Wt rescue was ∼72% GFP positive , as measured by FACS ( Fig . 5B ) . EBV DNA copy number per cell was determined by real time PCR for 293 cell pools at 8 and 16 weeks ( Fig . 5C ) . We found that ΔCTCF cell pools had ∼50% less EBV DNA per cell than Wt rescue containing pools as measured at 8 and 16 weeks . Isolation of EBV episomes by Hirt lysis revealed that ΔCTCF cell pools had ∼3 fold lower DNA than WT rescue at 8 weeks , and both pools were reduced at 16 weeks , indicating that ΔCTCF episomes are lost at a greater rate than Wt episomes ( Fig . 5D ) . These data suggest that the Qp CTCF binding site is important for the stable maintenance of EBV episomes in selected 293 cell pools . The loss of GFP expression and episome stability in 293 cell pools could be due to a deregulation of viral gene expression . Others have shown that EBV establishes a restricted pattern of latency gene expression in 293 cells , resembling a type I program with stable Qp utilization for EBNA1 expression [57] . To assess viral gene expression patterns in transfected 293 cells , we first assay mRNA expression of EBNA1 , EBNA2 , EBNA3A , and EBNA3C in 293 cell pools after 4 , 8 , and 16 weeks of selection ( Fig . 6 ) . RNA expression was measured by quantitative RT-PCR and normalized to bacmid expression of GFP mRNA . At 4 weeks , EBNA1 and EBNA2 mRNA levels were expressed at lower levels in ΔCTCF compared to Wt rescue cell pools ( Fig . 6C , top panel ) . At 8 weeks , EBNA1 levels were similar , while EBNA2 , EBNA3A , and EBNA3C levels were higher in ΔCTCF relative to Wt rescue cell pools ( Fig . 6C , middle panel ) . By 16 weeks , EBNA1 mRNA levels were maintained in the Wt rescue , but almost undetectable in ΔCTCF cell pools ( Fig . 6C , lower panel ) . EBNA2 , EBNA3A , and EBNA3C were expressed at very low levels in both Wt rescue and ΔCTCF cell pools at these later passages in culture . These observations are consistent with a previous study showing that EBV initially expresses EBNA2 , but eventually adopts an EBNA1 only , type I latency in 293 cells [57] . To better understand the failure of ΔCTCF bacmids to sustain EBNA1 mRNA expression , we investigated the promoter utilization at 4 , 8 , and 16 weeks after transfection ( Fig . 6D ) . EBNA1 mRNA has been shown to initiate from Wp/Cp in most type III latency , from Qp in most type I latency , and from Fp during lytic cycle gene expression ( Fig . 6A and B ) . We assayed the utilization of Wp/Cp , Qp , and Fp using quantitative RT-PCR with primers specific for each promoter ( Table S2 ) . At 4 weeks , we found that Wp/Cp was utilized at similar levels in Wt rescue and ΔCTCF containing cell pools ( Fig . 6D , top panel ) . Remarkably , Qp was utilized at relatively high levels in Wt rescue , but nearly undetectable in ΔCTCF . Interestingly , Fp utilization was detected in ΔCTCF , but undetectable in Wt rescue . At 8 weeks , Wt rescue containing cells utilized Qp predominantly , while ΔCTCF cells utilized all three promoters , with Fp dominating ( Fig . 6D , middle panel ) . By 16 weeks , Wt rescue genomes utilized Qp exclusively , while ΔCTCF genomes utilized Fp exclusively , although ∼ 10 fold less than Fp utilization at 8 weeks ( Fig . 6D , lower panel ) . Since Fp is typically associated with lytic gene activity , we tested whether ΔCTCF containing cells were expressing the lytic immediate early gene BZLF1 ( Fig . S3 and Fig . 6E ) . BZLF1 expression was undetectable at all time points tested for Wt or ΔCTCF 293 cell pools , as measured by quantitative RT-PCR ( Fig . S3 ) or by conventional PCR ( Fig . 6E ) . These findings suggest that latency promoter utilization is deregulated in ΔCTCF genomes . To better understand the ΔCTCF defects in viral gene expression , we assayed viral mRNA using exon specific PCR for transcripts ( Table S3 ) , initiating at Qp ( QUK ) , Cp ( C1C2W1W2 ) , Wp ( W0W1W2 ) , or within the lytic transcripts of BFLF1 [58] ( Fig . 6E ) . In addition , we measured the UK intron junction for EBNA1 , which is expressed in type I and type III cells . We also measure BZLF1 expression as an indicator of lytic cycle gene expression . As a control we assayed the expression levels of cellular GAPDH . RNA was isolated from pools after 8 weeks in culture . We found that QUK was expressed at slightly higher levels in Wt rescue relative to ΔCTCF cell pools ( Fig . 6E , top panel ) . C1C2W1W2 , W0W1W2 and FQUK transcripts were elevated in ΔCTCF relative to Wt rescue , consistent with RT-PCR data showing elevated levels of Wp/Cp and Fp utilization in bacmid lacking the CTCF binding site . UK transcripts were similar in ΔCTCF and Wt rescue , consistent with observations from quantitative RT-PCR ( Fig . 6C ) showing that EBNA1 mRNA levels were similar at 8 weeks after transfection . BZFL1 was not detected in either bacmid 293 cell pool , indicating that lytic gene activation or DNA replication was not indirectly responsible for these differences in gene expression . These exon-specific PCR studies further substantiate the real-time PCR data , and support the conclusion that CTCF mutations in Qp deregulate the latency type transcription pattern and promoter utilization . CTCF has been implicated in several functions , including chromatin insulation and boundary functions . To determine if the disruption of CTCF site at Qp altered the normal pattern of epigenetic marks surrounding Qp , we performed MeDIP and ChIP assays with a set of primers that probe the regions −1000 , −500 , +1 , and +800 relative to the Qp initiation site ( Fig . 7A–C ) . At 8 weeks post-transfection , MeDIP revealed that mCpG was enriched at −1000 and −500 position in Wt rescue , but undetectable in ΔCTCF genomes ( Fig . 7A , top panel ) . At 16 weeks post-transfection , mCpG was elevated at −1000 in WT rescue , but not at +1 or +800 , consistent with high levels of transcription initiating at Qp in these cell pools ( Fig . 6 ) . In contrast , mCpG was elevated at +1 position in ΔCTCF genomes , consistent with the lack of Qp transcription initiation at 16 weeks in ΔCTCF cells ( Fig . 6 ) . To examine potential changes in euchromatic or heterochromatic histone modifications surrounding Qp , we focused on H3mK4 or H3mK9 , respectively . At 8 weeks post-transfection , we found high levels of H3mK4 at the +1 and +800 positions in both Wt rescue and ΔCTCF genomes ( Fig . 7B , top panel ) . At 16 weeks , H3mK4 was elevated primarily in Wt rescue ( Fig . 7B , lower panel ) , consistent with persistent transcription from Qp in these cells ( Fig . 6 ) . The heterochromatic mark for histone H3mK9 was more revealing since it showed elevated levels upstream of Qp ( −1000 and −500 ) in Wt rescue genomes at 8 weeks ( Fig . 7C , top panel ) , and low levels in ΔCTCF where Fp is active . At 16 weeks , H3mK9 was highly enriched at +1 site of Qp in ΔCTCF genomes ( Fig . 7C , lower panel ) , consistent with the transcription inactivity of Qp in these cells at this time after transfection . The epigenetic pattern surrounding Cp and Wp was also examined ( Fig . S5 ) . We found that Cp and Wp were hypermethylated and enriched on H3mK9 both at 8 and 16 weeks in Wt rescue , consistent with a previous study demonstrating that Cp and Wp undergo transcription silencing in 293 cells [57] . In ΔCTCF genomes Cp and Wp are hypomethylated and enriched on H3mK4 at 8 weeks , consistent with transcription initiation . However , by 16 weeks , Cp and Wp show elevated mCpG and H3mK9 , consistent with the extinction of transcription initiation from these promoters at later passages . Thus , Cp and Wp undergo similar epigenetic silencing with Wt and ΔCTCF genomes , but Qp is protected from epigenetic silencing only in genomes were the CTCF site is intact ( Fig . 7D ) .
Stable gene expression programs , like those associated with cell-type differentiation , correlate with heritable epigenetic changes to the cellular chromosome . The meta-stable gene expression programs associated with Epstein-Barr virus latency types have also been shown to correlate with epigenetic changes in the viral genome . The most consistently observed epigenetic difference between type I and type III latency is the DNA methylation of the Cp and LMP1 promoter regions [26] , [43] , [59] . We and others have also explored histone modification patterns at the major latency control regions for EBV , and observed distinct patterns between cell lines carrying EBV genomes with either type I and type III latency [44] , [54] , [60] . In this work , we extended this approach to examine the pattern of histone H3 K4 and K9 methylation , DNA methylation , and CTCF binding across the complete EBV genome in cells carrying stable type I or type III EBV latent infections ( Figs 1 and 2 , Figs . S1 and S2 ) . We found that CTCF binding sites were located at or near regulatory regions , and commonly marked boundaries between euchromatic and heterochromatic marks ( Fig . 2 ) . The heterochromatic mark was typically H3mK9 in type III latency ( e . g . EBV regions 2–6 kb in Fig . 2D ) , and mCpG , especially in type I latency ( e . g . EBV regions 2–6 kb in Fig . 2B ) . This may reflect the natural history of epigenetic silencing where a partially repressive histone modification , like H3mK9 in type III , may eventually evolve into the more stable silencing modification associated with DNA methylation in type I . In some cases , the emergence of extensive CpG methylation correlates with the loss of CTCF binding ( e . g . EBV region ∼6 kb in Fig . 2E and 2H ) . CTCF binding is known to be sensitive to CpG methylation [61] . These observations suggest that CTCF plays a key role in organizing epigenetic marks along the EBV genome , and that CTCF binding and epigenetic patterns change in different latency types . To directly test the function of CTCF and other sequence specific binding factors in maintaining EBV latency type gene expression programs , we focused on the EBV Q promoter . We found that CTCF bound to Qp , and we used DNase I footprinting to map this binding site to a ∼20 bp region located ∼40 bp upstream of the EBNA1 binding sites ( Fig . 3 ) . In type I cells , CTCF appeared to separate a 3′ enrichment of H3mK4 that overlaps the Q transcription initiation site , from a 5′ enrichment of CpG methylation , that covers the neighboring lytic BFLF1 gene ( Fig . 2E–G , Fig . S1 ) . In type III cells , CTCF appears to spare the Qp transcription initiation site from surrounding H3 K9 methylation ( Fig . 2I ) . To directly test the function of CTCF binding at Qp , we used recombineering methods to engineer a mutation that disrupts CTCF binding in EBV bacmids ( Fig . 4 ) . Disruption of CTCF binding site at Qp caused a loss of stable GFP expression and loss of bacmid episomes after multiple cell divisions ( Fig . 5 ) . CTCF site disruption also caused an increase in Fp promoter utilization , with no other evidence of lytic gene activation ( Fig . 6C–E ) . Consistent with changes in gene expression , CTCF site disruption allowed for the formation of mCpG and H3mK9 methylation at the Qp initiation site ( Fig . 7 ) . These finding strongly suggest that CTCF contributes to the establishment and maintenance of an epigenetic pattern at Qp which is required for consistent expression of EBNA1 and episomal persistence in 293 cells pools . These findings also suggest that CTCF provides a barrier function that normally prevents Fp activation ( upstream ) and Qp silencing ( downstream ) during latent infection ( Fig . 7D ) . CTCF has been implicated in several gene regulatory and chromatin organizing activities [52] , [53] . At the H19/Igf2 imprinted loci , CTCF functions as an enhancer blocker [62] . At the paternal allele , DNA methylation prevents CTCF binding , and allows enhancer activation of the Igf2 promoter . In contrast , at the unmethylated maternal allele , CTCF binds to a cluster of sites and prevents enhancer activation of the Igf2 promoter . Our findings here and previously [55] suggest that CTCF may have enhancer blocking activity at the sites surrounding OriP ( e . g . the 5′ site at ∼6 kb and the 3′ site at 10 kb on the EBV genome ) . These sites are positioned to physically block OriP interactions with LMP1/2 and Cp , respectively . Interestingly , we found that CTCF binding at the 5′ site ( ∼6 kb on EBV ) is reduced in type I latency where the sites have been partially subject to CpG DNA methylation . Thus , latency type-specific DNA methylation patterns may regulate CTCF binding at some regulatory regions . Conversely , CTCF binding may prevent DNA methylation at other regulatory elements . The region surrounding the Qp initiation site has been shown to lack CpG methylation in all latency types [43] . The CTCF site we identified may function to prevent the spread of CpG methylation which is normally elevated in the regions upstream of Qp ( BFLF1 ORF ) in both type I and type III latency . Disruption of the CTCF site caused a significant increase in CpG methylation immediately over the Qp initiation site ( Fig . 7A ) . Thus , an essential function of CTCF may be to prevent CpG methylation at the Qp initiation site . This is consistent with the observation that CpG methylation is never detected at Qp in any EBV latency type [43] . Chromatin boundary factors , like CTCF , are thought to prevent the spread of processive histone modifications [48] . Our data suggests that CTCF has chromatin boundary activity at Qp that prevents inactivating heterochromatin , like H3mK9 from invading Qp . Elevation in H3mK9 at Qp was evident at 4 and 8 weeks post-transfection in ΔCTCF , suggesting that this mark is the first to cross into the Qp promoter region . At 16 weeks , mCpG is highly elevated at Qp , thus following H3mK9 and more completely silencing Qp transcription . Remarkably , the CTCF boundary functions in the reciprocal direction since the loss of CTCF leads to elevated transcription of Fp and BFLF1 , along with a corresponding increase in histone H3mK4 , and decrease in the normal H3mK9 and CpG methylation . It remains unclear what drives each modification at the specific sites . Our data strongly suggests that H3mK9 precedes the formation of mCpG , and that CTCF helps keep each modification in its proper place . Precisely what sets the pattern on each side of CTCF is not clear . CTCF may function in the recruitment of RNA polymerase II to Qp , and this may help to establish the correct orientation of the chromatin boundary at Qp . Loss of this boundary function leads to the loss of Qp transcription and the inappropriate regulation of other viral genes . CTCF also affected EBNA1 binding to Qp in vivo ( Fig . 4D ) , raising the possibility that EBNA1 may also contribute to some of these changes in chromatin and transcription at Qp . Furthermore , it is not known if CTCF or EBNA1 provide additional structural features that directly affect these other promoters , or if these are indirect effects of altering EBNA1 levels in the 293 cell model of latency DNA methylation , like histone modifications , can also spread across chromosomal regions to alter gene expression programs [63] . Epigenetic silencing due to promoter CpG methylation commonly arises at sights that have been actively repressed by histone deacetylation and K9 trimethylation . The mechanisms that restrict the drift of CpG methylation have not been completely elucidated . Our findings provide clear genetic evidence that CTCF can prevent the encroachment of CpG methylation at the Qp promoter of EBV . Our study shows that CpG methylation arises only after multiple generations at a region that is initially euchromatic and transcriptionally active . In the absence of CTCF , transcription initiation favors an alternative upstream promoter , Fp , which may prevent Qp utilization . Thus , CTCF may also facilitate promoter selection perhaps through its reported ability to interact with RNA polymerase II [64] . While the precise mechanism through which CTCF directs promoter selection and maintains chromatin boundaries remains to be discovered , our findings clearly indicate that CTCF provides an essential function in maintaining the epigenetic patterns at Qp . Our findings also indicate that protection of Qp by CTCF is essential for EBV genome stability during long-term latent infection . These findings also provide a framework for understanding the role of CTCF at other viral and cellular genes where protection from epigenetic drift and transcription silencing is critical for stable gene expression programs .
D98/HR1 , HeLa and 293 cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and antibiotics in a 5% CO2 incubator at 37°C . EBV positive Mutu I , Mutu-lymphoblastoid cell lines ( Mutu-LCL ) , Sav I and Sav III cells were cultured in suspension in RPMI 1640 medium supplemented with 10% fetal bovine serum and antibiotics in a 5% CO2 incubator at 37°C . Mutu-LCL were established by primary infection of peripheral blood mononuclear cells ( PBMCs ) with EBV virions derived from stimulated Mutu I cells . EBV bacmid was a generous gift of Dr . H . –J . Delecluse [65] . Mutations in EBV bacmid were generated by recombineering using the GalK marker gene insertion and negative selection method for its substitution as described previously ( http://recombineering . ncifcrf . gov/ ) [56] . The GalK gene was recombined into the Qp region at EBV coordinates 49927–50185 . The CTCF site at 50082 ggtcgctagatggcgcgggtgagg was mutated by single substitutions to ggtTgTtaTatTTTAcgggtgagg . The ΔEBNA1 binding site at 50142 gaaaag[gcgggatagcgtgcgctaccggatggcgggtaatacatgct]atccttaca was mutated by deletion and substitution with gaaag[tgcttgaaaaggcgcgg]atccttaca . Plasmid containing the Qp region ( 49712–50250 ) was subcloned by PCR into pBKSII using Asp718 and HinDIII sites . Recombinant human CTCF protein was expressed as an N-terninal hexa-histidine tagged fusion protein from a baculovirus expression virus in sf9 cells , as described previously[55] . ChIP assay followed the protocol provided by Upstate Biotechnology , Inc . , with minor modifications as previously described [54] . Additional modifications are as follows . DNAs were sonicated to between 200- and 350-bp DNA fragments on a Diagenode Bioruptor according to manufacturer's protocol , and real-time PCR was performed with SYBER green probe in an ABI Prism 7900 using 1/100 to 1/2 , 500 of the ChIP DNA according to manufacturer's specified parameters . Primer sequences for the EBV genome array are available upon request . Antibodies for H3 me2 K4 , H3 me3 K9 , CTCF , were purchased from Upstate Biotechnology . Primary antibodies to EBNA1 ( Advanced Biotechnologies , Inc . ) , CTCF ( Millipore ) , GFP ( Santa Cruz Biotecnology ) and PCNA ( Santa Cruz Biotecnology ) were used according to manufacturer's specifications EMSA assays with CTCF were described previously [55] . In a 20 µl reaction purified CTCF ( ∼100 ng ) was added to a reaction mixture containing 0 . 5 µg poly ( dI–dC ) , 5% glycerol , 0 . 1 mM ZnSO4 , and 10 , 000 cpm of 32P-labeled DNA probe ( ∼0 . 1 ng ) . Reaction mixtures were incubated for 30 min at 25°C , electrophoresed in a 5% nondenaturing , polyacrylamide gel at 110 V , and visualized by PhosphorImager . DNase I footprinting was performed as described previously [66] . 5′-end labeled DS probe was generated using 30 µCi of [−32P]dATP ( 6 , 000 Ci/mmol; Perkin-Elmer ) and 2 U of Klenow fragment ( Roche ) for 30 min at 25°C . Purified proteins were incubated in a reaction mixture containing 1XPBS , 5 mM MgCl2 , 0 . 1 mM ZnSO4 , 1 mM dithiothreitol , 0 . 1% NP-40 , 10% glycerol , 1 µg bovine serum albumin , 0 . 4 µg poly ( dI–dC ) , and 10 , 000 cpm of 32P-labeled probe . The protected probe was digested with different dilutions of DNase I ( Sigma ) and purified by phenol-chloroform extraction following proteinase K digestion . The DNA samples were then electrophoresed on a 7% denaturing , polyacrylamide sequencing gel at 33 mA and visualized by PhosphorImager . RNA was extract from 5×106 cells using Qiagen RNA extraction Kit according to manufacturer's protocol ( Qiagen ) . After the extraction the RNA was incubated with 2 U DNAse I at 37°C for 30 minutes , following by the inactivation of the enzyme at 65°C for 10 minutes . The RNA was quantified and 2 µg of RNA was reverse transcribed using Super Script II Reverse Transcriptase from Invitrogen . 50 ng of cDNA was then analyzed by real time or conventional PCR . Primer sequences used for real time and conventional PCR are listened in tables 2 and 3 ( Tables S2 and S3 ) . For EBV genome quantification Namalwa titration was used , assuming that Namalwa cell lines contain 2 copies of EBV genome and each cell contain 6 . 6×10−12 g of DNA . Namalwa cells and EBV positive cells were lysed in SDS lysis buffer ( 20 mM Tris pH 8 , 4 mM EDTA , 20 mM NaCl , 1% SDS ) following by incubation with Proteinase K for 2 h at 50°C . The DNA was extracted by phenol-chloroform extraction and precipitated by ethanol . Titration of 6 . 6×10−7–6 . 6×10−12 g of Namalwa DNA were used to obtain a calibration curve using primers for EBV genome ( 48779 – 48834 ) and β-Actin . 6 . 6×10−9 g of DNA from EBV positive cells were then analyzed by real time PCR for EBV genome copy quantification . Primer sequences used for real time PCR were listed in Tables S2 and S3 . 10×106 cells were resuspended in Lysis Buffer ( 20 mM Tris pH 8 , 4 mM EDTA , 20 mM NaCl , 1% SDS ) plus 0 . 7 µg/µl Proteinase K and incubated at 50°C overnight . The DNA was extracted by twice phenol-chloroform extraction followed by ethanol precipitation . The DNA was resuspended in 300 µl of TE buffer containing 20 µg/ml RNAse A and incubated at 37°C for 1 hr followed by DNA sonication to between 750 – 500 bp and DNA purification by phenol-chloroform extraction and ethanol precipitation . 8 µg of DNA were resuspended in Immunoprecipitation Buffer , IP buffer , [10 mM Na-Phosphate pH 7 , 140 mM NaCl , 0 . 05% Triton X-100 , and Proteinase inhibitor cocktail ( Sigma ) ] , denaturated at 95°C for 10 min and incubated with 5 µg of 5-methylCytodine antibody ( Abcam ) or 5 µg mouse IgG ( Upstate ) , overnight at 4°C . The immunocomplexes were precipited by adding 50 µl of Protein G Dynabeads ( Invitrogen ) for 2 hr at 4°C . The beads were collected by a magnetic rack and washed twice for 10 min with 1 ml of IP buffer . The DNA was eluted by incubating the beads with 250 µl of Proteinase digestion buffer ( 50 mM Tris pH 8 , 10 mM EDTA , 0 . 5% SDS , 0 . 3 µg/µl Proteinase K ) at 50°C for 3 hr with shacking . The DNA was then purified and then analyzed by real time PCR . Primer sequences are listed in Table S4 . 1 µg of genomic DNA was used as Input material . Viral episomal DNA was extracted using the Hirt lysis method [67] Briefly , Wt rescue and ΔCTCF 293 cells were pelleted by centrifugation and then resuspended in 800 µl Hirt's lysis buffer ( 0 . 6% SDS , 10 mM EDTA , 10 mM Tris-HCl , pH 7 . 4 ) containing 500 µg Proteinase K . The samples were incubated at 37°C for 1 h , followed by addition of 200 µl of 5 M NaCl and incubated at 37°C overnight . The samples were centrifuged at 14000 g for 20 minute and the supernatant was transferred in a new tube . DNA was then purified by phenol/chloroform/isoamyl alcohol extraction and ethanol precipitation . The DNA was resuspended in 50 µl of TE buffer and analyzed by real Time PCR by ΔΔCt method , using mitochondrial DNA as endogenous control and viral DNA extracted from Raji cell as a standard . | Epstein-Barr Virus ( EBV ) establishes a latent infection that is associated with several lymphoid and epithelial cell malignancies . The latent virus persists as a circular minichromosome in the nucleus of infected cells . Epigenetic modifications of the viral DNA and chromatin are known to control viral gene expression and genome stability , but the nature and mechanisms of these epigenetic marks are not known . Here , we use viral genome-wide analysis to characterize patterns of DNA and histone methylation , and how these are organized by the chromatin boundary factor CTCF . Mutation of one such CTCF site at the EBV Q promoter results in aberrant accumulation of DNA CpG methylation and histone H3 K9 trimethylation , and the consequent silencing of Qp transcription . We conclude that CTCF chromatin insulator function is required for the epigenetic programming and stable maintenance of latent viral infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biochemistry/transcription",
"and",
"translation",
"genetics",
"and",
"genomics/gene",
"function",
"virology/viruses",
"and",
"cancer"
]
| 2010 | CTCF Prevents the Epigenetic Drift of EBV Latency Promoter Qp |
Bacterial strains isolated from attine ants showed activity against the insect specialized fungal pathogen Escovopsis and also against the human protozoan parasite Leishmania donovani . The bioassay guided fractionation of extracts from cultures of Streptomyces sp . ICBG292 , isolated from the exoskeleton of Cyphomyrmex workers , led to the isolation of Mer-A2026B ( 1 ) , piericidin-A1 ( 2 ) and nigericin ( 3 ) . Nigericin ( 3 ) presented high activity against intracellular amastigotes of L . donovani ( IC50 0 . 129 ± 0 . 008 μM ) . Streptomyces puniceus ICBG378 , isolated from workers of Acromyrmex rugosus rugosus , produced dinactin ( 4 ) with potent anti-L . donovani activity against intracellular amastigotes ( IC50 0 . 018 ± 0 . 003 μM ) . Compounds 3 and 4 showed good selectivity indexes , 88 . 91 and 656 . 11 respectively , and were more active than positive control , miltefosine . Compounds 1–4 were also active against some Escovopsis strains . Compounds 1 and 2 were also produced by Streptomyces sp . ICBG233 , isolated from workers of Atta sexdens , and detected in ants’ extracts by mass spectrometry , suggesting they are produced in the natural environment as defensive compounds involved in the symbiotic interaction .
Leishmaniasis is designated as Neglected Tropical Diseases ( NTDs ) by the World Health Organization ( WHO ) . The visceral leishmaniasis is the most serious clinical form , produced by two Leishmania species , L . infantum and L . donovani [1] . There are between 50–90 thousands new cases and around 20–30 thousands deaths each year due to this form of leishmaniasis [2] . The treatment of leishmaniasis is still incomplete , since available drugs are toxic and expensive , have bioavailability issues , and need to overcome parasite resistance [3] . Miltefosine , originally launched as anticancer agent [4] , was the only drug approved against leishmaniasis between 1981 and 2014 [5] . Prospecting understudied sources of natural products can contribute to the discovery of new antiprotozoal pharmacophores . Streptomyces associated with insects have recently emerged as a prolific and underexplored source of antimicrobials [6] . In the quadripartite symbiosis in the fungus-growing ant ecosystem between three mutualists ( Attine ant , fungal garden and symbiotic actinomycetes ) and one parasite ( specialized pathogenic fungus Escovopsis sp . ) , some interspecies interactions are mediated by small molecules [7] . The ant associated actinobacteria produce secondary metabolites to inhibit the pathogen ( Escovopsis sp . ) but not the crop fungus ( phylum Basidiomycota ) [8] . This specific ecological function can guide the discovery of natural products potentially active against human pathogens [8] . Indeed , interesting bacterial-derived natural products have been reported with a wide spectrum of biological activities such as dentigerumycin [9] , 9-methoxyrebeccamycin [10] and selvamicin [11] . In an ongoing International Cooperative Biodiversity Group ( ICBG ) initiative [12] , we have isolated several actinobacteria strains from the exoskeleton of fungus-growing ants to prospect for antifungal and antiprotozoal compounds . There are some examples of compounds presenting both antifungal and antiprotozoal activities , such as azoles [13] and amphotericin B [14] . Therefore , bacterial symbionts of attine ants represent an underexplored ecosystem to search for antiprotozoal natural products based on their antifungal activity against Escovopsis in their niches .
RP HPLC was performed using a Shimadzu Prominence HPLC system and a Phenomenex Luna C6-Phenyl column ( 5μm , 250 x 10 mm ) . The mass spectrometry data for 2 and 3 were acquired with a Bruker MaXis Quadrupole Time-of-Flight MS coupled to a Waters Acquity UPLC system operated by Bruker Hystar software , and for 1 and 4 with an Accela UHPLC ( Thermo Scientific , USA ) apparatus with an 80 Hz photodiode array detector ( PDA ) coupled to a Q-Exactive Plus Orbitrap mass analyzer ( Thermo Scientific , USA ) . NMR spectra of 1–4 were obtained in CDCl3 with a Varian Unity-Inova 500 MHz spectrometer . The LC-HRESIMS and MS/MS spectra of organic extracts of Atta sexdens workers were acquired with a UPLC ( Shimadzu ) coupled to a micrOTOF II mass spectrometer ( Bruker Daltonics ) . About 1–10 grams of the fungal gardens of the attine ants’ colony were collected . Five ants from each colony were selected for isolation of actinobacteria . Ants were identified at genus level using genera identification keys [15–17] . Specialists in taxonomy identified respective species . Ten actinobacterial strains were isolated from Acromyrmex rugosus rugosus worker ants , eight strains were isolated from Cyphomyrmex workers and twelve strains from Atta sexdens soldier ants . The bacterium Streptomyces puniceus AB10 ( strain ICBG378 ) was isolated from A . rugosus rugosus ants collected at USP-Ribeirão Preto campus as previously described in Ortega et al . [18] . The bacterium Streptomyces sp . ICBG292 was isolated from the exoskeleton of Cyphomyrmex workers; and Streptomyces sp . ICBG233 from exoskeleton of A . sexdens workers . Cyphomyrmex and A . sexdens ants were collected in October of 2015 at the campus of the USP-Ribeirão Preto , as part of the ICBG-Brazil project [12] . Permits for collection of biological samples and research on genetic resources were issued by SISBIO ( authorization 46555–6 ) and CNPq ( 010936/2014-9 ) . Ants collected were washed with 500 μL of sterile deionized water , vortexed for 30 s and then plated on chitin medium supplemented with the antifungals nystatin and cycloheximide ( per liter: 4g chitin , 0 . 7g K2HPO4 , 0 . 3g KH2PO4 , 0 . 5g MgSO4·5H2O , 0 . 01g FeSO4·7H2O , 0 . 01g ZnSO4·7H2O , 0 . 01g MnCl2·4H2O , 20g of agar , 0 . 04 g/L nystatin , and 0 . 05 g/L cycloheximide ) . After two weeks of growth at 28°C , bacterial colonies were subcultured onto International Streptomyces Project Medium 2 ( ISP-2 ) agar with antifungals ( 0 . 04 g/L nystatin , and 0 . 05 g/L cycloheximide ) [19] . The DNA extraction procedure was modified from Kumar et al . [20] , in which the pellet was washed in 500 μL of 10 . 3% sucrose , centrifuged for 1 min at 10 , 000 g and the supernatant discarded . Then 450 μL of TSE + lysozyme were added and incubated for 20–30 min at 37°C . After , 13 μL of proteinase K was added and incubated for another 15 min at 55°C and then 250 μL of 2% SDS , gently mixed until formation of a clear solution . Then 300 μL of phenol: chloroform pH 8 . 0 were added and mixed and centrifuged for 10 min at 4°C . The supernatant was transferred to another tube and 60 μL of 3M NaOAc , pH 6 . 0 + 700 μL of isopropanol was added . The contents were mixed until "white strings" appeared and then centrifuged for 1 min to 10 , 000 g , and the supernatant discarded . The pellet was washed with 70% ethanol and centrifuged again at 10 , 000 g for 1 min . After being left overnight to completely dry the ethanol , the DNA was resuspended in 30 μL of deionized H2O . PCR amplification of the 16S rRNA gene of actinobacteria was performed using two primers: 27F ( 5'-AGAGTTTGATCMTGGCT-3' ) and 1492R ( 5'-TACGGYTACCTTGTTACGACTT-3' ) [21] . The EconoTac DNA Polymerase Kit ( Lucigen , USA ) was used and the final reaction volume of 15 μL contained: 8 μL Econotaq , 0 . 5 μL of each primer 27F and 1492R , 0 . 5 μL DMSO , 4 . 5 μL Deionized H2O and 1 μL DNA ( 10ng/μL ) . Amplification followed the following profile: an initial denaturation step at 94°C for 3 min followed by 32 cycles of amplification of 94°C for 30s , 60°C for 30s and 72°C for 2 minutes and a final extension step of 72°C for 5 min . The PCR product was detected by agarose gel electrophoresis and visualized by ultraviolet ( UV ) fluorescence after staining with ethidium bromide . The primers 27F and 1492R were used again for the sequencing of the 16S rRNA gene . The sequencing reaction of the PCR products contained: 1 . 5 μL 5X buffer , 1 μL primer ( 10 μM ) , 1 μL BigDye 3 . 1 ( Applied Biosystems ) , 0 . 5 μL DMSO , 1 μL PCR product DNA and deionized water to make up the total volume of 10 μL . The program used consisted of 95°C for 3 min , followed by 35 cycles of 96°C for 10s , 58°C for 3 min and a final extent of 72°C for 7 min . The sequencing reaction was purified with the Axyprep Mag Dyeclean purification kit ( Axygen ) in which 5 μL of magnetic beads solution and 31 μL of 85% ethanol were added for each reaction . The tubes were placed on a magnetic plate for 3 min and then the liquid was removed . 100 μL of 85% ethanol was added for 30s and then the liquid was discarded . 100 μL of 85% ethanol was added again for 30s and after discarded . The liquid was removed as much as possible with a pipette and left overnight to completely dry the ethanol . The DNA was resuspended in 25 μL of deionized H2O . Sequencing was performed at the Center for Genetics and Biotechnology at the University of Wisconsin—Madison ( Biotech Center , UW—Madison , WI , USA ) . The sequences were edited and used for assembly of the contigs in the SecMan Pro Software ( DNASTAR ) . Contigs were used to search for homologous sequences in the NCBI—GenBank ( https://blast . ncbi . nlm . nih . gov/Blast . cgi ) and Eztaxon ( http://www . ezbiocloud . net/eztaxon/identify ) . The sequences are deposited at NCBI GenBank under Accession numbers: MK118901 ( ICBG233 ) and MK118902 ( ICBG292 ) . Leishmania donovani axenic cultures ( strain MHOM/ET/67/HU3 ) were maintained in M199 medium ( pH 7 . 4 ) supplemented with 10% heat-inactivated fetal calf serum ( FCS ) and grown at 28°C [22] . Human leukemia cells ( THP-1 cell line ) were maintained in RPMI-1640 ( FCS 10% ) and grown at 37°C and 5% CO2 . Stock solutions of compounds 1–4 were prepared in 100% DMSO at 10 mM and tested in 2-fold serial dilutions ( 10 concentrations ) in 96-well flat-bottom microtiter plates . For the promastigote assay , L . donovani cells from axenic cultures in logarithmic growth were seeded at 1 x 105/well ( M199 , 80 μL ) and compounds were added in serial dilutions ( 20 μL ) . All plates included negative controls ( 100% parasite growth ) and miltefosine as a positive control . After 72 hours of incubation at 28°C , 10 μL of Alamar Blue ( 12 . 5 mg resazurin/100 mL distilled water ) [23] was added to each well and then the plates were incubated for 3 hours . This indicator of cell viability permeates into viable parasites , where it is reduced by NADPH and NADH enzymes to the highly fluorescent compound resorufin [24] . Following incubation , the plates were read with a microplate fluorometer under an excitation wave length of 536 nm and an emission wave length of 588 nm . If the test compound is inactive against the L . donovani , parasite remains viable and it is able to convert resazurin into resorufin , resulting in fluorescence emission . If the test compound is active against L . donovani , the number of viable parasites is reduced , thus resulting a decrease in fluorescence [25] . Growth inhibition was expressed as a percentage of the fluorescence of the negative control wells . IC50 values were determined using SigmaPlot . Dose-response curves were fitted using log ( inhibitor concentration ) vs . normalized response ( between 0% and 100% ) with variable slope , and IC50 values were automatically calculated . In the intracellular amastigote assay , THP-1 cells were seeded at 2 × 104/well ( RPMI-1640 , 100 μL ) with phorbol 12-myristate 13-acetate ( PMA ) at 20 ng/mL for differentiation into macrophages . After incubation for 72 hours ( 5% CO2 , 37°C ) , medium was aspirated and late-stage promastigotes were added ( 2 × 105/well , 100 μL ) . After 24 hours of incubation , medium was aspirated to clear extracellular parasites , compounds were added in serial dilutions ( 100 μL ) and the plates were incubated for 120 hours . All plates included negative controls ( 100% parasite growth ) and miltefosine as a positive control . Following incubation , medium was removed and the cells were fixed in methanol and stained with Giemsa . The average number of intracellular amastigotes per THP-1 cell was determined using an inverted microscope and a cell counter [26] . Growth inhibition was expressed as a percentage of the average number of amastigotes per macrophage in the negative control wells . IC50 values were determined as described above for the promastigote assay . For the selectivity assay , THP-1 cells were seeded at 2 × 104/well ( RPMI-1640 , 100 μL ) with PMA at 20 ng/mL for differentiation into macrophages [27] . After incubation for 72 hours ( 5% CO2 , 37°C ) , medium was aspirated , compounds were added in serial dilutions ( 100 μL ) and the plates were incubated for 120 h . All plates included negative controls and doxorubicin as a positive control . Following incubation , 10 μL of Alamar Blue was added to each well and then the plates were incubated for 3 hours . Next , the plates were read with a microplate fluorometer under an excitation wave length of 536 nm and an emission wave length of 588 nm . Growth inhibition was expressed as a percentage of the fluorescence of the negative control wells . IC50 values were determined as described above for the promastigote and amastigote assays . Seed cultures of Streptomyces sp . ICBG292 were initially grown in 40 mL of ISP-2 ( 4 tubes of 25 × 150 mm ) in a shaker for 7 days at 28°C and 200 rpm . The bacterium was inoculated into ISP-2 broth ( 4 g yeast extract , 10 g malt extract , and 4 g glucose per liter ) in a Fernbach flask ( 1 L of medium in a flask of 2 . 8 L + 70 g of HP20 ) for 7 days at 28°C and 200 rpm . The HP20 and cells were filtered and washed with water and extracted with acetone ( 2 L ) . The organic solvent was filtered and dried under vacuum . A liquid-liquid partition using ethyl acetate/water was carried out , the organic phase was separated and dried to give the crude organic extract ( 271 . 7 mg ) . The extract was purified by SPE-C18 ( 55 μm , 1 g ) using the following gradient: 10 mL ( 20% MeOH-H2O , A1: 55 . 1 mg ) ; 10 mL ( 40% MeOH-H2O , A2: 23 . 5 mg ) ; 10 mL ( 60% MeOH-H2O , A3: 14 . 5 mg ) ; 10 mL ( 80% MeOH-H2O , A4: 47 . 2 mg ) ; and 10 mL ( 100% MeOH , A5: 85 . 5 mg ) . Fractions A4 and A5 were active against Escovopsis ( S13 Fig ) , so they were combined and further purified by semi-preparative HPLC using the column C6-Phenyl ( 5μm , 250 x 10 mm ) and the following gradient at 4 mL/min: 1–20 min , linear gradient from 70% MeOH-H2O to 100% MeOH; 20–25 min , isocratic flow of 100% MeOH; 25–25 . 5 min , linear gradient from 100% MeOH to 70% MeOH-H2O; 25 . 5–30 . 5 min , isocratic flow of 70% MeOH-H2O to give 13 fractions [A5 . 1 ( 2 . 5 mg ) ; A5 . 2 ( 5 . 0 mg ) ; A5 . 3 ( 3 . 1 mg ) ; A5 . 4 ( 2 . 6 mg ) ; A5 . 5 ( 3 . 1 mg ) ; A5 . 6 ( 15 . 5 mg ) , A5 . 7 ( 10 . 9 mg ) , A5 . 8 ( 2 . 2 mg ) , A5 . 9 ( 14 . 3 mg ) , A5 . 10 ( 1 . 6 mg ) , A5 . 11 ( 0 . 9 mg ) , A5 . 12 ( 1 . 2 mg ) , A5 . 13 ( 4 . 3 mg ) ] . Fractions A5 . 2 , A5 . 6 and A5 . 9 were identified by NMR and HRESIMS as antibiotics Mer-A2026B ( 1 ) , piericidin-A1 ( 2 ) and nigericin ( 3 ) , respectively ( S1–S6 Figs ) . Purity of compounds was measured by HPLC as 99% for compound 1 , 97% for compound 2 and 93% for compound 3 . Seed culture of S . puniceus ICBG378 was initially grown in 10 mL of ISP-2 ( 25 × 150 mm tube ) and was mounted into a shaker for 2 days at 28°C and 200 rpm . The bacterium was inoculated in broth A-medium ( 20 g soluble starch , 10 g glucose , 5 g peptone , 5 g yeast extract , 5 g CaCO3 per liter ) in a baffled Erlenmeyer flask [2 x ( 100 mL of medium in a flask of 500 mL + 4 mL of seed culture + 7 g of HP20 ) ] for 7 days at 28°C and 200 rpm . The HP20 was filtered and washed with distilled water and acetone . The organic solvent was filtered and dried under vacuum to give the crude extract ( 235 . 68 mg ) , which was purified by SPE-ENV+ ( 55 μm , 1 g ) using the following gradient: 10 mL ( 25% MeOH-H2O , B1: 50 . 1 mg ) ; 10 mL ( 50% MeOH-H2O , B2: 20 . 5 mg ) ; 10 mL ( 75% MeOH-H2O , B3: 41 . 3 mg ) ; 10 mL ( 100% MeOH , B4: 102 . 5 mg ) . Fractions B3 and B4 were active against Escovopsis ( S13 Fig ) . They were mixed and purified by SPE-Si ( 55 μm , 500 mg ) with the gradient: 8 mL [100% Hexane , B4 . 1: 32 . 6 mg] , 8 mL [Hexane:EtOAc ( 8:2 ) , B4 . 2: 57 . 0 mg] , 8 mL [Hexane:EtOAc ( 6:4 ) , B4 . 3: 8 . 9 mg] , [Hexane:EtOAc ( 4:6 ) , B4 . 4: 6 . 7 mg] , 8 mL [Hexane:EtOAc ( 2:8 ) , B4 . 5: 3 . 5 mg] , 8 mL [100% EtOAc , B4 . 6: 2 . 9 mg] , and 8 mL [100% Methanol , B4 . 7: 15 . 3 mg] . The fraction B4 . 2 was identified by NMR and HRESIMS as dinactin ( 4 ) with 91% purity as measured by HPLC ( S7 and S8 Figs ) . Each bacterium-fungus and compound-fungal challenge was replicated two times on ISP-2 agar . Bacteria strains were initially screened against Escovopsis sp . ICBG1251 . Bacteria were placed in the center of ISP-2 agar Petri dishes and grown alone during 7 days; fungal strains were then point-inoculated near the edge of the culture ( microbial strains distant from each other around 3 cm ) . Two microliters of compounds ( 100 μg ) were placed in the center of Petri dishes and fungal strains were then point-inoculated near the edge of the plate . The positive control used was the miconazole . Challenges were monitored each 7 days and inhibition zone was measure after 21 days [8] . Four different fungal strains were used for testing the pure compounds: Escovopsis sp . ICBG711 ( from Trachymyrmex colony ) , Escovopsis sp . ICBG740 ( from Acromyrmex colony ) , Escovopsis sp . ICBG1251 ( from Atta colony ) and Trichoderma sp . ICBG1100 ( from attine colony ) . Atta sexdens colonies , collected at USP-Ribeirão Preto campus , were kept under laboratorial conditions . A total of 25 A . sexdens individuals , obtained from these colonies , were mechanically cleaned using small forceps and extracted with 50 mL of methanol . The extracts were filtered and evaporated to dryness . The crude extracts were evaluated for the presence or absence of compounds 1 and 2 by LC-HRESIMS , as described at general procedures .
Ten bacterial strains were isolated from A . rugosus rugosus ants , eight strains were recovered from Cyphomyrmex ants and twelve from A . sexdens . All 30 bacterial strains were challenged in antagonism assays against Escovopsis sp . , the specialized pathogenic fungus of Attine ants , and bioactive strains were identified through 16S rRNA sequencing . Streptomyces puniceus ICBG378 from A . rugosus rugosus , Streptomyces sp . ICBG292 from Cyphomyrmex sp . , and Streptomyces sp . ICBG233 from A . sexdens showed high inhibition of Escovopsis , and were selected for scale up culturing and antiprotozoal assays . Crude extracts and fractions of cultures of the three selected Streptomyces strains inhibited the growth of L . donovani promastigotes ( inhibition higher than 90% ) . Therefore , they were selected for the isolation and characterization of biologically active natural products . The fractionation of extracts was guided by the antifungal assay against Escovopsis ( S1 Fig ) and led to the isolation of the known antibiotics mer-A2026B [28] ( 1 ) , piericidin-A1 [29] ( 2 ) , nigericin [30 , 31] ( 3 ) , produced by Streptomyces sp . ICBG292 ( Fig 1 ) ; and dinactin [32] ( 4 ) , produced by S . puniceus ICBG378 . Compounds 1 and 2 were also isolated from Streptomyces sp . ICBG233 , associated with A . sexdens ants . Structures were established on the basis of NMR and HRESIMS data and comparison with literature ( S2–S14 Figs ) . Compounds 1–4 were active against Escovopsis sp . ICBG740 ( Fig 2 ) . Compound 1 showed higher antagonist activity against four different Escovopsis strains compared to compounds 2–4 ( Fig 2 and S15–S17 Figs ) , with inhibition zone similar to the positive control ( miconazole ) . Compound 1 was also active against the fungus Trichoderma sp . ( Fig 3 ) . All compounds were active against L . donovani promastigotes while compounds 1 , 2 and 4 were also active against intracellular amastigotes ( Table 1 ) . This is the first report of the antileishmanial activity of antibiotics Mer-A2026B ( 1 ) , piericidin-A1 ( 2 ) and dinactin ( 4 ) . L . donovani lives in the sandfly gut as promastigotes . Promastigotes are the infective stage of Leishmania sp . , being transmitted to humans via the bite of sandflies . Skin macrophages phagocyte the promastigotes , where the promastigotes differentiate into amastigote form . Intracellular amastigotes reproduce within the macrophages , eventually rupturing the host cell to infect other surrounding macrophages [33] . In addition to being involved in different stages of the life cycle of the parasite , promastigotes and amastigotes differ morphologically . Promastigotes are flagellated elongated cells , while amastigotes are rounded non-flagellated cells [26] . Compounds 3 and 4 were more active against both L . donovani forms than the positive control miltefosine ( Table 1 ) . Although intracellular amastigotes are the clinically relevant form , assessing the activity against both parasite stages can provide important information for further studies on the mechanism of action of these compounds . These activity data can be useful to investigate which biochemical pathways are modulated and understand the role played by the respective molecular targets in each stage of the parasite life cycle . The selectivity index , which is the ratio between the activity against THP-1 macrophages and intracellular amastigotes , indicates whether the compounds are selective for L . donovani over the human host cells . The probability of a compound to elicit cytotoxic effects in the human host decreases as the selectivity index increases . Therefore , the selectivity index is an important safety metric , and was assessed for compounds 3 and 4 . The selectivity indexes of 3 and 4 were 88 . 91 and 656 . 11 , respectively , suggesting their safety . A potent vasodilating activity has been reported for mer-A2026B ( 1 ) [34]; and insecticidal , antimicrobial and cytotoxic activities for piericidin-A1 ( 2 ) [35–37]; while strong antibacterial and anticancer activities have been found for nigericin ( 3 ) and dinactin ( 4 ) [38–40] . The high activities and good selectivity indexes obtained for nigericin ( 3 ) and dinactin ( 4 ) in our experiments ( Table 1 ) are in agreement with previous data for nigericin monosodium salt and nonactin , an analogue of 4 , using ex-vivo splenic explant culture system from hamsters infected with L . donovani [41] . Compounds 3 and 4 are considered ionophores that reversibly bind and transport ions across biological membranes [42] . Nigericin ( 3 ) has been shown to move sodium and potassium ions through membranes [43] . When bound to a cation , nigericin loses a proton and generates an uncharged species that can permeate into cell membranes , acting as a carrier . The molecule can also permeate into membranes as a protonated noncomplexed molecule . Nigericin can promote an exchange of K+ for H+ that results in the modification of the ion gradient across the membranes involved in the energetic metabolism [44] . Dinactin ( 4 ) is one member of the family of macrotetrolide nactins with ability to selectively complex a wide variety of cations [45] . Few ionophore compounds have been described to inhibit L . donovani . One example is the ionophore A23187 that binds Ca2+ and kills intracellular Leishmania in the presence of lipopolysaccharide ( LPS ) , mediated by generation of L-arginine-dependent nitrogen oxidation products [46] . Another ionophore , named calcimycin , has been described to kill Leishmania promastigotes by activating parasite nitric oxide synthase [47] . The Leishmania cell death is accompanied by the loss of mitochondrial polarization and plasma membrane integrity and can be blocked by specific inhibitors of constitutive Ca2+/calmodulin-dependent nitric oxide synthase [47] . The most recognized mechanism of action of miltefosine against L . donovani is the inhibition of phospholipid synthesis and cytochrome c oxidase , but recently another mechanism has been described based in the abrupt increase in the intracellular Ca2+ concentration in the L . donovani [48] , a similar property of ionophores . L . donovani lives in the sandfly gut as promastigotes and in mammalian macrophages as amastigotes [49] . This protozoan extrudes protons through H+-ATPase to regulate intracellular pH and to facilitate nutrient uptake [49] . This proton extrusion is enhanced by the addition of K+ [49] . This could be one mechanism by which nigericin controls the growth of Leishmania parasite . The mechanisms of action of nigericin ( 3 ) and dinactin ( 4 ) against L . donovani have not been described . Compounds 1 and 2 showed drug-like properties according to several rules such as Lipinski and Veber filters [50 , 51] , while 3 and 4 exceed the ideal molecular weight and number of hydrogen-bond acceptors ( HBA ) ( Table 2 ) . The computational predictions were run using SwissADME [52] and Stardrop ( Optibrium ) [53] . Lipinski’s rule of five states that compounds showing more than 5 hydrogen-bond donors , 10 hydrogen-bond acceptors , molecular weight greater than 500 and LogP greater than 5 , are likely to show poor gastrointestinal absorption [50] . However , several natural products that do not comply with Lipinski´s rules have been approved as drugs , such as paclitaxel , rapamycin , cyclosporine A , and others . In general , natural products are considered as exceptions to Lipinski´s rules . However , the properties LogP and hydrogen-bond donors are very important for predicting bioavailability . A possible explanation is that nature can maintain low hydrophobicity and intermolecular hydrogen-bond donating potential when it needs to produce active compounds with high molecular weight and rotatable bonds; and natural products could also take advantage of active transport mechanisms since they contain biosynthetic moieties that resemble endogenous metabolites [54] . So , dinactin ( 4 ) could be an interesting compound for further pharmacological studies in the treatment of leishmaniasis based on the high selectivity index against L . donovani and on LogP and HBD values that comply with Lipinski´s rules . Furthermore , given their remarkable in vitro activity , compounds 3 and 4 are suitable starting points for molecular optimization aiming to pursue molecules that fit into the drug-like concept . Compounds 1–4 can join the chemical cocktail used by actinobacteria to control the growth of the pathogenic fungus Escovopsis sp . and other opportunistic fungi such as Trichoderma sp . in fungus-growing ant colonies . Compounds 1 and 2 were also identified from Streptomyces sp . ICBG233 associated to workers of Atta sexdens and from the organic extract of these ants by mass spectrometry ( S18–S23 Figs ) , confirming their production in the natural environment . Compound 2 and other piericidin derivatives together with nigericin ( 3 ) have also been reported from Candidatus Streptomyces philanthi symbiont of solitary beewolf digger wasps ( Philanthus triangulum , Hymenoptera , Crabronidae ) as antibiotic protectors of their larval offspring against pathogens [55 , 56] . Authors argue that the mixture of these antibiotics could help in the evolutionary stable defense against different pathogens [55 , 56] , and the current identification of the same compounds in bacterial symbionts of attine ants reinforces this hypothesis . Considering the remarkable activity against L . donovani shown by the identified compounds and that the treatment for visceral leishmaniasis suffers from several drawbacks , the results reported herein can contribute to the development of novel therapeutic agents for this NTD . Moreover , most current drug development approaches are based on high-throughput screening ( HTS ) of synthetic compound collections . HTS platforms can screen libraries containing thousands of molecules , whose chemical diversity are provided by methods such as combinatorial chemistry . Natural products can provide further structural diversity and novel chemotypes that differ from those obtained via combinatorial chemistry . In this context , natural products are a rich source of structural diversity that offers unique chemical matter to be used as reference for the design of novel leishmanicidal agents . Our results also validate the ecological approach of screening antifungal natural products from actinobacteria associated to attine ants as a good strategy for discovering antileishmanial compounds . | Visceral leishmaniasis , caused by Leishmania infantum and L . donovani , is characterized by high rate mortality worldwide . Current treatments for this disease suffer from toxicity , variable efficacy , requirements for parenteral administration and length of treatment regimens . New chemical entities and development of new drugs are important to overcome the impact of this protozoan disease . Actinobacterial strains , such as Streptomyces , have been a source of most naturally derived antibiotics , as well as anticancer , anthelmintic , and antifungal drugs . These microorganisms also produce small molecules important in symbiotic interactions with insects , such as fungus-growing ants , fungus-growing termites , beetles and wasps against pathogens . Several novel compounds have been reported from these microorganisms with promising biological activities . In this work we show an interesting ecologic approach for drug discovery that also shows promise for the identification of antileishmanial natural products from fungus-growing ant ecosystem . Two compounds isolated from Streptomyces strains showed potent activity against L . donovani , higher than the positive control ( miltefosine ) with high selectivity indexes . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"and",
"discussion"
]
| [
"blood",
"cells",
"antimicrobials",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"drugs",
"immunology",
"microbiology",
"antifungals",
"parasitic",
"protozoans",
"animals",
"protozoan",
"life",
"cycles",
"developmental",
"biology",
"streptomyces",
"protozoans",
"leishmania",
"pharmacology",
"promastigotes",
"fungal",
"pathogens",
"mycology",
"white",
"blood",
"cells",
"hymenoptera",
"animal",
"cells",
"ants",
"medical",
"microbiology",
"microbial",
"pathogens",
"life",
"cycles",
"leishmania",
"donovani",
"insects",
"amastigotes",
"arthropoda",
"eukaryota",
"cell",
"biology",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"protozoology",
"cellular",
"types",
"macrophages",
"organisms"
]
| 2019 | Antifungal compounds from Streptomyces associated with attine ants also inhibit Leishmania donovani |
Sensitive diagnostic tools are required for an accurate assessment of prevalence and intensity of helminth infections in areas undergoing regular deworming , and for monitoring anthelmintic drug efficacy . We compared the diagnostic accuracy of the Kato-Katz and FLOTAC techniques in the frame of a drug efficacy trial . Stool samples from 343 Zanzibari children were subjected to duplicate Kato-Katz thick smears and the FLOTAC basic technique in a baseline screening in early 2009 . The FLOTAC showed a higher sensitivity than the Kato-Katz method for the diagnosis of Trichuris trichiura ( 95% vs . 88% , p = 0 . 012 ) and Ascaris lumbricoides ( 88% vs . 68% , p = 0 . 098 ) , but a lower sensitivity for hookworm diagnosis ( 54% vs . 81% , p = 0 . 006 ) . Considering the combined results from both methods as ‘gold’ standard , the prevalences of T . trichiura , hookworm and A . lumbricoides were 71% ( 95% confidence interval ( CI ) : 66–75% ) , 22% ( 95% CI: 17–26% ) and 12% ( 95% CI: 8–15% ) , respectively . At follow-up , 3–5 weeks after 174 among the 269 re-examined children were administered anthelmintic drugs , we observed cure rates ( CRs ) against A . lumbricoides , hookworm and T . trichiura of 91% ( 95% CI: 80–100% ) , 61% ( 95% CI: 48–75% ) and 41% ( 95% CI: 34–49% ) , respectively , when using the Kato-Katz method . FLOTAC revealed lower CRs against A . lumbricoides ( 83% , 95% CI: 67–98% ) and T . trichiura ( 36% , 95% CI: 29–43% ) , but a higher CR against hookworm ( 69% , 95% CI: 57–82% ) . These differences , however , lacked statistical significance . Considerable differences were observed in the geometric mean fecal egg counts between the two methods with lower egg reduction rates ( ERRs ) determined by FLOTAC . Our results suggest that the FLOTAC technique , following further optimization , might become a viable alternative to the Kato-Katz method for anthelmintic drug efficacy studies and for monitoring and evaluation of deworming programs . The lower CRs and ERRs determined by FLOTAC warrant consideration and could strategically impact future helminth control programs .
Current estimates suggest that soil-transmitted helminths might still affect over a quarter of the world's population and inflict a huge public health burden , particularly on rural and deprived urban communities in the developing world [1]–[3] . Efforts are underway to reduce morbidity due to soil-transmitted helminthiasis , with an ambitious target to administer anthelmintic drugs regularly to at least 75% and up to 100% of school-aged children and other high-risk groups [4] . One region where this goal was already met in 2006 is Zanzibar , a group of islands forming part of the United Republic of Tanzania [5] . Indeed , helminth control programs have been implemented in Zanzibar for 20 years and have resulted in marked decreases in prevalence , intensity and morbidity due to soil-transmitted helminthiasis , urinary schistosomiasis , and lymphatic filariasis [6]–[9] . The diagnosis of soil-transmitted helminth infections with direct parasitological methods based on egg detection in stool is unreliable among infected individuals who harbor only a few intestinal worms , since egg output is much lower than among heavily infected individuals . Hence , in settings where helminth control programs have been implemented and infection intensities dropped as a result of regular deworming , diagnostic methods with a high sensitivity are needed for an accurate assessment of the actual prevalence and intensity of soil-transmitted helminth infections [10] . Sensitive diagnostic tools are also mandatory for monitoring drug efficacy and to detect the emergence and spread of drug resistance [11] , [12] . The Kato-Katz method [13] is a widely used diagnostic tool , which provides estimates of population prevalence and infection intensity , and facilitates anthelmintic drug efficacy assessment in clinical trials and monitoring and evaluation of community-based control programs . However , due to the small amount of stool examined ( i . e . , 41 . 7 mg ) , the Kato-Katz method shows a low sensitivity when infection intensities are light , which is common after deworming [14] . The sensitivity can be improved by examining multiple Kato-Katz thick smears produced from a single stool sample or by examining multiple stool samples [15]–[17] . With the recently developed FLOTAC method , up to 1 g of stool can be examined – 24 times more than with a single Kato-Katz thick smear – and standard protocols are now available [18] . The FLOTAC technique showed a higher sensitivity than multiple Kato-Katz thick smears for the diagnosis of soil-transmitted helminth infections in previous studies [19]–[22] . Moreover , FLOTAC outperformed the McMaster technique qualitatively and quantitatively for hookworm diagnosis [23] . What is not known , though , is how the FLOTAC technique performs within anthelmintic drug efficacy trials , and whether it might be utilized for monitoring and evaluation of helminth control programs that are shifting the tactic from morbidity control to infection and transmission control , and eventually local elimination . The objective of this study was to compare the accuracy of the Kato-Katz method with the FLOTAC basic technique for the detection of soil-transmitted helminth infections within a randomized controlled trial on anthelmintic drug efficacy and safety carried out in Zanzibar in early 2009 [24] . The diagnostic accuracy of each method was determined before and after treatment and anthelmintic drug efficacy was estimated according to the method used .
The study presented here was embedded in a randomized controlled trial with the protocol being reviewed by the institutional research commission of the Swiss Tropical and Public Health Institute ( Basel , Switzerland ) . Ethical approval was given by the ethics committee of Basel ( EKBB , reference no . 13/09 ) and the Ministry of Health and Social Welfare of Zanzibar ( MoHSW , reference no . ZAMEC/0001/09 ) . The study is registered at controlled-trials . com , identifier ISRCTN08336605 . The directors and teachers of the primary schools in Kinyasini and Kilombero were informed about the purpose and procedures of the study . The study was explained in lay terms to the school children in their local language ( Kiswahili ) . An informed consent sheet , including study information and the fact that participation was voluntary , was distributed to each child . Written informed consent was obtained from parents or guardians prior to stool sampling . Additionally , oral assent was sought from children . At the end of the study in May 2009 , all children attending the two schools were offered free anthelmintic drugs , as part of the regular deworming done by the Helminth Control Laboratory Unguja ( HCLU ) . Single oral doses of albendazole ( 400 mg ) and praziquantel ( 40 mg/kg ) were administered to school-aged children for preventing morbidity due to soil-transmitted helminthiasis and urinary schistosomiasis , respectively . The study was carried out on Unguja island , Zanzibar , in the first half of 2009 within the frame of a randomized controlled trial to assess the efficacy and safety of different anthelmintic drugs against Trichuris trichiura and other soil-transmitted helminth infections [24] . For the assessment of anthelmintic drug efficacy according to the diagnostic method used , we aimed at enrolling at least 200 T . trichiura-positive individuals [25] . In view of the sample size calculations for our clinical trial , we aimed at screening 2000 children to identify 600 T . trichiura-infected subjects , assuming a prevalence of 30% [8] . All children attending the primary schools of Kilombero and Kinyasini in district North A , located 30–40 km from Zanzibar Town , were eligible to submit stool samples . To reach the number of 200 T . trichiura-infected children for assessing anthelmintic drug efficacy , not only with the Kato-Katz , but also with the FLOTAC technique , we systematically preserved ∼1 g of every third stool sample collected at the baseline screening for subsequent FLOTAC examinations . Since the T . trichiura prevalence turned out to be >50% , we stopped recruiting children during our baseline survey after we had included 1240 children in the study . Among these children , 1066 ( 86% ) had submitted one stool sample; 750 from Kinyasini and 316 from Kilombero . The baseline screening was carried out as follows: in early March 2009 , over a period of 3 weeks , every morning between 08:00 and 09:30 hours , ∼120 children were called from class and given a stool collection container labeled with unique identification numbers ( IDs ) . Children were asked to return the container filled with a lime-sized own fresh morning stool sample the following day . Upon collection , filled stool containers were ordered by increasing IDs in a wooden transport-shelf and promptly transferred to the HCLU in Zanzibar Town . At HCLU , duplicate Kato-Katz thick smears were prepared from each stool sample , using standard 41 . 7 mg templates [13] . All Kato-Katz thick smears were examined quantitatively by one of four experienced laboratory technicians for hookworm eggs after a clearing time of 20–40 min in the morning and by another one of four experienced laboratory technicians for T . trichiura and Ascaris lumbricoides eggs in the afternoon . Slides were numbered with the child's ID plus letter A or B , and each microscopist adhered to either the A or B series to avoid duplicate reading of the same stool sample by the same technician . With regard to the FLOTAC method , ∼1 g of stool , systematically obtained from each third stool sample in the transport-shelf , was weighed in a plastic tube , filled with 10 ml of 5% formaldehyde . Stool samples were suspended with a wooden spatula and stored at room temperature until further use . Three to 9 weeks after the collection of the first stool sample , children who had their stool samples examined both by the Kato-Katz and FLOTAC techniques were invited to submit a second stool sample . Samples were again processed with duplicate Kato-Katz , and ∼1 g of stool was preserved in 5% formaldehyde and stored at room temperature . Of note , children with a T . trichiura infection as identified by duplicate Kato-Katz thick smear readings at baseline and meeting other inclusion criteria , were treated with one of the following four drug regimens: ( i ) albendazole ( 400 mg ) plus placebo; ( ii ) albendazole plus ivermectin ( 200 µg/kg ) ; ( iii ) mebendazole ( 500 mg ) plus placebo; and ( iv ) mebendazole plus ivermectin . Drugs were administered shortly after the end of the baseline screening in late March 2009 . Follow-up stool samples from children who had received anthelmintic drugs were collected within 3–5 weeks after treatment . The administration of albendazole and praziquantel to all school children in Kinyasini and Kilombero ( and other schools ) regardless of whether or not children participated in our study was carried out by members of the HCLU in late May 2009 , when all available follow-up stool samples had been collected . After completion of the trial in late May 2009 , in the last 2 weeks of the study , all formaldehyde-preserved stool samples were examined with the FLOTAC basic technique [18] . Since FLOTAC had not been implemented at HCLU before , all 22 laboratory workers , including eight microscopists , underwent a 3-day training workshop with two FLOTAC specialists from Italy to become acquainted with this new diagnostic procedure . We used flotation solution no . 4 ( FS4; sodium nitrate: NaNO3 315 g plus 685 ml H2O; specific gravity ( s . g . ) = 1 . 20 ) in light of our preceding results for the diagnosis of soil-transmitted helminth infections [18] , [19] , [21] . Each preserved stool suspension was pressed through a tea sieve using a wooden spatula and adding 10 ml of 0 . 9% NaCl . The supernatant was equally distributed in two labeled 15 ml plastic tubes and centrifuged for 3 min at 170 g in a Hettich EBA centrifuge ( Tuttlingen , Germany ) . Subsequently , the supernatant was discarded and each tube filled to the 6 ml mark with FS4 . The pellets were suspended by pipetting the solution up and down , and 5 ml of the suspension were transferred into one of the two chambers of the labeled FLOTAC apparatus . Next , the FLOTAC apparatus was centrifuged for 5 min at 120 g in a Hettich Universal 320 centrifuge ( Tuttlingen , Germany ) . Finally , after translation of the top portion of the FLOTAC apparatus , the observation grids of both chambers were examined for soil-transmitted helminth eggs at 100× magnification . Fecal egg counts ( FECs ) for each helminth species were recorded separately for each Kato-Katz thick smear and each of the two FLOTAC observation grids . For quality control , 10% of the slides and observation grids were re-examined by a senior laboratory technician . In case the senior technician detected one or several eggs of a helminth species that had not been recorded in the original reading , the former result was considered as false-negative and replaced by the result of the senior technician . Moreover , in case of deviations in FECs of more than 10% , the original egg count was replaced by the result of the senior technician . In both cases , the microscopist was advised to read more carefully the following days . All Kato-Katz thick smears would have been re-read if there were discrepancies in the FECs in more than 20% of the re-examined slides , but this was never the case over the course of our study . The microscopists reading the FLOTAC observation grids were blinded to the results derived from the Kato-Katz method . Data were entered twice in Microsoft Excel version 10 . 0 ( 2002 , Microsoft Corporation; Redmond , WA , USA ) and checked for consistency with EpiData version 3 . 1 ( EpiData Association; Odense , Denmark ) . Discrepancies were removed by consulting original data records . Data sets for Kato-Katz and FLOTAC results from the baseline screening and follow-up were merged by ID . Statistical analyses were carried out with STATA version 10 ( StataCorp . ; College Station , TX , USA ) . For method comparisons , only individuals whose stool samples were subjected to duplicate Kato-Katz and FLOTAC at baseline or follow-up were included . We used the combined results of duplicate Kato-Katz and two FLOTAC observation grids as diagnostic ‘gold’ standard . A child with egg-positive microscopic test results in any Kato-Katz thick smear or FLOTAC observation grid was considered a true-positive . We assumed 100% specificity , and hence the complete absence of false-positive results for Kato-Katz and FLOTAC on the basis of unambiguously identifiable soil-transmitted helminth eggs under a microscope by experienced technicians . The sensitivity ( proportion of true-positives detected by the test [26] ) was calculated for duplicate Kato-Katz thick smears or the FLOTAC basic technique in relation to our ‘gold’ standard . The agreement between the results of the FLOTAC basic technique and duplicate Kato-Katz thick smear readings examined at baseline and follow-up was assessed for the diagnosis of A . lumbricoides , hookworm , and T . trichiura using kappa ( κ ) statistics [27] . Interpretation of κ statistics was as follows: <0 . 00 indicating no agreement , 0 . 00–0 . 20 indicating poor agreement , 0 . 21–0 . 40 indicating fair agreement , 0 . 41–0 . 60 indicating moderate agreement , 0 . 61–0 . 80 indicating substantial agreement , and 0 . 81–1 . 00 indicating almost perfect agreement [28] . The McNemar test was used to assess the inter-method differences in sensitivities , considering only individuals who were identified as positive for A . lumbricoides , hookworm , or T . trichiura according to the diagnostic ‘gold’ standard [29] . Statistical significance was given for P-values<0 . 05 . The difference in intra-method sensitivity assessed before and after treatment was determined based on the assumption that non-overlapping 95% confidence intervals ( CIs ) indicate statistical significance . Helminth-specific FECs of each individual were expressed as eggs per gram of stool ( EPG ) , calculated by multiplying the sum of the two FECs from duplicate Kato-Katz thick smears by a factor 12 . For FLOTAC , the FECs obtained from the two observation grids were added and multiplied by a factor ( 1/weight of stool sample ) and expressed as EPG . The geometric mean ( GM ) EPG of the study cohort was calculated using the normal logarithm of the EPG plus 1 ( GM = exp ( ( ∑ log ( EPG+1 ) ) /n ) −1 ) , where log ( EPG+1 ) is the sum of the logarithm of each individual EPG , and one egg is added to each count to permit the calculation of the logarithm in case of EPG = 0 [30] . We calculated 95% CIs for sensitivity and the arithmetic mean ( AM ) EPGs and GM EPGs of the study cohort . Participants with complete data from the baseline and follow-up survey , who received treatment and who were identified as positive for A . lumbricoides , hookworm , or T . trichiura according to our ‘gold’ standard at the baseline survey were included in the calculation of cure rate ( CR ) and egg reduction rate ( ERR ) . The CR was determined as the percentage of children excreting eggs before treatment according to the ‘gold’ standard who became negative after treatment according to either the Kato-Katz or the FLOTAC method . CRs derived by the Kato-Katz or FLOTAC method were compared using a two-sample test of proportion . The ERR determined with Kato-Katz and FLOTAC from the treated children was calculated according to World Health Organization ( WHO ) guidelines [30] , as follows: ERR = ( ( GM EPG before treatment – GM EPG after treatment ) /GM EPG before treatment ) ×100 . The group GM EPG used to determine the ERR was calculated from the group of individuals identified as positive for A . lumbricoides , hookworm , or T . trichiura according to the diagnostic ‘gold’ standard at the baseline survey .
Consent to participate in our trial was given by the parents and guardians of 1240 children , among whom 1066 children provided a stool sample at baseline . For FLOTAC examinations , 385 ( 36% ) stool samples were preserved in 5% formaldehyde at baseline ( Figure 1 , left arm ) . Due to inaccurate preparation , a sudden power cut , and the flotation of stool debris , which hindered subsequent microscopic examinations , and because of erroneous labeling , the results from 32 stool samples preserved at baseline were not available . Additionally , five IDs did not match the IDs from the Kato-Katz results . For another five IDs only a single instead of duplicate Kato-Katz results were available . Hence , 343 among the 385 individuals ( 89% ) had complete FLOTAC and duplicate Kato-Katz results at baseline . Among them 182 ( 53% ) were girls and 161 ( 47% ) boys . The age ranged between 6 and 20 years with a median of 11 years . A second stool sample was preserved from 288 among the 385 individuals ( 75% ) 3–9 weeks after the collection of the first stool sample ( Figure 1 , right arm ) . Among them 204 children were given one of four different anthelmintic treatments . Results from 18 preserved stool samples were lost due to the flotation of debris or incorrect labeling . Hence , 270 preserved stool samples were examined with the FLOTAC basic technique at follow-up . Complete FLOTAC and duplicate Kato-Katz results were available from 269 ( 70% ) individuals at follow-up . Since 29 IDs from the first and second stool examination data set did not match , complete examination data from the baseline and follow-up survey were available from 240 among the 385 originally selected study participants ( 62% ) . Among them , 174 were given anthelmintic drugs . Results presented in Tables 1 and 2 show that the FLOTAC basic technique detected T . trichiura and A . lumbricoides infections with a higher sensitivity than duplicate Kato-Katz thick smears , but was less sensitive in detecting hookworm eggs , both at baseline and follow-up . At baseline , the sensitivity of FLOTAC for the diagnosis of T . trichiura , A . lumbricoides and hookworm was 95 . 0% , 87 . 5% and 54 . 1% , respectively , whereas the respective sensitivity of Kato-Katz was 88 . 0% , 67 . 5% and 81 . 1% . At follow-up , the sensitivity of FLOTAC for the diagnosis of A . lumbricoides , T . trichiura and hookworm was 97 . 4% , 93 . 3% and 61 . 2% , respectively , whereas the sensitivity of Kato-Katz was 41 . 7% , 84 . 9% and 77 . 6% , respectively . The inter-method sensitivity between FLOTAC and Kato-Katz differed significantly for the detection of T . trichiura at baseline and follow-up ( P = 0 . 012 and P = 0 . 030 ) and for hookworm at baseline ( P = 0 . 006 ) . The intra-method sensitivity assessed at baseline and follow-up differed significantly for the diagnosis of A . lumbricoides only , according to non-overlapping 95% CIs . Moderate-to-substantial agreement between the two diagnostic techniques was observed for all helminths investigated , before and after treatment . The highest agreement ( κ = 0 . 74 ) was observed for T . trichiura both before and after treatment , whereas the lowest agreement was noted for hookworm diagnosis at baseline ( κ = 0 . 44 ) . In line with a higher sensitivity of the FLOTAC basic technique for the diagnosis of A . lumbricoides and T . trichiura , the observed prevalences of T . trichiura and A . lumbricoides infections determined with FLOTAC were higher compared to the ones derived by the Kato-Katz method . The opposite was observed for hookworm . At baseline , T . trichiura , hookworm and A . lumbricoides infections were detected in 67 . 1% , 11 . 7% and 10 . 2% of the children , respectively , when using FLOTAC . The respective prevalences according to Kato-Katz were 62 . 1% , 17 . 5% , and 7 . 9% ( Table 1 ) . Considering the results from the two methods combined , the respective prevalences were 70 . 6% , 21 . 6% , and 11 . 7% . The GM EPGs revealed with the Kato-Katz method were higher than those obtained with FLOTAC , showing values of 18 . 9 EPG for T . trichiura , 1 . 2 EPG for hookworm , and 0 . 9 EPG for A . lumbricoides vs . 9 . 7 EPG , 0 . 3 EPG and 0 . 6 EPG , respectively . At follow-up , after 193 out of the 269 children ( 72% ) had received experimental treatment , observed prevalences of T . trichiura , hookworm and A . lumbricoides infections had decreased to 57 . 3% , 11 . 2% and 4 . 1% , respectively , according to the FLOTAC technique . The respective prevalences according to the Kato-Katz method were 52 . 0% , 14 . 1% and 1 . 9% ( Table 2 ) . Results of both methods combined revealed prevalences of 61 . 3% , 18 . 2% and 4 . 5% , respectively . As expected , the GM EPGs were lower at follow-up than at baseline . The Kato-Katz method revealed GM EPGs of 10 . 5 EPG , 0 . 8 EPG and 0 . 2 EPG for T . trichiura , hookworm and A . lumbricoides , respectively . The FLOTAC method revealed respective GM EPGs of 4 . 4 EPG , 0 . 2 EPG and 0 . 2 EPG . Table 3 shows diagnostic method-specific CRs and ERRs estimated for those children who were treated , had complete data records and were identified as positive for A . lumbricoides , hookworm , or T . trichiura according to the ‘gold’ standard at the baseline survey . Employing duplicate Kato-Katz thick smears before and after treatment revealed CRs of 91 . 3% , 61 . 2% and 41 . 4% against A . lumbricoides , hookworm and T . trichiura infections , respectively . The estimated CRs using FLOTAC were lower for A . lumbricoides ( 82 . 6% ) and T . trichiura ( 36 . 2% ) , but higher for hookworm ( 69 . 4% ) . However , none of the differences showed statistical significance . The ERRs determined with the Kato-Katz method for A . lumbricoides , hookworm and T . trichiura infections were 99 . 9% , 89 . 9% and 87 . 6% , respectively , and with the FLOTAC method 99 . 4% , 65 . 5% and 80 . 7% , respectively . A total of 66 children had a stool sample examined with duplicate Kato-Katz thick smears and the FLOTAC basic technique at baseline and follow-up , without treatment in between . Among them , eight , seven and four children were identified to be infected with T . trichiura , hookworm and A . lumbricoides , respectively , by the Kato-Katz method at baseline ( Table 4 ) . The FLOTAC technique identified 22 , five and four positive children , respectively . At follow-up , 24 and nine children were infected with T . trichiura and hookworm , respectively , according to the Kato-Katz method . The FLOTAC technique identified 26 children with a T . trichiura infection , 10 with hookworm eggs in their stool and one case of A . lumbricoides .
We found a significantly higher sensitivity of the FLOTAC basic technique compared to the Kato-Katz method for the diagnosis of T . trichiura , both before and after anthelmintic drug administration . The sensitivity of FLOTAC was also higher for the diagnosis of A . lumbricoides at both time points , but the difference showed no statistical significance . With regard to hookworm diagnosis , FLOTAC showed a significantly lower sensitivity before experimental chemotherapy . The intra-method sensitivity assessed before and after treatment showed considerable heterogeneity for A . lumbricoides diagnosis , but not for the other two soil-transmitted helminth species investigated , notwithstanding significantly lower FECs after treatment for each species . In general , the GM EPGs obtained with the Kato-Katz method were several-fold higher than those derived from the FLOTAC method . ERRs determined by FLOTAC after anthelmintic treatment were lower than those derived by the Kato-Katz method . There was no statistically significant difference in the CRs as determined by the Kato-Katz or FLOTAC method . However , the FLOTAC revealed somewhat lower CRs than the Kato-Katz method for both T . trichiura ( 36% vs . 41% ) and A . lumbricoides ( 83% vs . 91% ) . The opposite was found for hookworm ( 69% vs . 61% ) . The higher sensitivity of the FLOTAC basic technique for A . lumbricoides and T . trichiura diagnosis compared to the Kato-Katz method is in line with previous studies [21] , [22] . The lower sensitivity for detecting hookworm eggs reported here , however , is in contrast to prior investigations performed with stool samples from Côte d'Ivoire and Zanzibar [19] , [21] , [22] . In our hands now , the sensitivity of FLOTAC for hookworm diagnosis was as low as 54% at baseline and slightly higher at follow-up ( 61% ) , whereas in the previous studies , sensitivities above 80% were reported [19] , [21] . Four issues are offered for consideration , which might explain these observations . First , in the current study , we rigorously adhered to examining Kato-Katz thick smears within 20–40 min after preparation for hookworm egg counts to avoid over-clearance due to glycerol-soaked cellophane strips [31] . This had likely benefited the sensitivity outcome of Kato-Katz . Indeed , a limitation of our previous studies had been that we examined the slides for hookworm eggs only after 40–60 min post-preparation , which might have resulted in hookworm egg over-clearance [19] , [21] . Second , the stool samples in the previous studies were preserved in sodium acetate-acetic acid-formalin ( SAF ) , whereas in the current study 5% formaldehyde was used . A potential negative impact of the stool preservation media and FS on fragile hookworm eggs have been discussed before [18] , [20] , [22] . Third , the higher sensitivity of FLOTAC for hookworm diagnosis at follow-up compared to baseline is pointing to a negative impact of the duration of stool preservation on hookworm eggs ( samples collected at follow-up had at least a 3-week shorter preservation period than samples preserved at the baseline survey ) , which is in line with findings from Côte d'Ivoire [20] , [22] . Fourth , floated organic debris might have averted the accurate detection of transparent hookworm eggs in some of our stool samples , and hence negatively impacted on the sensitivity of FLOTAC for hookworm diagnosis . This latter problem was recently observed in stool samples collected from school children in Côte d'Ivoire and Pemba island , where it was overcome by including a washing step with ether or ethyl acetate to remove the organic debris or by a higher dilution of stool samples using tap water [22] , [23] . The comparable sensitivities of either method at baseline and follow-up , despite a considerable decrease in FECs , suggest that a decrease in sensitivity only occurs if FECs fall under the lower detection limit of a method ( i . e . , 12 EPG for duplicate Kato-Katz thick smears , 24 EPG for a single Kato-Katz thick smear and 1 EPG for the FLOTAC basic technique ) . This suggestion is supported by the finding that those seven individuals found A . lumbricoides-positive by FLOTAC , but not with duplicate Kato-Katz thick smears at follow-up showed FECs of 9 . 8 EPGs and below , which likely explains the significant difference in the intra-method sensitivity for A . lumbricoides diagnosis before and after treatment . The considerably lower numbers in the GM EPGs of our study cohort derived by FLOTAC in comparison to Kato-Katz are in line with previous studies [19] , [21] , [22] . Since there is no evidence of an upper detection limit of eggs of the FLOTAC method or of an artificial distortion in FECs associated with the smaller amount of biological material examined with the Kato-Katz method , but rather a linear relationship between FECs detected by the Kato-Katz and FLOTAC method , the following two hypotheses are offered for consideration . First , the amount of fecal material used in the Kato-Katz template ( 41 . 7 mg ) is filtered , which might act like a concentration step , and hence contains a higher number of eggs [32] , [33] , whereas the amount of stool used for FLOTAC ( ∼1 g ) is measured before filtering and contains heavy fibers , seeds and other undigested foodstuffs , but there is no concentration of eggs . Second , the FLOTAC procedure might not bring all helminth eggs into flotation in the FS-stool suspension , but only a certain proportion . Hence , some eggs would remain in the stool debris pellet . This might happen due to a variety of reasons , including the physical damage of eggs or slightly different densities of fertilized and unfertilized eggs . Clearly , additional investigations are warranted to elucidate these hypotheses . The somewhat lower observed CRs against T . trichiura and A . lumbricoides and the lower ERRs of all soil-transmitted helminth infections identified by FLOTAC requires further study , as it might strategically impact on future helminth control programs . For example , if one considers that anthelmintic drug efficacy is lower than generally assumed , one might conclude that preventive chemotherapy fails to bring prevalence and infection intensities to sufficiently low levels , and hence more emphasis should be placed on other preventive measures , such as health education , the implementation of sewage systems , and improving sanitation and access to clean water . Additionally , in case anthelmintic drugs are less efficacious than assumed , then the risk of resistance development is likely higher than expected , since a larger proportion of helminths survive chemotherapy , which might select for resistant strains [34] . Before generalizing these results one must consider , however , that our study design suffers from the following shortcomings: ( i ) the study was not adequately powered for clinically important findings; ( ii ) CRs and ERRs were estimated only for individuals who were found T . trichiura-positive by the Kato-Katz method at baseline and occasionally co-infected with A . lumbricoides or hookworm; ( iii ) the number of individuals infected with A . lumbricoides or hookworm at baseline and treated with anthelmintic drugs was low; ( iv ) only a single stool sample was examined per person at baseline and follow-up with the FLOTAC basic technique , therefore not accounting for day-to-day variation in helminth egg output [35]; and ( v ) perfect specificities were assumed for the Kato-Katz and FLOTAC method . Hence , the current sampling scheme , dictated by the primary outcome measure of the randomized controlled trial ( i . e . , drug efficacy against T . trichiura ) [24] might have biased our results and points ( iv ) and ( v ) might have resulted in an inaccurate estimate of the test sensitivity [14] , [36] . There are several causes for the high loss of samples from the baseline to the follow-up survey . First , 97 children did not submit a stool sample at follow-up . Among them 87 ( 89 . 7% ) were from the T . trichiura-negative , and hence non-treatment group , which we did not follow as rigorously as the treated children who were part of a randomized controlled trial [24] . Second , ∼1% of the stool samples could not be analyzed due to the flotation of stool debris directly under the examination grid of the FLOTAC apparatus . Third , another 1% of the samples were lost due to a sudden power cut . Fourth , a considerable number of results was not recorded or lost due to inappropriate labeling of the tubes , the FLOTAC apparatus or the Kato-Katz slides . Points two and three constitute limitations of the FLOTAC technique that should be taken into consideration in future studies . The flotation of debris can be overcome by an additional ether washing step of the stool sample to remove organic compounds . The ether washing step results in a more clearly examinable grid of the FLOTAC apparatus and improves the detection of A . lumbricoides and T . trichiura eggs [22] . It seems , however , to impact negatively on the detection of hookworm eggs [22] , and hence a more appropriate way to lower the contamination of the FLOTAC examination grid has to be found . The problems of sudden power cuts in resource-constrained countries can be overcome by the use of mirror-operated microscopes and by hand-operated centrifuges [37] . Of course , these options are more laborious and time consuming , and can hence not be considered as ideal solutions for large-scale surveys . Point four implicates human failure . Since five labeling or recording steps ( i . e . , preservation tube , weight records , centrifugation tube , apparatus , and result records ) are needed for FLOTAC , but only two ( i . e . , slide and result records ) for Kato-Katz , the FLOTAC method is more error-prone , especially if large numbers of stool samples are processed under time constraints . In general , the application of the FLOTAC technique is more complicated and expensive than the Kato-Katz method [38] . This needs to be considered when applying the FLOTAC in field laboratories and large-scale epidemiological surveys , where ease of examination is beneficial . While it is still too early to generalize the results reported here , and the FLOTAC technique might need further optimization for reliable diagnosis of hookworm infections , this new copro-microscopic technique holds promise for simultaneous detection of the three common soil-transmitted helminths , S . mansoni and intestinal protozoa infections [21] , [22] . If these issues are solved , we are confident that the FLOTAC can serve as a viable alternative to the Kato-Katz method for anthelmintic drug efficacy assessment and for monitoring and evaluation of deworming programs , particularly in settings where infections intensities have come down to low levels after repeated treatment . The lower CRs and ERRs identified by FLOTAC warrant more investigation and , if confirmed , could strategically impact on future helminth control programs . | In areas where parasitic worm infections have been successfully reduced as a result of deworming campaigns , the level of infections and drug efficacy must be carefully monitored . For this purpose , diagnostic methods with a high sensitivity are needed . We compared the accuracy of the widely used Kato-Katz method with the more recently developed FLOTAC technique for the diagnosis of parasitic worms . Our study was done with children on Zanzibar island , Tanzania , within the frame of an anthelmintic drug efficacy study . We collected stool samples from 343 children in two primary schools before and after treatment and examined the stool samples with both methods . FLOTAC showed a higher sensitivity than Kato-Katz for the diagnosis of roundworm and whipworm , but a lower sensitivity for hookworm diagnosis . The cure rates determined by FLOTAC were lower for roundworm and whipworm when compared with Kato-Katz . The opposite was found for hookworm . Egg reduction rates were generally lower when the FLOTAC technique was used . Our results suggest that the FLOTAC method , after additional optimization , can become a viable alternative to the Kato-Katz method for anthelmintic drug efficacy studies and for monitoring and evaluation of deworming programs . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"public",
"health",
"and",
"epidemiology"
]
| 2011 | Diagnostic Accuracy of Kato-Katz and FLOTAC for Assessing Anthelmintic Drug Efficacy |
In all sexually reproducing organisms , cells of the germ line must transition from mitosis to meiosis . In mice , retinoic acid ( RA ) , the extrinsic signal for meiotic initiation , activates transcription of Stra8 , which is required for meiotic DNA replication and the subsequent processes of meiotic prophase . Here we report that RA also activates transcription of Rec8 , which encodes a component of the cohesin complex that accumulates during meiotic S phase , and which is essential for chromosome synapsis and segregation . This RA induction of Rec8 occurs in parallel with the induction of Stra8 , and independently of Stra8 function , and it is conserved between the sexes . Further , RA induction of Rec8 , like that of Stra8 , requires the germ-cell-intrinsic competence factor Dazl . Our findings strengthen the importance of RA and Dazl in the meiotic transition , provide important details about the Stra8 pathway , and open avenues to investigate early meiosis through analysis of Rec8 induction and function .
Most eukaryotes reproduce sexually , with life cycles that display an alternation of diploid and haploid phases . The generation of haploid cells from diploid cells is achieved through meiosis , featuring a single round of DNA replication ( meiotic S ) followed by two rounds of division ( meiosis I and meiosis II ) . In all sexually reproducing organisms , including fungi , plants , and animals , cells of the germ line activate the meiotic program when conditions are opportune and appropriate to the species' reproductive strategies . In yeast , for example , the meiotic program is initiated only when diploid cells are starved for nutrients and cannot proliferate . In mammals , the meiotic program is initiated only after the specialized cells of the germ line have migrated to the gonad . The timing of mammalian meiotic initiation differs dramatically between the sexes . In males , meiotic initiation does not commence until a spermatogonial stem cell population has been established , well after birth . In females , meiosis is initiated shortly after the germ cells have entered the gonad , during fetal development . In mice , the published data are consistent with a model whereby an extrinsic meiosis-initiating signal – retinoic acid ( RA ) – induces transcription and expression of a single meiotic factor – Stra8 – which in turn governs the meiotic program [1]–[4] . In the ovary , induction of Stra8 in fetal germ cells expressing Dazl , an intrinsic factor , is required for meiotic DNA replication and the subsequent events of meiotic prophase [2] , [5] , [6] . In fetal testes , this process is temporarily blocked: CYP26B1 , a cytochrome p450 enzyme , degrades RA , preventing expression of Stra8 and thus precluding meiotic initiation [1] , [3] , [7] . After birth , RA induces Stra8 in testicular germ cells , leading to meiotic initiation [3] , [4] . Although the currently accepted model in mice postulates that RA induction of Stra8 may be necessary and sufficient for meiotic initiation [8] , evidence suggests that other , independent factors are also at play: germ cells in Stra8-deficient fetal ovaries express Rec8 [2] , encoding a meiosis-specific component of the cohesin complex . Rec8 is required for completion of sister chromatid cohesion , proper synapsis , and chiasmata formation [9] , [10] . We decided to examine how Rec8 expression is regulated during the meiotic transition and whether RA plays a role in its expression . Our investigation proceeded by first comparing the patterns and regulation of Rec8 and Stra8 expression and then exploring important differences with respect to their roles in driving meiotic initiation . We discovered that RA activates meiosis in two independent ways , both of which require Dazl expression in the germ cells .
We first sought to investigate how Rec8 expression is initiated in the germ cells of fetal ovaries . If Rec8 is regulated like Stra8 and other early meiotic markers , it should initiate expression in an anterior-to-posterior pattern between E12 . 5 and E16 . 5 [5] , [11] , [12] . Using whole mount in situ hybridization , we discovered that Rec8 expression does unfold this way from E13 . 0 to E16 . 0 ( Figure 1A ) . These findings suggested that Rec8 , like Stra8 , could be a target of RA signaling . Furthermore , since Dazl expression is required for ovarian germ cells to respond to RA signaling , perhaps , as with Stra8 expression , expression of Rec8 requires both DAZL and RA . We tested this new model ( Figure 1B ) in fetal ovaries , fetal testes and adult testes . We examined if RA signaling was required for Rec8 expression in the germ cells of fetal ovaries . We harvested ovaries at E12 . 5 and cultured them for two days in the presence of the RA receptor pan-antagonist BMS-204493 and then evaluated expression of both Stra8 and Rec8 using quantitative RT-PCR . BMS-204493 antagonizes all three RAR isotypes [13] and prevents RA signaling in fetal ovaries without killing the germ cells . We discovered that BMS-204493 dramatically lowered Rec8 expression , similar to Stra8 ( Figure 2A ) , indicating that , in wild-type fetal ovaries , RA signaling is required for the germ cells to express Rec8 . Taking these results together with our laboratory's previous finding that Stra8-deficient fetal ovaries express Rec8 [2] , we conclude that RA induces Rec8 in fetal ovaries independently of Stra8 . We next considered whether RA regulation of Rec8 expression resembles that of Stra8 in other respects . Germ cells in wild-type fetal testes express Stra8 when exposed to high levels of exogenous RA [3] , but germ cells in Dazl-deficient testes do not [6] . Thus , during meiotic initiation , the germ cells must express Dazl in order to respond to RA signaling . We tested whether RA-mediated upregulation of Rec8 expression similarly requires Dazl . We used quantitative RT-PCR to compare Rec8 expression levels in E12 . 5 Dazl-deficient testes cultured for two days with or without RA added to the medium ( Figure 2B ) . We found that , unlike Stra8 , Rec8 is expressed , albeit at very low levels , in wild-type and Dazl-deficient testes . However , similarly to Stra8 , Rec8 expression was significantly upregulated by RA treatment in wild-type but not Dazl-deficient testes ( Figure 2B ) . Thus RA-induced upregulation of Rec8 in embryonic testes depends on Dazl . RA also regulates Stra8 expression and meiotic initiation in germ cells of postnatal testes [3] , [4] . We examined whether Rec8 followed a similar pattern to Stra8 here as well . Since retinoic acid is a metabolite of vitamin A , vitamin A-deficient ( VAD ) mice can be used to evaluate the effects of dramatically reduced RA signaling on postnatal testes . We removed testes from several vitamin A-deficient adult males and VAD males with restored RA signaling ( 24 hours post RA injection ) and evaluated Rec8 and Stra8 transcripts by quantitative RT-PCR . Like Stra8 , Rec8 transcription was dramatically increased 24 hours after injection of RA ( Figure 2C ) . These results demonstrate that RA regulates Rec8 transcription in adult testes in vivo , as in fetal ovaries; this signaling event is shared between the sexes . To test whether this Rec8 upregulation in postnatal testes was Stra8-dependent , we examined Rec8 expression in Stra8-deficient , VAD testes before and after injection of RA . While the Stra8-deficient , RA-deficient VAD testes expressed very little Rec8 , restoration of RA resulted in dramatically increased expression of Rec8 ( Figure 2D ) . Thus , as in fetal ovaries , RA induces Rec8 expression in postnatal testes independently of Stra8 . Germ cells in Cyp26b1-deficient fetal testes express Stra8 and several other early meiotic factors at the same time as they do in fetal ovaries because of uninhibited RA signaling [1] , [7] , [14] ( Figure S1 ) . However , whether STRA8 protein is expressed and , if so , whether it influences other early meiotic factors has not been determined . We developed a system of single- and double-mutant mice with which to analyze in vivo the effects of RA signaling on germ cells in the presence and absence of STRA8 . We found that STRA8 protein is expressed in Cyp26b1-deficient fetal testes but not in double-mutant Cyp26b1-deficient/Stra8-deficient testes ( Figure 3A ) . We then assayed Rec8 expression in single- and double-mutant fetal testes using quantitative RT-PCR . In both cases , Rec8 expression is higher than in wild type , achieving similar levels in single- and double-mutant samples ( Figure 3B ) . High expression levels in the double mutant indicate that RA induction of Rec8 in Cyp26b1-deficient fetal testes is independent of Stra8 . In our studies above , we have established that RA regulates Rec8 , and that it does so in parallel to its other known target , Stra8 , in fetal ovaries , adult testes and in Cyp26b1-deficient fetal testes ( Figure 1B ) . Drawing on the comparative model we used to examine Rec8 expression in fetal testes , we explored whether RA regulates other early meiotic factors in parallel to Stra8 . We first tested whether ectopic RA signaling is sufficient to drive DNA replication in germ cells of fetal testes , and , if so , whether this effect is also mediated through STRA8 . The thymidine analog 5-bromo-2-deoxyuridine ( BrdU ) can be incorporated into newly synthesized DNA during S phase . We injected BrdU into pregnant females , dissected E16 . 5 fetal gonads and immunostained them with anti-GCNA ( a germ cell marker ) and anti-BrdU antibodies . In wild-type animals , testicular germ cells have arrested in G0/G1 by E16 . 5 . We can therefore detect ectopic germ cell proliferation in response to STRA8 upregulation by assaying for ongoing DNA replication in E16 . 5 fetal gonads . BrdU incorporation was evident in germ cells of Cyp26b1-deficient testes ( Figure 4A ) , consistent with transition towards meiosis . In contrast , GCNA-positive germ cells of double-mutant Cyp26b1-deficient/Stra8-deficient testes were uniformly negative for BrdU at E16 . 5 . We conclude that the DNA replication observed in germ cells of Cyp26b1-deficient fetal testes at E16 . 5 depends on and is mediated through STRA8 ( Figure 4A ) . We then determined if RA is sufficient in fetal testes to induce DNA double strand breaks ( DSBs ) , which are required for meiotic recombination [15]–[19] , and if the generation of these DSBs is mediated through STRA8 induction . We assayed for the presence of γH2AX , a phosphorylated histone variant that localizes to DSBs , by immunostaining at E15 . 5 , when DSBs are first observed [20] . Cyp26b1-deficient testes displayed many germ cells positive for γH2AX , suggesting that DSBs are induced by RA ( Figure 4B ) . In contrast , we rarely observed γH2AX-positive germ cells in double-mutant Cyp26b1-deficient/Stra8-deficient testes ( Figure 4B ) . This result suggests that the induction of DSBs in Cyp26b1-deficient testes is driven by ectopic RA and STRA8 . Since DSBs arise not only during meiotic recombination but also during apoptosis [21] , and apoptosis has been reported in Cyp26b1-deficient testes [7] , we tested whether γH2AX-positive germ cells observed in Cyp26b1-deficient testes represent meiotic and not simply apoptotic events . We first generated double mutant Cyp26b1-deficient/Bax-deficient embryos . Bax is a proapoptotic gene , and its deletion has been shown to suppress apoptosis in germ cells [14] , [22] , [23] ( Figure S2 ) . Staining in double-mutant Cyp26b1-deficient/Bax-deficient testes revealed many γH2AX-positive germ cells ( Figure 4B ) , confirming that most γH2AX-positive germ cells observed in Cyp26b1-deficient testes represent meiotic rather than apoptotic DNA DSBs . Formation of meiotic DNA DSBs thus represents another portion of the meiotic pathway that is STRA8-mediated . Meiotic DSBs are processed by DMC1 , an ortholog of the bacterial strand exchange protein RecA , which commences expression early during meiotic initiation . We compared the effects of RA on Dmc1 expression in Cyp26b1-deficient testes and in double-mutant ( Cyp26b1-deficient/Stra8-deficient ) testes . The Cyp26b1-deficient testes displayed increased levels of Dmc1 , while levels of Dmc1 in double-mutant testes were similar to controls ( Figure 4C ) . Thus , RA is sufficient to drive Dmc1 expression in fetal testes in vivo , but this induction requires mediation by STRA8 . In summary , it appears that RA induction of STRA8 in fetal testes is required for all of the above-tested markers/processes during early meiosis , with the notable exception of RA-regulated Rec8 expression . To exclude the possibility that induction of Stra8 and its downstream target Dmc1 depends on Rec8 function , we examined Stra8 and Dmc1 expression in Rec8-deficient ( Rec8mei8/mei8 ) ovaries and testes [9] . As expected , we found no significant difference in Stra8 and Dmc1 expression levels between control and Rec8-deficient E13 . 5 ovaries ( Figure 5A ) . Similarly , we detected STRA8 and DMC1 proteins in both control and Rec8-deficient adult testes ( Figure 5B ) . We conclude that RA induction of Stra8 , and its downstream targets , is independent of and occurs in parallel with RA induction of Rec8 .
Our findings lead us to conclude that RA plays a broad and encompassing role in regulating and coordinating the transition from mitosis to meiosis in mouse germ cells , in both fetal ovaries and postnatal testes . Surprisingly , RA accomplishes this by independently inducing both Stra8 and Rec8 , which both play critical roles in the earliest stages of meiosis . The discovery that RA induction of Stra8 in Cyp26b1-deficient fetal testes mediates DNA replication , DSB formation , and the expression of recombinase Dmc1 provides critical details about the Stra8 pathway . Moreover , Stra8 induction was recently shown to be required for SYCP3 expression in Cyp26b1-deficient testes [24] . Rec8 induction is the first component of the molecular program of meiotic initiation shown to be Stra8-independent in mice . Now that Rec8's independent induction has been established , its expression pattern and function invite deeper investigation . How Rec8 expression is induced by RA remains elusive . Stra8's promoter region contains two putative RA Response Elements ( RAREs ) , suggesting that RA could be turning on this gene directly [25] . A chromatin immunoprecipitation-sequencing ( ChIP-Seq ) study in embryonic stem cells identified RAR binding sites in both Stra8 and Rec8 promoter regions , suggesting that Rec8 may also be regulated by RA directly [26] . Intriguingly , in the same study , Dmc1 , which is dependent on STRA8 , does not show such RAR binding sites , consistent with Stra8 and Rec8 being regulated directly , unlike Stra8's downstream targets . What purpose does RA upregulation of REC8 serve ? It may ensure that Rec8 is expressed during pre-meiotic S phase so that its product can be incorporated into the meiotic cohesin complex that joins sister chromatids . Indeed , germ cells in Rec8-deficient mice later show defects that can be traced to its cohesion function – incorrect synapsis topology and failure at chromosome segregation and chiasmata formation [9] , [10] . Recent studies also suggest a role for cohesins in direct regulation of gene expression by novel mechanisms involving DNA looping [27] , [28] . It is presently unknown if Rec8 is a direct transcriptional regulator . However , Rec8 null animals exhibit partial embryonic lethality and fail to thrive [10] , phenotypes hard to reconcile with an exclusive role in germ cell meiotic cohesion . The mechanisms that govern meiotic initiation have been explored most thoroughly in yeast , and these studies offer interesting parallels to our findings in mice . In both yeast and mice , the decision to initiate the meiotic program is taken prior to pre-meiotic DNA replication [2] , [29] . Our finding that RA regulates Rec8 is consistent with an early role of RA in this transition , since at least in budding yeast , REC8 associates with chromosomes from late G1 phase [30] . In addition , in both yeast and mice , the decision to initiate meiosis requires an extrinsic signal and an intrinsic competence factor [1] , [3] , [6] , [31] , [32] . In yeast , the extrinsic signal – nutrient depletion – activates multiple molecular pathways in parallel , and these converge on IME1 , which is required for upregulating the expression of meiosis-specific transcripts . However , IME1 is not sufficient to induce meiosis in yeast [33] , [34] . Our studies show that , analogously , RA activates at least two pathways by regulating Stra8 and Rec8 independently . While many early meiotic processes described so far hinge on STRA8 , STRA8 may not be sufficient for meiosis in mice . The search for additional RA targets will likely yield further insights into the networks governing transition from mitosis to meiosis in mammals .
All experiments involving mice were approved by the Committee on Animal Care at the Massachusetts Institute of Technology . Cyp26b1-deficient mice were generated by deleting a 2 . 9-kb portion of the gene ( including exons 4 , 5 , 6 , and the coding region of exon 7 ) by homologous recombination in embryonic stem ( ES ) cells ( Figure S1 ) . A Cyp26b1/PGK-Neo targeting construct was assembled using PCR products amplified with Advantage HF2 polymerase ( Clontech ) using mouse ( C57BL/6J ) genomic BAC RP24-470O13 ( GenBank Accession AC159337 ) as template . The targeting construct was linearized and electroporated into v6 . 5 ES cells [35] . Cells harboring the construct were selected using neomycin ( Invitrogen ) . ES cell colonies were screened by PCR for homologous integration at both the 5′ and 3′ arms of the construct . Clones that tested positive by both PCR assays were confirmed by Southern blot analysis using EcoRV and Nde1 restriction endonucleases . Correctly targeted ES cell clones were injected into Balb/c or C57Bl/6N blastocysts and transferred to pseudopregnant Swiss Webster females . Germline transmission was obtained with one clone , and the resulting homozygous embryos displayed anomalies of limb , eye , and facial development and died at birth , as previously described [7] , [36] . Embryos were genotyped by PCR , ( primer sequences available in Note S1 ) . Mice carrying the DazlTM1Hgu allele [37] were generously provided by Howard Cooke , MRC Human Genetics Unit , Western General Hospital , Edinburgh , UK , and Dazl-deficient mice were generated as described previously [6] , [38] . Stra8-deficient mice were generated as described previously [2] , [4] . Bax-deficient mice were generated by mating Baxtm1Sjk/+ mice obtained from The Jackson Laboratory ( Bar Harbor , ME ) . Rec8-deficient mice were generated by mating Rec8mei8/+ mice [9] , which were generously provided by John Schimenti , Cornell University , Ithaca , New York . Mouse embryos used in whole mount in situ hybridizations and gonad cultures were obtained from matings between CD1 random bred mice ( Charles River Labs ) . Noon of the day when vaginal plug was recorded was considered E0 . 5 . Whole mount in situ hybridizations with the Stra8 probe were performed as previously described [3] , [39] . Digoxigenin riboprobe for Rec8 was generated by amplifying cDNA fragments by RT-PCR from Rec8 ( NM_020002 . 2: bases 274–865 ) , and inserting them into TA cloning vector pCR4-TOPO ( Invitrogen ) . Plasmid was linearized with Spe1 or Not1 and transcribed with T7 or T3 respectively to make the antisense and sense probes . For experiments involving Rec8-deficient mice , total RNAs were prepared from gonads using the RNeasy plus Micro RNA isolation kit ( QIAGEN ) , and reverse transcription was carried out using the high-capacity cDNA reverse transcription kit ( Applied Biosystems ) . For all other experiments , total RNAs were prepared using TRIzol ( Invitrogen ) extraction followed by DNase ( Ambion ) treatment , and reverse transcription was carried out using the RETROscript reverse transcription kit ( Life Technologies ) . The resulting total cDNAs were analyzed quantitatively using SYBR Green PCR reagents ( Applied Biosystems ) with primers for Dmc1 , Rec8 , Stra8 , or Dazl . Expression profiles were tested in triplicate on at least two litters of embryos on an ABI 7500 instrument ( Applied Biosystems ) . Data were analyzed using the comparative Ct ( ΔΔCt ) method and one-tail , unpaired student T test ( significance cutoff p<0 . 01 ) . Results were normalized to Rps2 ( VAD experiments on adult testis ) , Dazl ( Rec8-mutant experiments on embryonic ovary ) , and Hprt ( all other experiments ) . Primers were selected from PrimerBank [40] ( Note S1 ) . Fetal gonads were dissected in phosphate buffered saline ( PBS ) , fixed in 4% paraformaldehyde overnight at 4°C , embedded in paraffin and sectioned . Slides were incubated with anti-GCNA IgM ( courtesy of G . Enders , undiluted supernatant ) , anti-STRA8 ( Abcam . 1∶100 ) , and anti-phosphoH2A . X ( Upstate Cell Signaling Solutions , 1∶250 dilution ) . Colorimetric staining was performed using ABC reagents ( Vector Laboratories ) and developed with DAB peroxidase substrate ( Vector Laboratories ) . Sections were mounted in Vectashield Medium with DAPI ( Vector Laboratories ) , and fluorescent staining was obtained using Texas-Red or FITC-conjugated secondary antibodies ( Jackson Immunoresearch Laboratories , 1∶500 dilution ) . Adult testes were fixed in Bouin's solution overnight at 4°C , washed with PBS and 70% ethanol , embedded in paraffin , and sectioned at 5 µm thickness . Slides were matured overnight , de-waxed , rehydrated , and heated in 10 mM sodium citrate buffer ( pH 6 . 0 ) for antigen retrieval . Sections were incubated in 3% hydrogen peroxide for 5 min and blocked in 2 . 5% normal horse serum ( Vector Laboratories ) for 80 minutes at room temperature . Later , slides were incubated overnight with anti-STRA8 ( Abcam , 1∶500 ) or anti-DMC1 ( Santa Cruz Biotechnology , 1∶50 dilution ) . The following day , slides were washed three times in PBS and incubated with anti-rabbit ImmPRESS peroxidase reagent ( Vector Laboratories ) for 30 minutes . The slides were later developed using a DAB substrate kit ( Vector Laboratories ) for 1 minute . The slides were counterstained with Mayer's hematoxylin for 5 minutes and washed in running water , dehydrated , and mounted with Permount ( Fisher Scientific ) . Apoptotic cells were detected in paraffin sections of fetal testes using the Fluorescein in situ Cell Death Detection Kit ( Roche Applied Science ) and mounted in Vectashield Medium with DAPI ( Vector Laboratories ) . Pregnant females were injected with 5-bromo-2-deoxyuridine ( BrdU ) solution ( 50 mg/kg ) at 18 . 5 days post coitum . Six hours later , fetal gonads were dissected . Gonads were then fixed in 4% paraformaldehyde overnight at 4°C , embedded in paraffin , and sectioned . Prior to antibody application , sections were treated with denaturing reagent ( 3 . 5N HCl ) for 2 min . Incorporated BrdU was detected using anti-BrdU ( Accurate Chemical & Scientific Corp . , 1∶500 dilution ) in anti-GCNA IgM supernatant . Pregnant female mice were sacrificed by cervical dislocation and embryos were removed into PBS solution . After determining tail somite number , fetal ovaries and mesonephroi were dissected . One gonad from each embryo was then placed in a 35 µl droplet of culture media ( DME +10% FBS ) supplemented with either 5 µM pan-RAR inhibitor BMS-204493 ( Bristol-Myers Squibb ) or all trans RA ( Sigma ) dissolved in ethanol in a Petri plate . Control media contained vehicle ( ethanol ) alone . Petri plates were then inverted and placed within larger plates containing water and incubated at 37°C with 5% CO2 . Media was replaced after 24 hours . After 48 hours , tissue was removed from media , mesonephroi were dissected off and ovaries were placed individually into TRIzol reagent ( Invitrogen ) . Samples were then processed for quantitative RT-PCR as described above . Adult female mice ( 129/SvJ ) were fed a Vitamin-A-Deficient ( VAD ) diet ( Harlan Teklad , Indianapolis ) for at least 2 weeks before mating and throughout pregnancy . Their male offspring were fed a VAD diet for 13–14 weeks . In the first experiment with wild-type animals , one testis was removed from each animal and cut into two pieces; one fixed in Bouin's solution for histological assessment of spermatogenesis and the other placed in TRIzol ( Invitrogen ) for RNA extraction to serve as a pre-injection control in RT-PCR analysis . Incisions were sutured and the animals recovered for 24 h . Three animals with similarly deficient spermatogenesis ( as judged by pre-injection testicular histology ) were injected with 100 µl of 7 . 5 mg/ml all-trans retinoic acid ( Sigma ) in 10% ethanol/90%sesame oil solution . The animals' remaining testes were harvested 24 h after injection . In contrast , both testes were harvested from two Stra8-deficient VAD animals at the same time ( one was analyzed histologically to confirm depletion ) and compared to testes harvested from two RA-restored Stra8-deficient animals . Quantitative RT-PCR analysis was performed , in triplicate , using Stra8 and Rec8 primers , and Rps2 was used as a normalization control ( primer sequences in Note S1 ) . | The transition from mitosis to meiosis is a defining feature of germ cells , the precursors of eggs and sperm . In mice , retinoic acid ( RA ) , a vitamin A derivative , induces expression of the gene Stra8 , which in turn is required for the first critical steps of meiosis . The timing of Stra8 expression in mammalian germ cells is influenced by an RA-degrading enzyme , CYP26B1 , that is normally expressed in fetal testes to delay meiosis in males . It is unknown if Stra8 is RA's only meiosis-inducing target in germ cells or if other such genes are regulated by RA independently of Stra8 . To investigate this question , we generated two lines of mice: Cyp26b1 mutants and Stra8 mutants . Our genetic experiments comparing germ cell development in these two mutants revealed a new RA target , Rec8 . We demonstrate that Rec8 upregulation by RA occurs in the same temporal and spatial manner as Stra8 , but Rec8 expression is independent of Stra8 . Rec8 , like Stra8 , plays a critical role during early meiotic processes , suggesting that RA induces meiosis in at least two independent pathways . These findings expand our understanding of the gene regulatory network involved in meiotic initiation in mammals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"animal",
"models",
"animal",
"genetics",
"model",
"organisms",
"gene",
"regulatory",
"networks",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"research",
"and",
"analysis",
"methods",
"gene",
"function"
]
| 2014 | Retinoic Acid Activates Two Pathways Required for Meiosis in Mice |
As most of the heritability of complex traits is attributed to common and low frequency genetic variants , imputing them by combining genotyping chips and large sequenced reference panels is the most cost-effective approach to discover the genetic basis of these traits . Association summary statistics from genome-wide meta-analyses are available for hundreds of traits . Updating these to ever-increasing reference panels is very cumbersome as it requires reimputation of the genetic data , rerunning the association scan , and meta-analysing the results . A much more efficient method is to directly impute the summary statistics , termed as summary statistics imputation , which we improved to accommodate variable sample size across SNVs . Its performance relative to genotype imputation and practical utility has not yet been fully investigated . To this end , we compared the two approaches on real ( genotyped and imputed ) data from 120K samples from the UK Biobank and show that , genotype imputation boasts a 3- to 5-fold lower root-mean-square error , and better distinguishes true associations from null ones: We observed the largest differences in power for variants with low minor allele frequency and low imputation quality . For fixed false positive rates of 0 . 001 , 0 . 01 , 0 . 05 , using summary statistics imputation yielded a decrease in statistical power by 9 , 43 and 35% , respectively . To test its capacity to discover novel associations , we applied summary statistics imputation to the GIANT height meta-analysis summary statistics covering HapMap variants , and identified 34 novel loci , 19 of which replicated using data in the UK Biobank . Additionally , we successfully replicated 55 out of the 111 variants published in an exome chip study . Our study demonstrates that summary statistics imputation is a very efficient and cost-effective way to identify and fine-map trait-associated loci . Moreover , the ability to impute summary statistics is important for follow-up analyses , such as Mendelian randomisation or LD-score regression .
Genome-wide association studies ( GWASs ) have been successfully applied to reveal genetic markers associated with hundreds of traits and diseases . The genotyping arrays used in these studies only interrogate a small proportion of the genome and are therefore typically unable to pinpoint the causal variant . Such arrays have been designed to be cost-effective and include only a set of tag single nucleotide variants ( SNVs ) that allow the inference of many other unmeasured markers . To date , thousands of individuals have been sequenced [1 , 2] to provide high resolution haplotypes for genotype imputation tools such as IMPUTE and minimac [3 , 4] , which are able to infer sequence variants with ever-increasing accuracy as the reference haplotype set grows . Downstream analyses such as Mendelian randomisation [5] , approximate conditional analysis [6] , heritability estimation [7] , and enrichment analysis using high resolution annotation ( such as DHS ) [8] often require genome-wide association results at the highest possible genomic resolution . Summary statistics imputation [9] has been proposed as a solution that only requires summary statistics and the linkage disequilibrium ( LD ) information estimated from the latest sequencing panel to directly impute up-to-date meta-analysis summary statistics [10] . Because summary statistics imputation uses summarised data as input , it is not bounded to privacy restrictions related to the use of individual data . Another advantage is its substantially lower computation time compared to genotype imputation . For example , for imputation of the UK Biobank data , it is about 500 times faster ( 4200 vs 8 . 3 CPU days comparing Minimac [4] to our SSIMP software [11] ) . This study compares summary statistics imputation directly to genotype imputation and focuses on its practical advantages using real data . In particular , we evaluated two experiments: 1 ) we ran a GWAS on both simulated traits and human height using data from 120′086 individuals from the UK Biobank and compared the performances of summary statistics imputation and genotype imputation , using direct genotyping/sequencing as gold standard; 2 ) we imputed association summary statistics from a HapMap-based GWAS study [12] using the UK10K reference panel to explore new potential height-associated variants which we validated using results from Marouli et al . [13] and the UK Biobank height GWAS ( n = 336′474 ) . We extended summary statistics imputation [9 , 14] which yields increased imputation accuracy by accounting for variable sample sizes . For all applications presented in this manuscript we are using this improved version of summary statistics imputation .
By combining summary statistics for a set of variants and the fine-scale LD structure in the same region , we can estimate summary statistics of new , untyped variants at the same locus . We assume a set of univariate effect size estimates ai are available for SNVs i = 1 , … , I from a linear regression between a continuous phenotype y and the corresponding genotype gi measured in N individuals . Without loss of generality we assume that both vectors are normalised to have zero mean and unit variance . Thus a i = ( g i ) ′ · y N and a= ( a1 , a2 , … , aI ) ′∼N ( α , Σ ) . Σ represents the pairwise covariance matrix of effect sizes of all i = 1 , … , I SNVs . To estimate the univariate effect size αu of an untyped SNV u in the same sample , one can use the conditional expectation of a multivariate normal distribution . The conditional mean of the effect of SNV u can be expressed using the effect size estimates of the tag SNVs [9 , 15]: a ^ u = a u | M = α u + Σ u M Σ M M - 1 ( a - α ) , ( 1 ) where M is a vector of so-called tag SNVs , Σ u M represents the covariance between SNV u and all M markers and Σ M M represents the covariance between all M markers . We assume that estimates for the two covariances are available from an external reference panel with n individuals and denote them s = Σ ^ M u , S = Σ ^ M M . The corresponding correlation matrices are γ and Γ , with c = N ⋅ s and C = N ⋅ S being the estimates for the correlation matrices . Further , by assuming that SNV u and the trait are independent conditioned on the M markers , i . e . α u - Σ u M Σ M M - 1 α = 0 , Eq ( 1 ) becomes a ^ u = a u | M = s ′ S - 1 a = c ′ C - 1 a ( 2 ) One can also choose to impute the Z-statistic instead , as derived by Pasaniuc et al . [9]: z ^ u | M = c ′ C - 1 z ( 3 ) with z = a N , when the effect size is small ( as is the case in typical GWAS ) . Similar to Pasaniuc et al . [9] , we chose M to include all measured variants within at least 250 Kb of SNV u . To speed up the computation when imputing SNVs genome-wide , we apply a windowing strategy , where SNVs within a 1 Mb window are imputed simultaneously using the same set of M tag SNVs the 1 Mb window plus 250 Kb flanking regions on each side . We simulated genetic data on 25’000 individuals was used . In brief , we used data from the five European subpopulations CEU , GBR , FIN , TSI and IBR of the 1000 Genomes reference panel [1] . We chose to up-sample chromosome 15 using HAPGEN2 [29] to 5′000 individuals for each subpopulation , yielding a total of 25′000 individuals . Of these , half of the data was used to estimate the LD structure C and the other half to simulate the association study with an in silico phenotype . The simulation procedure is described in more detail in S1 Appendix . Forty regions were selected with one non-HapMap causal variant in each and all HapMap SNVs were used as tag SNVs . Sample size distributions were drawn from two published GWAS studies ( on HDL [30] and T2D [31] ) . Missingness was assigned at random positions while respecting the missingness correlation parameter θmiss , with zero value reflecting missingness at random and one corresponding to the maximum possible sample overlap between SNVs . To estimate LD structure in C and c ( Eq ( 2 ) ) we used 3′781 individuals from UK10K data [32 , 33] , a reference panel of British ancestry that combines the TWINSUK and ALSPAC cohorts . All analysis was performed with R-3 . 2 . 5 [34] programming language , except GWAS summary statistics computation for UK Biobank genotype and genotype imputed data , for which SNPTEST-5 . 2 [35] was used . For summary statistics imputation we used SSIMP [11] .
The conventional estimate of the standardised effect of a SNV u , a ^ u ( c o n v ) , ( Eq ( 2 ) ) is unbiased , under certain assumptions , but can have large variance when there is variation in the sample sizes recorded in N M . In this section , we used upsampled 1000 Genomes data [1] and simulated phenotype with known standardised effect α and various missingness design . We compare the MSE of the conventional estimation to the MSE of two other estimators , Eq ( 13 ) using D ( dep ) and D ( ind ) , derived in the method section . In general , the size of the overlap is unknown and we recommend using the assumption of maximum dependence ( D ( dep ) ) as it is the most conservative assumption . An alternative is to assume randomly distributed missingness ( D ( ind ) ) . Most pairs of SNVs in GIANT attain close to the maximum possible missingness-overlap ( S10 Fig ) and therefore this assumption is not overly-conservative . The results in Fig 3 demonstrate that the conventional method has the largest MSE across all the simulation parameters tested . Where the variance in sample size is very large ( top row of Fig 3 ) , the true correlation is often very close to zero . Both of our methods effectively make this same ( correct ) assumption of low correlation and therefore they both perform equally well . Where the variation in sample size is less extreme , as in the simulations on the bottom row of Fig 3 , there is less shrinkage of correlation and the simulated missingness correlation becomes more relevant . Where the simulated data has the maximum possible missingness correlation ( on the right hand side of the subplots in Fig 3 ) , i . e . the sample overlap between each pair of SNVs is as large as possible given their two sample sizes , D ( dep ) performs better ( as expected ) . With lower overlap ( first column ) D ( ind ) performs better . By having two types of genetic data at hand , genotype and imputed genotype data , we were able to compare summary statistics of 37′467 typed SNVs resulting from ( 1 ) associations calculated from original genotype data ( ground truth ) ; ( 2 ) associations calculated from imputed genotype data ( genotype imputation ) and ( 3 ) associations imputed from summary statistics calculated using genotype data ( Fig 1 ) . For our analysis , we defined 706 genomic regions in total , among which 535 contain SNVs associated with height [12 , 13] , while the remaining 171 regions were selected to be free of any known height associated SNVs . We examined imputation results for different SNV categories . These were grouped based on ( i ) their association status ( being correlated with the causal SNV vs . null SNVs ) with the lead SNV of each of the 535 height-associated regions ( 6′080 variants were correlated , 31′567 were not ) ; ( ii ) frequency ( MAF: 1% < low-frequency ≤ 5% < common; 13′857 and 23′790 variants , respectively ) ; and ( iii ) imputation quality based on summary statistics imputation ( r ^ pred , adj 2: low ≤ 0 . 3 < medium ≤ 0 . 7 < high; 724 , 9′792 , and 27′131 variants , respectively ) . S1 and S2 Figs show the distribution of SNV counts in each of these twelve subgroups . We term the 6′080 SNVs correlated with a height-associated lead SNV as associated SNVs . Conversely , we refer to the 31′567 SNVs that are not correlated with any height-associated lead SNV as null SNVs . For both , null and associated SNV groups , the largest group of analysed variants were common and well-imputed ( S1 Fig ) . The fraction of SNVs with low quality imputation increases with lower minor allele frequency ( S2 Fig ) . However , the number of rare variants ( MAF < 1% ) were too small ( 2′411 variants , among these only 13 associated variants ) , similar to the number of badly-imputed SNVs ( 724 variants , among these only one associated variant ) to draw meaningful conclusions and hence we limited our analysis to common and low-frequency , and medium- and well-imputed variants . We focused on two aspects of the imputation results . First , we compared how summary statistics imputation and genotype imputation perform relative to the ground truth ( direct genotyping ) . For this we used four measures: the root mean squared error ( RMSE ) , bias , the linear regression slope , and the correlation . Second , we calculated power and false positive rate for genotype imputation and summary statistics imputation directly . While previous studies have examined the role of ( common ) HapMap variants for height [12 , 36] , the impact of rare coding variants could not be investigated until bespoke genotyping chips ( interrogating low-frequency and rare coding variants ) were designed to address this question in a cost-effective manner . Such an exome chip based study was conducted by the GIANT consortium in 381′000 individuals and revealed 120 height-associated loci , of which 83 loci were rare or low-frequency [13] . These association results enabled us to compare the usefulness of imputation-based inference with direct genotyping done in Wood et al . [12] , since the two studies are highly comparable in terms of ancestry composition and statistical analysis , evidenced by S6 Fig confirming very high concordance between summary statistics for the subset of 2′601 SNVs correlated to a height-associated variant which were available in both studies .
Imputation accuracy is affected by the varying sample size across tag SNVs . If two SNVs were observed in two different samples , the correlation between the summary statistics will decrease with the number of individuals in common between the two samples . Our approach addresses this problem by shrinking the correlation matrix according to sample size overlap . We present two ways of estimating this overlap: D ( ind ) for independent missingness , which is randomly distributed; and D ( dep ) for dependent missingness , which is highly correlated . To evaluate the performance ot these two methods we simulated data with two different distributions of missingness ( narrow or wide range of sample sizes ) and varying correlation in missingness between variants ( from completely random to maximal overlap , Fig 3 ) . We then compared the performances of conventional summary statistics imputation and our proposed dependent ( D ( dep ) ) and independent ( D ( ind ) ) approaches . Overall , replacing C and c with D and d yields a lower RMSE . Furthermore , we note that the dependent approach has lower RMSE when the sample size variance is low and the missingness correlation approaches one . S15 Fig shows the comparison between the conventional estimation and using D ( dep ) for imputing GIANT height association summary statistics . Ideally , for any pair of SNVs that are in LD with each other , we would know the exact number of individuals that are in the overlap , i . e . the number of individuals for which both SNVs were genotyped . Using the individual study missingness and sample sizes from the Genetic Investigation of ANthropometric Traits ( GIANT ) consortium , we demonstrate in Fig . S10 Fig that the size of the overlap is generally larger than would be the case under a strict ‘missing independently at random’ assumption . Furthermore , the correlation of missingness is typically positive ( N k ∩ l > N k N l N m a x ) and often approaches the maximum possible overlap ( Nk∩l = min ( Nk , Nl ) ) . The reason for this is that SNPs are either entirely missing from a study or being available for all study participants depending on its genotyping chip or imputation panel , which induces positive missingness correlation between markers . We compared summary statistics imputation and genotype imputation by using individual-level data from the UK Biobank . In general , imputation using summary statistics imputation leads to a larger RMSE than genotype imputation in all twelve SNV subgroups investigated ( Fig 6 ) . Among associated SNVs , summary statistics imputation performs similar to genotype imputation for well-imputed SNVs , but shows a trend for underestimation of the Z-statistics and lower correlation with the true effect size for medium-imputed SNVs ( Fig 4 ) . Conversely , genotype imputation has more consistent results for most of the twelve SNV subgroups ( Figs 4 and 5 ) , that is reflected in a correlation close to one between Z-statistics from genotype data and genotype imputation data . When investigating power and FPR for both methods ( Fig 7 ) we observe that for a given significance threshold , summary statistics imputation has lower power compared to genotype imputation , an effect that is amplified for SNVs with lower imputation quality ( r ^ pred , adj 2 ≤ 0 . 7 ) and lower MAF ( MAF ≤ 5% ) . Ultimately , the underestimation of imputed Z-statistics with summary statistics imputation leads to a lower type I error . This effect is amplified for SNV groups with lower imputation quality ( r ^ pred , adj 2 < 1 ) . For associated SNVs with r ^ pred , adj 2 < 1 we expect an underestimation for associated SNVs due to the fact that we are imputing summary statistics under the null model , whereas for null SNVs with r ^ pred , adj 2 < 1 we expect an underestimation due to decreased variance of the summary statistics imputation estimation . Ideally , for an unbiased estimation of causal and null SNVs , the imputed Z-statistics ( Eq ( 2 ) ) should be divided by r ^ 2 . However , as the imputation quality r ^ pred , adj 2 is noisily estimated from small reference panels ( discussed below ) and it is not guaranteed that the SNV we impute is causal , we risk to overestimate the summary statistics of associated SNVs . This is the reason why refrain from doing so . S9 Fig shows the P-value distribution of summary statistics imputation for null SNVs with an accumulation of low P-values for well-imputed SNVs and an accumulation of high P-values for badly-imputed SNVs . We think that two factors are in play here . First , mostly due to polygenicity , the genomic lambda for height is λGC = 1 . 94 , therefore we expect even seemingly null variants to show inflation . Second , for null SNVs , the sample variance of the imputed Z-statistics should be proportional to the average imputation quality . We calculated for each of the null SNV subgroups the ratio between the sample variance for Z-statistics from summary statistics imputation and the sample variance for Z-statistics from genotype data . For common null SNVs we observe a ratio that gradually decreases with imputation quality ( 0 . 86 for perfectly- , 0 . 61 for medium- and 0 . 32 for badly imputed SNVs ) . For low-frequency null variants the ratio is up to 0 . 6 lower ( 0 . 80 for perfectly- , 0 . 54 for medium- and 0 . 30 for badly imputed SNVs ) . The inflation for well-imputed SNVs can be explained by the genomic lambda , while for badly-imputed SNVs it is aggravated by the underestimated standard error . Because the number of associated SNVs with MAF < 1% was too low ( 13 variants ) to draw any meaningful conclusions , we refrained from analysing this MAF group . One other reason to exclude rare variants from this analysis is , that the reference panel used ( UK10K ) contains 3′871 individuals and therefore estimations for LD of rare variants are unreliable and rare variants can ( in theory ) only be covered down to MAF = 1/ ( 2 ⋅ 3′871 ) . We believe improving summary statistics imputation for rare variants will require not only larger reference panels to allow estimation of LD of rare variants , but also methods which would allow non-linear tagging of variants . It should be kept in mind that , just like for genotype imputation , even with very large reference panels , one will not be able to impute variants with extremely rare allele counts . To investigate these SNVs full genome sequencing is indispensable [42] . We find that our imputation quality measure r ^ pred , adj 2 is conservative and probably underestimates the true imputation quality ( S4 Fig ) . To calculate the imputation quality r ^ pred , adj 2 , we need—similar to imputing summary statistics in Eq ( 2 ) —to compute correlation matrices c and C estimated from a reference panel ( Eq ( 8 ) ) and therefore encounter similar challenges as summary statistic imputation itself due to difficulties of reliable LD estimation . The discrepancy in imputation quality metric between summary statistics imputation and genotype imputation ( S4 Fig ) can be explained by the fact that: ( 1 ) genotyped variants that were imputed too , were also used for phasing , ( 2 ) it is indeed more difficult to impute summary statistics using summary statistics imputation , and therefore the imputation quality is shifted towards zero , and ( 3 ) r ^ pred , adj 2 is an estimation that can either be erroneous due to choosing the wrong reference panel ( and therefore r ^ pred , adj 2 does not represent the true imputation quality ) or it can be imprecise due to small sample size of the reference panel . For example , UK10K contains 3′871 individuals and is too small to precisely estimate these matrices ( the standard error for a correlation estimated from n = 3′871 is 0 . 016 ) , which becomes problematic in cases of low correlation . As a showcase of the utility of summary statistics imputation we imputed Wood et al . [12] to higher genomic resolution ( limited to variants with MAF ≥ 0 . 1% as well as 111 previously reported exome variants ) [13] , then selected imputed variants that act independently from all variants reported in Wood et al . and from each HapMap SNP , we then replicated these using ( independent ) UK Biobank data . While Wood et al . [12] is the largest height study to date in terms of number of markers ( covering HapMap variants in 253′288 individuals ) , Marouli et al . [13] exceeds their sample size by more than 100′000 individuals , but is limited to 241′419 exome variants . The similarity between both GIANT studies made the exome chip study ideal for replication . We chose the UK Biobank as a second replication dataset , despite its limitation to individuals of British ancestry , as it covers more variants than the exome chip study . We identify 35 regions , of which one had already been identified in the recent GIANT height exome chip study ( rs28929474 ) and 19 replicated in UK Biobank ( at α = 0 . 05/35 level ) . Two candidate loci ( #2 , #3 in Table 2 ) that replicate in UK Biobank have borderline significant HapMap signals in close proximity ( P-value between 10−6 and 10−8 in [12] ) and were therefore not reported in the study in 2014 . The 15 non-replicating candidate loci were on average on a lower allele frequency spectrum ( ten are rare , three are low-frequency variants , and two are common ) . Allele frequency was higher among the 20 replicating candidate variants ( 19 were common and one a low-frequency variant ) . We also ran an additional approximate conditional analysis , where we conditioned each of the 35 variants onto their neighbouring HapMap SNP ( one-by-one ) . The resulting maximum conditional P-value per locus , is provided as an additional column S1 Table . Correcting for the testing of 529 windows ( α = 0 . 05/529 ) we find evidence that 18 of the 35 variants are not only independent from all [12] reported SNPs , but also of each HapMap variant too . For replication of summary statistics from European individuals we use the UK Biobank , which represents only a subset of all European ancestries and is genotype-imputed ( instead of sequenced ) , but on the other hand provides a reliable resource due to its sample size . Furthermore , in UK Biobank , genotype imputation done for genotyped variants can only partially be compared to genotype imputation for untyped variants , as genotyped variants were used for phasing ( therefore genotype imputation of genotyped variants is easier and leads imputation qualities close to one , S4 Fig ) . Due to the small number of height-associated rare variants ( 13 ) we can not draw meaningful conclusions for this group and hence avoided their analysis . The choice of the reference panel to conduct summary statistics imputation depends on the fine balance between maximising the sample size of the reference panel ( which determines the error in estimated LD ) and matching the population diversity of the conducted GWAS . At the first glance , 1000 Genomes reference panel could have been used to appropriately match GIANT allele frequencies , however , the 8-fold higher sample size of UK10K panel offers a larger benefit , ultimately reducing the RMSE [43] . For the simulation study comparing standard summary statistics imputation to our method taking into account variable missingness , we used an upsampling technique called HAPGEN2 [29] , which limits the lower bound of the global allele frequency to 1/ ( 2 ⋅ 503 ) . Furthermore , the outcome used for the simulated GWAS is based on one causal variant with an explained variance of 0 . 02 , therefore it might not be fully representative for a polygenic phenotype with more than one causal variant . The summary statistics imputation method itself has several limitations too . Due to the size of publicly available sequenced reference panels we can not explore the performance of rare variants ( MAF < 1% ) . The imputation of summary statistics of an untyped SNV is essentially the linear combination of the summary statistics of the tag SNVs ( Eq ( 2 ) ) . Such a model cannot capture non-linear dependence between tag- and target SNVs [10] , which is often the case for rare variants [44 , 45] . In contrast , genotype imputation is able to capture such non-linear relationships by estimating the underlying haplotypes ( a non-linear combination of tagging alleles ) . Furthermore , in case of genotype imputation it is sufficient that the relevant haplotypes are present in the reference panel , but the overall allele frequency does not need to match the GWAS allele frequency . Summary statistics imputation relies on fine tuning of parameters , such as shrinkage of the correlation matrix . Any λ > 0 will make the correlation matrix invertible , but a stronger shrinkage can compensate for estimation error . We hypothesised that optimal shrinkage depends on local LD structure , and sought to optimise λ for each genomic region using the effect sizes of tag SNVs as training data set in a leave-one-out fashion . When looking at null variants , however , maximum shrinkage ( λ = 1 ) usually leads to the smallest RMSE . Therefore , when looking at a region with a mixture of null and associated SNVs , the selected λ will be shifted towards 1 and shrink the estimation of associated SNVs towards 0 , which is not ideal . The imputation quality metric r ^ pred , adj 2 tends to be inaccurate in case of small reference panels . The metric is commonly estimated as the total explained variance of a linear model given the reference panel , where the unmeasured SNV is regressed onto all measured markers in the reference panel ( Eq ( 7 ) ) . We noticed that for reference panel sizes smaller than 1000 individuals , the conventional estimation of imputation quality in Eq ( 7 ) is biased towards overestimation . We extend the existing imputation quality ( Eq ( 7 ) ) by accounting for sample size and the effective number of variants ( Eq ( 8 ) ) . The most accurate imputation quality estimations are obtained using an out-of-sample prediction after model selection by fitting a ridge regression model for each unmeasured SNV ( r 2 ^ ridge ) . However , due to the computational complexity , the calculation takes longer than the actual imputation . We provide a more detailed analysis in S2 Appendix . | Genome-wide association studies ( GWASs ) quantify the effect of genetic variants and traits , such as height . Such estimates are called association summary statistics and are typically publicly shared through publication . Typically , GWASs are carried out by genotyping ∼ 500′000 SNVs for each individual which are then combined with sequenced reference panels to infer untyped SNVs in each’ individuals genome . This process of genotype imputation is resource intensive and can therefore be a limitation when combining many GWASs . An alternative approach is to bypass the use of individual data and directly impute summary statistics . In our work we compare the performance of summary statistics imputation to genotype imputation . We observe that genotype imputation shows a 3- to 5-fold lower RMSE compared to summary statistics imputation , as well as a better capability to distinguish true associations from null results . Furthermore , we demonstrate the potential of summary statistics imputation by presenting 34 novel height-associated loci , 19 of which were confirmed in UK Biobank . Our study demonstrates that given current reference panels , summary statistics imputation is a very efficient and cost-effective way to identify common or low-frequency trait-associated loci . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
]
| [
"genome-wide",
"association",
"studies",
"variant",
"genotypes",
"sociology",
"social",
"sciences",
"alleles",
"genetic",
"mapping",
"mathematics",
"statistics",
"(mathematics)",
"genome",
"analysis",
"molecular",
"biology",
"techniques",
"genotyping",
"research",
"and",
"analysis",
"methods",
"mathematical",
"and",
"statistical",
"techniques",
"statistical",
"methods",
"consortia",
"molecular",
"biology",
"genetic",
"loci",
"heredity",
"meta-analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"genomics",
"statistics",
"computational",
"biology",
"human",
"genetics"
]
| 2018 | Evaluation and application of summary statistic imputation to discover new height-associated loci |
The structural flexibility or ‘breathing’ of the envelope ( E ) protein of flaviviruses allows virions to sample an ensemble of conformations at equilibrium . The molecular basis and functional consequences of virus conformational dynamics are poorly understood . Here , we identified a single mutation at residue 198 ( T198F ) of the West Nile virus ( WNV ) E protein domain I-II hinge that regulates virus breathing . The T198F mutation resulted in a ~70-fold increase in sensitivity to neutralization by a monoclonal antibody targeting a cryptic epitope in the fusion loop . Increased exposure of this otherwise poorly accessible fusion loop epitope was accompanied by reduced virus stability in solution at physiological temperatures . Introduction of a mutation at the analogous residue of dengue virus ( DENV ) , but not Zika virus ( ZIKV ) , E protein also increased accessibility of the cryptic fusion loop epitope and decreased virus stability in solution , suggesting that this residue modulates the structural ensembles sampled by distinct flaviviruses at equilibrium in a context dependent manner . Although the T198F mutation did not substantially impair WNV growth kinetics in vitro , studies in mice revealed attenuation of WNV T198F infection . Overall , our study provides insight into the molecular basis and the in vitro and in vivo consequences of flavivirus breathing .
Flaviviruses are enveloped , positive-stranded RNA viruses typically transmitted to humans via infected ticks or mosquitoes . As many members of the flavivirus genus are emerging , they constitute a significant threat to global health . For example , approximately 390 million humans worldwide are infected annually with one of the four serotypes of dengue virus ( DENV ) [1] . West Nile virus ( WNV ) was introduced into North America in 1999 [2] and rapidly became the leading cause of arbovirus-related encephalitis in the United States [3]; Zika virus ( ZIKV ) emerged from Asia and Africa for the first time in 2007 and has since caused epidemics in French Polynesia [4] , Oceania [5] , and most recently , the Americas [6 , 7] . Despite causing significant morbidity , licensed vaccines or therapeutic agents to protect humans against many flaviviruses are lacking . However , highly effective vaccines for some flaviviruses such as yellow fever virus , Japanese encephalitis virus , and tick-borne encephalitis virus are in use . The induction of a neutralizing antibody ( NAb ) response is a correlate of protection for these vaccines [8–12] . While a live-attenuated tetravalent DENV vaccine was recently licensed , its efficacy and durability varied by DENV serotype , pre-existing flavivirus immune status , and age of vaccine recipient [13–15]; the relationship between neutralization titer and protection for this vaccine is less clear . Because of the importance of antibodies for flavivirus immunity , a detailed understanding of flavivirus antigenic structure as well as the mechanisms of antibody-mediated neutralization is critical [16] . Assembled flavivirus particles are composed of three structural proteins: capsid ( C ) , pre-membrane ( prM ) , and envelope ( E ) . The E protein , which consists of three structural domains ( DI , DII , and DIII ) connected to the viral membrane via a helical anchor , has critical roles in directing both the assembly of virions and their entry into cells . Flexible hinges between E protein domains enable conformational changes necessary for many steps of the viral life cycle , including fusion and maturation [17 , 18] . Flaviviruses bud into the lumen of the endoplasmic reticulum as immature , non-infectious particles with a spiky surface composed of 60 icosahedrally arranged prM-E heterotrimers [19 , 20] . During virus egress through the acidic environment of the trans-Golgi network , conformational changes in E expose a cleavage site within prM , which is recognized by host furin-like proteases . Cleavage of prM in the trans-Golgi network and release of the pr peptide in the extracellular environment give rise to mature and infectious virus particles covered by antiparallel E homodimers . NAbs can target epitopes in all three E protein structural domains and in quaternary structures composed of multiple domains within or across E dimers [21–31] , and may block multiple steps in the virus entry pathway [32–35] . Cryo-electron microscopic ( cryo-EM ) reconstructions of DENV [36 , 37] , ZIKV [38 , 39] , and WNV [40] have provided detailed models for the structure of mature virions on which 90 E dimers lie flat against the viral membrane in a herringbone pattern . However , several lines of evidence suggest that many infectious flaviviruses exist as structures beyond those captured by high-resolution cryo-EM reconstructions [16] . Flaviviruses are released from infected cells as a heterogeneous population containing varying amounts of uncleaved prM due to incomplete maturation [41] . While the structures of these partially mature virions are not fully defined , they appear to contain regions that display mature- and immature-like arrangements of E proteins in varying proportions [42] . The presence of uncleaved prM on virions modulates the accessibility of many E protein epitopes recognized by NAbs [43–47] . Additionally , prM retention on virions allows for recognition by prM-reactive antibodies incapable of efficiently neutralizing infectivity . These prM-specific antibodies can enhance DENV infection and potentially contribute to severe clinical disease [48–52] . Flavivirus heterogeneity also arises from conformational flexibility of viral proteins that allows virions to sample an ensemble of conformations at equilibrium [53] . As with changes in virus maturation efficiency , virus conformational dynamics or ‘breathing’ has the potential to modulate antibody recognition and potency . Prolonged virus-antibody incubation reveals time-and temperature-dependent changes in antibody potency , the degree of which correlates generally with predictions of epitope accessibility on the mature virion; the neutralization potency of antibodies targeting cryptic epitopes can be enhanced more significantly compared to antibodies targeting highly accessible epitopes [54] . In this context , prolonged incubation of WNV and DENV in the absence of antibody does not irreversibly render them more sensitive to neutralization , suggesting that antibody binding may stabilize transiently sampled virion conformations [47] . The role of virus breathing in modulating antibody recognition is exemplified by the DENV1-specific human monoclonal antibody ( mAb ) E111 , which neutralizes two DENV1 strains , Western Pacific ( WP ) and 16007 , with a ~200-fold difference in potency [55] . Structural analyses revealed that E111 binds to an epitope in DIII that is predicted to be inaccessible based on existing cryo-EM models . The strain-dependent neutralization potency of E111 was neither explained by antibody binding affinity nor sequence variation within the epitope . Instead , a single residue in DII outside of the antibody footprint that differed between DENV1 WP and 16007 was responsible for modulating sensitivity to neutralization by E111 [56] . Thus , natural variation at this residue regulates the conformational dynamics of DENV1 in a way that affects exposure of the distal E111 epitope . The determinants and functional consequences of a dynamic virion are poorly understood . In this study , we describe a single residue within the E protein DI-DII hinge that alters the neutralization sensitivity and stability of WNV and DENV virions through changes in conformational dynamics . Mutation at this residue in the WNV E protein attenuated infection and pathogenesis in mice , suggesting that changes in virus breathing have relevant consequences in vivo .
To identify epitopes targeted by NAbs in flavivirus-immune sera , we created a library of WNV reporter virus particles ( RVPs ) [57] containing mutations at solvent-accessible E protein residues for use in neutralization studies . We focused on residues within the DI-DII hinge region , which has been shown to be an important target for many potently neutralizing antibodies against both WNV and DENV [23 , 24 , 33] . Because virion maturation state and conformational dynamics may modulate epitope accessibility and neutralization sensitivity indirectly , we performed a series of control experiments , as described previously [26] , to identify pleiotropic effects of mutations on the overall antigenic structure of virions . These experiments identified a mutation at E residue 198 ( T198F , Fig 1A ) that unexpectedly modulated sensitivity to neutralization by the DII-fusion loop ( DII-FL ) -reactive mAb , E60 , despite being located distally from the predicted epitope [58] . RVPs incorporating T198F were 68-fold more sensitive to E60 neutralization than were wild type ( WT ) RVPs ( average EC50 = 13 . 1 and 881 ng/ml for T198F and WT RVPs , respectively , p< 0 . 0001 , Fig 1B and 1D ) . In contrast , there was little difference in sensitivity to neutralization by mAb E16 ( average EC50 = 5 . 9 and 8 . 1 ng/ml for T198F and WT RVPs , respectively , Fig 1C and 1E ) , which binds an accessible epitope on the DIII lateral ridge ( Fig 1A ) [59–62] . Neutralization studies with antibodies that bind poorly accessible epitopes such as those in DII-FL [46 , 58 , 63] often reveal incomplete neutralization even at saturating antibody concentrations [45] . As flavivirus neutralization occurs when the number of antibodies bound to the virion exceeds a stoichiometric threshold [62] , incomplete neutralization may reflect structural heterogeneity among a genetically homogeneous RVP population that limits epitope accessibility . Virions that are not bound by antibodies at a stoichiometry sufficient for neutralization remain infectious [62] . We noted that a significantly smaller fraction of T198F RVPs remained infectious at the highest E60 concentration tested ( 10 μg/ml ) compared to WT ( 5 . 5% and 26 . 7% for T198F and WT RVPs , respectively , p = 0 . 001 , Fig 1B ) . These results demonstrate that the T198F mutation increases the accessibility of a cryptic epitope in DII-FL . The neutralizing activities of many mAbs , including those targeting DII-FL , can be modulated by the efficiency of virion maturation [45 , 46 , 64] . Specifically , virions that retain uncleaved prM often are more sensitive to neutralization by antibodies that bind epitopes predicted to be poorly accessible on the mature virion . For example , increasing the efficiency of prM cleavage ( Fig 2A ) reduced the sensitivity of WT WNV RVPs to neutralization by E60 relative to a standard WT RVP preparation ( ~70-fold reduction in EC50 , p = 0 . 02 , Fig 2B and 2C ) , as detailed previously [26 , 45 , 47] . To investigate whether T198F modulates the efficiency of prM cleavage during virion maturation , we analyzed the prM content of RVPs produced using standard conditions ( Std-RVP ) and in the presence of over-expressed human furin ( Furin-RVP ) . Although three independent preparations of Std T198F RVPs contained an average of five times the level of uncleaved prM compared to that of WT RVPs prepared in parallel , furin over-expression in RVP producing cells resulted in efficient prM cleavage for both WT and T198F RVPs ( Fig 2A ) , suggesting that the increased prM content of Std T198F RVPs was not due to an inability of the virion to adopt conformations in which the prM cleavage site was accessible during egress . Moreover , increased prM content was not sufficient to explain the greater sensitivity of T198F RVPs to neutralization by E60 , as Furin T198F RVPs were more sensitive to neutralization by E60 than were Furin WT RVPs ( 148-fold reduction in EC50 , p = 0 . 002 , Fig 2B and 2C ) , despite undetectable levels of uncleaved prM ( Fig 2A ) . Consistent with prior studies [26 , 45 , 47] , increasing the efficiency of prM cleavage increased the proportion of WT RVPs that were resistant to E60 neutralization at the highest mAb concentration tested ( 22% and 55% for Std and Furin WT RVPs , respectively , p = 0 . 03 , Fig 2B and 2D ) . In contrast , we observed only a minimal difference between the proportion of Std and Furin T198F RVPs resistant to neutralization at saturating concentrations of E60 ( 5% and 11% , respectively , p = 0 . 12 , Fig 2B and 2D ) . Additionally , in contrast to the 70-fold difference in EC50 between Std and Furin WT RVPs , there was a much smaller ( 7-fold ) difference in EC50 between corresponding preparations of T198F RVPs ( p = 0 . 09 , Fig 2B and 2C ) . Altogether , these data suggest that the large increase in sensitivity of T198F RVPs to neutralization by E60 was not simply due to increased retention of uncleaved prM . Furthermore , the reduced impact of maturation state on E60 recognition of WNV T198F suggests that the E60 DII-FL epitope is distinctly displayed by this variant . We investigated whether the increased neutralization sensitivity of WNV T198F was dependent on amino acid chemistry at this position . In addition to T198F , we produced Std WNV RVPs in which the threonine at position 198 was replaced with amino acids containing small ( A ) , nucleophilic ( C , S ) , hydrophobic ( L , M ) , acidic ( D ) , basic ( K ) , or amide ( N ) side chains . Each of these variants resulted in infectious RVPs , with titers within 2-fold of the WT control produced in parallel ( p = 0 . 47 , S1 Fig ) . Compared to WT , all T198 RVP variants were similarly sensitive to neutralization by mAb E16 ( <2-fold difference in EC50 , Fig 3A and 3C ) . However , when tested against E60 , neutralization sensitivity varied among RVP variants , although there was no clear correlation with amino acid chemistry . For example , T198F , T198M , T198K resulted in a similar ( ~50-fold ) reduction in EC50 despite incorporating amino acids with distinct side chain characteristics ( Fig 3B and 3D ) . Importantly , every T198 variant except for T198A and T198S resulted in a significant increase in sensitivity to neutralization by E60 to varying extents ( ~10 to 100-fold reduction in EC50 compared to WT , Fig 3B and 3D ) . Thus , the increased sensitivity of T198F RVPs to neutralization by E60 was not linked to a particular change in amino acid chemistry at WNV E residue 198 . We hypothesized that the increased accessibility of the E60 DII-FL epitope on T198F virions was due to changes in virus conformational dynamics or ‘breathing , ’ which allows the transient display of poorly exposed epitopes [47 , 54] . A potential consequence of virus breathing is a reduction in virus stability that can be inferred by the loss of infectivity over time ( or ‘intrinsic decay’ ) at physiological temperatures , as has been established for picornaviruses [65 , 66] . Among the ensemble of conformations sampled by a virion at equilibrium , a subset may result in irreversible structural changes that are incompatible with infectivity for a given cell type . We therefore investigated whether the T198F mutation altered the functional stability of WNV RVPs . Following incubation at 37°C for up to 72 hours , the infectivity of WT and T198F RVPs collected periodically was determined by titrating viruses on Raji-DCSIGN-R cells ( Fig 4A ) . In agreement with previous findings , virion maturation state modulated the rate of intrinsic decay [47 , 67]: the infectivity of Std RVPs decayed more rapidly compared to that of Furin RVPs for both WT and T198F . For Std RVPs , the T198F substitution resulted in a ~3-fold reduction in half-life relative to WT ( average half-life of 5 versus 16 hours , respectively , p<0 . 0001 , Fig 4B ) . Similarly , the infectivity of Furin T198F RVPs decayed at a rate that was approximately twice as fast as Furin WT RVPs ( average half-life of 10 versus 22 hours , respectively , p = 0 . 001 ) . These findings support the hypothesis that the T198F mutation alters the ensemble of conformations sampled by virions at equilibrium . The threonine at residue 198 in WNV and the phenylalanine found at the corresponding residue in DENV ( 193 ) and ZIKV ( 198 ) is highly conserved ( 99 . 9 to 100% ) among 1989 , 2692 , and 104 sequences of WNV , DENV , and ZIKV naturally occurring isolates , respectively , available in the Virus Variation database [68] . We introduced the reciprocal mutation at this residue ( F193T ) into the Western Pacific strain of DENV1 to investigate whether it similarly affected antigenic structure and dynamics of the virion . The infectivity of standard RVP preparations of DENV1 F193T was reduced by ~10-fold as compared to WT DENVI ( p = 0 . 001 , S1 Fig ) . While prM cleavage was much less efficient for DENV than WNV , as reported previously [69 , 70] ( Figs 2A and 5A ) , the F193T mutation in DENV1 did not alter maturation efficiency; three independent standard preparations of WT and F193T DENV1 RVPs contained a similar level of prM ( 1 . 1-fold difference , Fig 5A ) . We next investigated the effect of F193T on the neutralization of DENV1 by E60 [58] . Similar to results with WNV T198F , DENV1 F193T RVPs were more sensitive to neutralization than WT RVPs ( average EC50 = 25 ng/ml and 200 ng/ml for F193T and WT RVPs respectively , p = 0 . 0006 , Fig 5B and 5C ) . Moreover , the F193T mutation resulted in a 3-fold reduction in the half-life of infectivity of DENV1 RVPs ( average half-life of 2 . 5 and 0 . 8 hours for WT and F193T , respectively , p<0 . 0001 , Fig 5D and 5E ) . In contrast to our findings with WNV and DENV1 , mutation at the analogous residue ( F198 ) of ZIKV E protein had no effect on sensitivity to neutralization by E60 or virion stability in solution ( Fig 5F and 5G ) . Together , these results suggest that a single residue in the DI-DII hinge of the E protein alters the exposure of a cryptic DII-FL epitope and the stability of flavivirus particles in solution in a context-dependent manner . To explore further the possibility that the T198F mutation modulates WNV conformational dynamics , we performed kinetic neutralization assays , in which virus-antibody complexes were used to infect Raji-DC-SIGN-R cells immediately following a 1 h pre-incubation at room temperature , or further incubated at 37°C prior to addition of target cells ( Fig 6 ) . We studied only Furin RVP preparations containing little or no prM to eliminate the confounding effects of heterogeneity in virion maturation state on neutralization sensitivity ( Fig 2B ) [26 , 45 , 47] . As observed previously , increasing the virus-antibody incubation time for mAb E16 , which binds an accessible epitope on DIII , resulted in only modest increases in neutralization potency against both WT and T198F RVPs [54] . In contrast , kinetic changes in neutralizing antibody potency were more pronounced with mAb E60 . Increased sensitivity due to the T198F mutation was observed at all time points tested . For example , after 8 hours of incubation , we observed complete neutralization of Furin T198F RVPs , whereas a large proportion of Furin WT RVPs remained infectious at the highest E60 concentration tested . Even after 24 hours of incubation with E60 , WT RVPs remained much less neutralized compared to T198F RVPs . These findings suggest that the T198F mutation alters accessibility of the E60 DII-FL epitope . To extend our findings with RVPs capable of only a single round of infection , we investigated effect of the T198F mutation on standard preparations of fully infectious WNV encoding GFP [71] . As observed with RVPs ( Figs 1 and 2 ) , the T198F mutation reduced the efficiency of prM cleavage ( Fig 7A ) , and resulted in increased sensitivity to neutralization by E60 ( Fig 7B ) . Following prolonged incubation at 37°C , T198F reduced the infectious half-life of virions by ~2-fold relative to WT ( average half-life of 2 . 8 and 6 . 5 hours , respectively , p = 0 . 002 , Fig 7C ) , consistent with our observations with RVPs ( Fig 4 ) . As expected , raising the temperature at which viruses were incubated to 40°C reduced the infectious half-life of both WT and T198F RVPs ( average half-life of 3 . 6 and 1 . 6 hours , respectively , Fig 7C ) compared to incubation at 37°C; the 2-fold decrease in the half-life of T198F relative to WT viruses was maintained at 40°C ( p<0 . 0001 ) . Despite reduced stability in solution , T198F viruses nonetheless displayed similar growth kinetics as WT in Vero cells at 37°C and 40°C , and in mosquito ( C6/36 ) cells at 28°C ( Fig 7D ) , suggesting that the effect of T198F on virus stability in solution might be masked under conditions that allow efficient cell-cell spread of infection . Because the T198F mutation did not impair WNV replication in vitro , we investigated its impact on pathogenesis in a well-established mouse model of infection [72] . We infected 5-week old C57BL/6J mice with either WT or T198F WNV and monitored survival for three weeks . In contrast to the high mortality rate ( 13 of 15 ) observed among mice infected with WT WNV , only 2 of 15 T198F-infected mice succumbed to infection ( p<0 . 0001 , Fig 8A ) . As expected , WT- and T198F-infected mice that died experienced rapid weight loss beginning at day 6 post-infection ( Fig 8B ) . Weight loss was observed to a much lesser extent among mice that survived infection . Type I interferon ( IFN ) and antibody responses have been shown to be critical for protection against lethal WNV infection [72–75] . We therefore investigated the outcome of T198F infection of mice treated with MAR1-5A3 , a previously described blocking antibody against IFN-α/β receptor that prevents type I IFN-induced intracellular signaling in vitro and inhibits antiviral responses in mice [76 , 77] , and of congenic C57BL/6J mice that were genetically deficient in mature B cells and antibody ( μMT strain ) . The attenuated phenotype of T198F in WT mice was not observed in MAR1-5A3-treated WT or μMT mice . Although T198F infection remained attenuated relative to WT infection ( 1/10 versus 7/10 deaths , respectively , p = 0 . 004 , Fig 8C ) in mice treated with GIR-208 , an isotype control antibody targeting human IFNγ receptor , MAR1-5A3-treated mice were equally susceptible to lethal infection with WT or T198F virus ( 9/10 vs 8/10 deaths , respectively , p = 0 . 36 , Fig 8C ) . Analogously , 8/8 μMT mice infected with either WT or T198F virus succumbed to lethal infection by day 12 and 13 post-infection , respectively ( p = 0 . 08 , Fig 8D ) . Thus , the T198F mutation attenuates WNV in mice in a manner that is dependent on type I IFN signaling or B cell responses . To investigate whether the T198F virus was attenuated due to rapid clearance , we collected serum samples at days 2 and 4 following infection of WT mice with either WT or T198F WNV . At day 2 post-infection , serum viral load was 7-fold lower in T198F- compared to WT-infected mice ( p = 0 . 009 , Fig 8E ) ; a similar reduction in serum T198F infectious titer was observed at this time point ( p = 0 . 06 , Fig 8F ) . However , by day 4 post-infection , T198F serum viremia reached WT levels ( p = 0 . 82 , Fig 8E ) . Moreover , there was no difference in the infectious titer of WT and T198F viruses harvested from spleens at 4 days post-infection ( p = 0 . 79 , Fig 8F ) . Despite similar levels of WT and T198F viremia by day 4 post-infection , viral burden in the brain of T198F-infected mice was severely reduced compared to WT-infected mice . By day 6 , WT-infected mice had a median virus titer of 104 PFU/g in the brain , whereas no infectious virus was detectable in the brain of T198F-infected mice ( p = 0 . 001 , Fig 8G ) . Infectious virus became detectable in the brain of 2 of 8 T198F-infected in mice by day 8 , although at levels that were over 10-fold lower than those found in WT-infected mice on the same day ( p = 0 . 02 , Fig 8G ) . Sequence analyses of viruses isolated from the brain of T198F-infected mice revealed no reversion to WT . These results suggest that WNV containing the T198F mutation is suppressed early in infection and is attenuated for neuroinvasion or neurovirulence . As shown above , viremia in T198F-infected mice was reduced as early as day 2 , but not by day 4 post-infection ( Fig 8E and 8F ) . Because WNV-specific antibodies become detectable 4 days after infection [72 , 73] , we hypothesized that natural antibodies might accelerate the rate of T198F virus decay relative to WT . To test this hypothesis , we compared the stability of WT and T198F viruses incubated at 37°C in heat-inactivated serum samples obtained from naïve WT or μMT mice; viruses incubated in media were included as controls . We observed a ~2 . 3- and ~1 . 5-fold reduction in half-life of WT and T198F viruses in these serum samples , respectively , compared to incubation in media ( Fig 9 ) , suggesting that antibodies and other heat-resistant serum factors modulate the infectious half-life of WNV . Notably , the 2-fold reduction in half-life of T198F viruses compared to WT was observed following incubation in WT serum , μMT serum , and media ( Fig 9D ) , suggesting that natural antibodies do not differentially modulate the rate of decay of WT and T198F WNV . To investigate whether neutralizing antibodies play a role in the attenuation of WNV T198F , we pooled serum samples from five WT- or T198F-infected WT mice for use in neutralization studies with WT and T198F RVPs . We compared serum samples obtained at days 6 and 9 after infection to distinguish neutralizing activity mediated by IgM and IgG , which become detectable at 4 and 8 days after infection , respectively [72 , 73] . At both days 6 and 9 , there were minimal differences ( maximum of 1 . 2 fold change in EC50 ) in the ability of sera from WT- and T198F-infected mice to neutralize WT or T198F RVPs ( Fig 10A–10D ) . To distinguish IgM- versus IgG-mediated neutralizing activity , pooled serum samples from infected mice were either treated with 2-mercaptoethanol ( 2-ME ) , which preferentially degrades IgM [72 , 78] , or were used for IgG purification . As expected , treatment with 2-ME resulted in a large reduction ( 74–196 fold , Fig 10A and 10B ) in the neutralization potency of serum samples obtained from WT- and T198F-infected mice at day 6 , during which IgM , but not IgG is present [72 , 73] . At day 9 , when IgG is present , 2-ME treatment resulted in a smaller reduction in serum neutralization potency ( 4–8 fold , Fig 10C and 10D ) . As observed with untreated sera , at both days 6 and 9 , there were minimal differences ( maximum of 1 . 8 fold change in EC50 ) in the ability of WT- or T198F-immune sera treated with 2-ME to neutralize WT and T198F RVPs . Although T198F RVPs were slightly more sensitive ( 2–4 fold ) to neutralization by IgG purified from day 9 sera compared to WT RVPs , this was observed for both WT- and T198F-immune IgG ( Fig 10E and 10F ) , suggesting that infection with WNV T198F did not uniquely elicit NAbs that preferentially neutralized T198F viruses . Finally , to directly study the impact of the T198F mutation on immunogenicity , we immunized WT C57BL/6J mice with WT or T198F RVPs capable of only a single round of infection . Pooled sera from WT or T198F RVP-immunized mice at either day 10 or 21 displayed limited differences in their ability to neutralize WT and T198F RVPs ( S2C–S2F Fig ) . Similar results were observed with individual serum samples at day 21 post-immunization; although T198F was neutralized with a 3–4 fold greater potency compared to WT RVPs , this was observed for sera obtained from both WT and T198F RVP immunization groups ( S2G Fig ) . These results suggest that neither infection nor immunization with T198F elicited unique NAb responses in mouse polyclonal sera .
Our study demonstrates that a single residue in the E protein DI-DII hinge regulates conformational dynamics in distinct flaviviruses , with relevant consequences in vivo for WNV infection and pathogenesis . Although conformational flexibility has been described for different virus families [53] , the first evidence of the dynamic properties of flaviviruses came from structural studies of mAb 1A1D-2 , which is capable of neutralizing multiple DENV serotypes , yet binds to an epitope in the β-strand of DIII that is not predicted to be fully accessible on the mature virus particle [79] . Monoclonal antibody 1A1D-2 can bind to DENV particles at 37°C , but not at 4°C , suggesting that at an elevated temperature , this mAb trapped a conformation on which the DIII β-strand epitope was otherwise not accessible at lower temperatures [79] . Consistent with the conformational flexibility of flaviviruses , subsequent studies showed that exposure of DENV2 virions to physiological temperatures in the absence of antibody results in the formation of an expanded ‘bumpy’ structure , on which E protein dimers are more loosely arranged and are rotated outwards relative to their orientation on fully mature particles [80 , 81] . However , this ‘bumpy’ structure was not observed for all DENV strains or serotypes [81 , 82] , suggesting that sequence variation contributes to the structural pathways sampled by virus breathing at equilibrium . The molecular determinants that govern flavivirus breathing have not been defined . We recently demonstrated that natural variation at residue 204 in DII of the DENV1 E protein explained large genotypic differences in sensitivity to neutralization by a mAb targeting a cryptic epitope in DIII [56] . In our current study , we identified a single mutation in the E protein DI-DII hinge of WNV ( T198F ) and DENV1 ( F193T ) that increased sensitivity to neutralization by mAb E60 , which targets a poorly accessible epitope that includes the DII-FL . Because flavivirus neutralization occurs once the number of antibodies bound to the virion exceeds a stoichiometric threshold , neutralization potency depends not only on antibody affinity , but also on epitope accessibility [62] . Based on this model , antibodies targeting poorly exposed epitopes may not achieve complete neutralization even at saturating concentrations [45 , 62] , as observed for E60 against WT WNV . We observed that the T198F mutation markedly reduced the proportion of neutralization resistant WNV virions at high E60 concentrations . Compared to WT , increased accessibility of this epitope on T198F virions was less dependent on maturation state , which has been shown to indirectly modulate epitope accessibility [45 , 46 , 64] . Following prolonged incubation with E60 for up to 24 hours , WT WNV virions remained much less sensitive to neutralization than T198F virions . These results demonstrate that the E60 DII-FL epitope is displayed uniquely on T198F virus particles . Notably , increased sensitivity to neutralization by E60 also was observed by introducing a mutation at the corresponding residue of DENV1 , but not ZIKV , suggesting that the molecular mechanisms governing conformational flexibility and/or FL exposure may be distinct for ZIKV , in agreement with recent neutralization studies with mAbs [31 , 83–85] . The structural basis for these functional data cannot be inferred directly from our studies . The phenylalanine at position 193 and 198 of DENV and ZIKV , respectively play a space filling role ( S4 Fig ) , while for WNV , the analogous threonine at this position projects outwards and is solvent exposed . Thus , the local structural environment likely contributes to the effects of amino acid substitutions at this position . Moreover , while the structures of mature ZIKV and DENV particles share many similarities , a distinguishing feature is an extended loop surrounding the glycan at ZIKV E residue 154 , which has been hypothesized to limit accessibility of the adjacent DII-FL on the neighboring E protein [38] . Time-dependent increases in E60 neutralization potency were still apparent for T198F virions , suggesting that while this mutation increased the overall accessibility of a cryptic DII-FL epitope , it did not result in a grossly open ‘ground-state’ conformation . In support of this hypothesis , neutralization studies with an expanded panel of mAbs revealed that T198F ( and DENV1 F193T ) did not uniformly confer large increases in the potency of antibodies targeting distinct epitopes throughout the E protein ( S3 Fig ) . Indeed , T198F resulted in a relatively modest increase in the neutralization sensitivity of WNV to mAb E53 , which targets residues within the nearby DII bc-loop in addition to those within DII-FL [46 , 58] . Together , these results suggest that T198F alters the ensemble of conformations sampled by WNV to increase the accessibility of poorly accessible epitopes within DII-FL . Consistent with alterations in conformational dynamics , T198F also impacted the functional stability of WNV virions . Although the molecular basis for the loss of virus infectivity ( intrinsic decay ) following prolonged incubation is not understood [47 , 65 , 66] , we hypothesize this could be a consequence of virus breathing . Among the ensemble of conformations sampled by a dynamic virus , a subset may lead to irreversible changes in the E protein that impair viral infectivity . The more rapid intrinsic decay of T198F virions suggests that this mutation alters the conformational landscape in a manner that more frequently leads to irreversible changes in the E protein that are incompatible with infectivity . Although studies of antibody reactivity and intrinsic decay have provided clues into the dynamic properties of flaviviruses , the consequences of virus breathing for viral replication and pathogenesis remain poorly understood . Our finding that the T198F mutation in WNV reduced the efficiency of prM cleavage from virions prepared under standard conditions suggests that virus breathing may affect the accessibility of the prM cleavage site during Golgi transit , thus contributing to the heterogeneity in the maturation state of released virus particles [41] . While the corresponding mutation in DENV1 ( F193T ) increased both sensitivity to E60 neutralization and the rate of intrinsic decay , prM cleavage efficiency in the context of DENV1 was unaffected . We previously demonstrated that the rate of intrinsic decay differs between WNV and DENV [47 , 86] , suggesting that sequence variation and the presence of uncleaved prM may alter the structural pathways sampled by flaviviruses . The possibility that virus breathing affects prM cleavage efficiency , perhaps by modulating access to the furin cleavage site on prM , further adds to the complex interplay among the determinants of flavivirus structural heterogeneity . Indeed , the reduced efficiency of prM cleavage of both WT and F193T DENV1 compared to WT WNV may reflect differences in the structural flexibility of DENV and WNV . We demonstrated that altered virus breathing impacts pathogenesis . The T198F mutation attenuated WNV pathogenesis in WT mice , but not in mice treated with a monoclonal antibody targeting the IFN-α/β receptor or in congenic C57BL/6J mice deficient in B cells and antibody , suggesting that T198F attenuation is dependent on type I IFN- or B cell-mediated immunity . Our finding highlights the role of both innate and adaptive immune responses in protection against WNV lethal infection . Specifically , type I IFN signaling has been shown to be important in priming and enhancing B cell responses , in addition to its established role in innate antiviral defense [87–89] . Prior studies have demonstrated that both neutralizing and non-neutralizing WNV-specific antibodies can protect against lethal infection [72–74 , 90] . For weakly neutralizing antibodies targeting DII-FL , protection is dependent on non-neutralizing mechanisms [90] . Our data indicate that T198F attenuation is not likely due to increased susceptibility to NAbs , suggesting a possible role for antibody effector functions . T198F viremia was reduced as early as day 2 post-infection , before WNV-specific antibodies become detectable [72 , 73] , suggesting that this early viral suppression also might be due to differential effects of innate immune responses . Additionally , we previously found that the presence of even low concentrations of WNV-specific antibody can decrease the infectious half-life of virions in vitro [54] , perhaps by trapping conformations that are incompatible with infectivity . Although we demonstrated that natural antibodies did not differentially affect the rate of intrinsic decay of WT and T198F viruses in vitro , it is possible that in the presence of other immune factors , even low concentrations of WNV-specific and/or natural antibodies may facilitate rapid viral clearance in vivo to limit dissemination to vital organs , as has been shown for other viruses [91] . Indeed , the suppression of T198F viremia very early in infection , though transient , was sufficient to limit CNS dissemination , which typically occurs between days 4 and 5 after infection [72 , 74] . Finally , the stoichiometric requirements of prM cleavage for the production of infectious virus particles have not been defined . Although increased prM retention on T198F virus particles did not significantly impair infectivity in vitro ( Fig 7 ) , it is possible that decreased maturation efficiency of T198F may result in lower infectivity of key target cells in vivo , thus contributing to its attenuation . Beyond impacting pathogenicity , the in vivo consequences of virus conformational dynamics are unexplored . The vector competence and extrinsic incubation period ( time from infected blood meal to transmission ) for both WNV [92–95] and DENV [96–99] are temperature-dependent , which could correspond to changes in the extent of virus breathing . It is intriguing to consider that the reduced infectious half-life of T198F WNV and F193T DENV1 virions in solution may result in less efficient transmission to mosquitoes during an infected blood meal , especially from a febrile animal , given that the rate of intrinsic decay is accelerated at elevated temperatures ( Fig 7C and [86] ) . Indeed , sequence analyses reveal that WNV E residue 198 and the analogous DENV E residue 193 are highly conserved in nature [68] . The impact of changes in conformational dynamics on virus attachment to target cells also is unexplored . In addition to increased susceptibility to immune clearance , changes in conformational flexibility also might impair T198F virus interaction with host attachment factors in the central nervous system or in the vessels lining the blood-brain barrier [100] . T198F was neutralized slightly more potently than WT RVPs by both WT- and T198F-immune mouse sera ( Fig 10 and S2 Fig ) , suggesting that infection or vaccination with T198F did not skew the NAb response to preferentially neutralize T198F . Thus , although the T198F mutation impacts antigenicity as measured by changes in accessibility of a cryptic DII-FL epitope , its effects on immunogenicity are unclear . These results suggest , however , that antibodies targeting DII-FL do not significantly contribute to the overall neutralizing activity of polyclonal sera in mice . As the specificity of the polyclonal antibody repertoire elicited by flavivirus infection likely differs between mice and humans [22 , 101–103] , how changes in E protein conformational flexibility alter immunogenicity in humans remains to be determined . Recently , a number of potently neutralizing human monoclonal antibodies that target quaternary epitopes within or across flavivirus E protein dimers have been identified following natural infection or vaccination [23 , 24 , 27 , 30 , 31 , 33 , 34] . We speculate that the dynamic properties of E proteins have the potential to impact the exposure of these epitopes , and thus the induction of antibodies against them . The conformational flexibility of envelope proteins has been shown to modulate antibody recognition of HIV ( [104]; among the different structures sampled by HIV envelope trimers at equilibrium , broad and potent NAbs preferentially target the highly ordered , ‘closed’ trimer conformation [104–106] . Based on these observations , current HIV immunogen design strategies to elicit broad and potent NAbs are focused on stably presenting the closed form of native envelope trimers [106–108] . Whether limiting conformational flexibility is a suitable strategy for flavivirus vaccine design awaits further studies . Our ongoing studies aim to identify additional residues throughout the E protein that regulate conformational flexibility to facilitate studies on the impact of flavivirus breathing on immunogenicity and other aspects of flavivirus biology , including maturation , replication , and the pH threshold of fusion [109 , 110] . We hypothesize that residues at the E protein hinge regions and dimer interface play critical roles in regulating virus breathing by virtue of their conformational flexibility [17 , 18] and potential interactions that contribute to overall virion stability [111] , respectively . The existence of antiviral compounds that inhibit virus breathing of selected picornaviruses suggests an important role for structural flexibility in the lifecycle of viruses [112–117] . Structural flexibility contributes to heterogeneity in the antigenic structure of virions by governing the exposure of cryptic epitopes that may be immunodominant [47 , 54] . For example , antibodies that bind the conserved DII-FL are cross-reactive , poorly neutralizing antibodies with the potential to contribute to antibody dependent enhancement at high concentrations , which is especially relevant in the context of DENV infection [52 , 118] . We have shown here for WNV and elsewhere for DENV [56] that exposure of cryptic epitopes can be modulated by amino acid substitutions at a distance . Thus , an improved understanding of the molecular determinants that regulate flavivirus breathing and the consequences of conformational dynamics on flavivirus biology has the potential to inform both the design of novel vaccines and identification of antiviral compounds .
HEK-293T ( ATCC ) and Vero ( ATCC ) cells were maintained in Dulbecco’s Modified Eagle medium ( DMEM ) containing 25 mM HEPES ( Invitrogen ) supplemented with 7% fetal bovine serum ( FBS; Invitrogen ) and 100 U/ml penicillin-streptomycin ( P/S; Invitrogen ) . C6/36 ( ATCC ) cells were similarly cultured , except with the addition of 1X non-essential amino acids ( Invitrogen ) . Raji-DC-SIGN-R cells ( Raji B lymphoblast [ATCC] engineered to stably express DC-SIGN-R , Pierson lab [45 , 54 , 62 , 119] ) were cultured in RPMI 1640 medium containing Glutamax ( Invitrogen ) supplemented with 7% FBS and 100 U/ml P/S . HEK-293T , Vero , and Raji-DC-SIGN-R cells were maintained at 37°C in the presence of 7% CO2 . C6/36 cells were maintained at 28°C in the presence of 7% CO2 . We used a previously described expression vector encoding the structural genes ( C-prM-E ) of the WNV NY99 strain [57] as a template for mutagenesis . Initially , threonine at residue 198 of the WNV E protein was replaced with phenylalanine by site-directed mutagenesis using the Pfu Ultra DNA polymerase system ( Agilent Technologies ) . The reciprocal mutation ( F193T ) was introduced into a previously described expression vector encoding the structural genes ( C-prM-E ) of the DENV1 Western Pacific strain [86] . Mutation at the analogous residue of the ZIKV E protein ( F198T ) was introduced into a plasmid encoding the structural genes of the ZIKV strain H/PF/2013 [120] . This plasmid is described elsewhere [121] . Additional amino acid variants were introduced at position 198 of the WNV E protein using primers containing a degenerate codon ( NNN ) . PCR cycling parameters were as follows: 1 cycle of 95°C for 1 min; 18 cycles of 95°C for 50 s , 60°C for 50 s , and 68°C for 9 min; and 1 cycle of 68°C for 7 min . PCR products were treated with DpnI ( New England BioLabs ) for 3 h at 37°C , prior to transformation into Stbl2 cells ( Invitrogen ) and propagation at 30°C . The entire C-prM-E region of each construct was sequenced to ensure that no additional mutations were present . RVPs were produced by complementation of a GFP-expressing WNV sub-genomic replicon with plasmids encoding the structural genes of WNV , DENV , or ZIKV , as described previously [121 , 122] , with slight modifications . Briefly , HEK-293T cells were pre-plated in a low-glucose ( 1 g/liter ) formulation of DMEM containing 25 mM HEPES ( Invitrogen ) , 7% FBS , and 100 U/ml P/S , transfected with plasmids encoding the replicon and structural genes at a 1:3 ratio by mass using Lipofectamine 3000 ( Invitrogen ) , and incubated at 37°C . For each microgram of DNA , 2 μl of Lipofectamine 3000 was used . Four hours post-transfection , cells were transferred to 30°C . Supernatant was harvested at 72–96 h post-transfection , passed through a 0 . 22 μm filter ( Millipore ) , and stored at –80°C . To produce mature preparations of WNV RVPs containing low to undetectable prM , RVPs were produced as above by co-transfecting plasmids encoding the replicon , structural genes , and human furin at a 1:3:1 ratio . To detect prM in RVP preparations , a modified structural gene construct that encodes prM and E , with a V5 tag immediately downstream of the prM signal cleavage site [123] was used to complement a plasmid encoding capsid [57] . For immunization studies , 180 ml of transfection supernatant containing WT or T198F RVPs was passed through a 0 . 22 μm filter , layered on 20% sucrose ( pH 7 . 4 ) , and pelleted by ultracentrifugation at 32 , 000 rpm at 4°C for 5 h . The virus pellet was resuspended in 0 . 5 ml of PBS containing 1% BSA . Infectious WNV encoding a GFP reporter gene was produced using a previously described molecular clone system in which a DNA fragment encoding WNV structural genes is ligated into a GFP-expressing WNV replicon plasmid ( pWNV-GFP-backbone V3 ) and transfected directly into HEK-293T cells [71] . Briefly , 1 μg each of the backbone and structural gene plasmids was digested with BamHI and BssHII , and ligated with T4 DNA ligase ( New England Biolabs ) in a final volume of 40 μl at 16°C overnight . Next , the entire unpurified ligation mixture was transfected directly into HEK-293T cells using Lipofectamine 3000 ( Invitrogen ) . Cells were incubated at 37°C in the presence of 7% CO2 . Viral supernatant was harvested at 48 and 72 h post-transfection , filtered using a 0 . 22 μm filter ( Millipore ) , and stored at –80°C . To detect prM in fully infectious virus preparations , a DNA fragment encoding WNV structural genes was modified to express a V5 tag immediately downstream of the prM signal cleavage site and was used for virus production as described above . Clarified virus-containing supernatant was serially diluted 2-fold in a total volume of 100 μl and used to infect 5 x 104 Raji-DC-SIGN-R cells in an equal volume at 37°C . Cells were fixed in 1 . 8% paraformaldehyde at 48 h or 16 h following infection by RVPs or fully infectious viruses , respectively , and GFP-positive cells enumerated using flow cytometry . Virus titer was calculated using the linear portion of the virus-dose infectivity curve using the following formula: Infectious units ( IU ) /sample volume = ( % GFP-positive cells ) x ( number of cells ) x ( dilution factor ) . Viruses produced in HEK-293T cells using WNV-GFP-backbone V3 [71] as described above were used to inoculate Vero or C6/36 cells at a multiplicity of infection ( MOI ) of 0 . 05 for 2 h at the indicated temperatures , after which supernatant was collected to confirm the input virus titer . After washing twice with PBS to remove unbound virus , cells were further incubated at 37°C ( Vero ) , 40°C ( Vero ) , or 28°C ( C6/36 ) . At the indicated time points , virus supernatant was collected and clarified by centrifugation at 2000 rpm for 5 min . Virus titers were determined on Raji-DC-SIGN-R as described above . RVP or fully infectious virus stocks were diluted to a level of infectivity that ensures antibody excess ( ~5 to 10% ) and incubated with serial dilutions of mAbs or heat-inactivated ( 56°C for 1 h ) sera for 1 h at room temperature before addition of Raji-DC-SIGN-R cells . To investigate the kinetics of neutralization , virus-antibody complexes were further incubated for additional lengths of time at 37°C as indicated prior to addition of Raji-DC-SIGN-R cells . All infections were performed in duplicate at 37°C . At 48 h ( RVP ) or 16 h ( fully infectious virus ) post-infection , infectivity was scored as a percentage of GFP-positive cells by flow cytometry . Antibody dose-response curves were analyzed using non-linear regression with a variable slope ( GraphPad Prism v 6 . 0g , GraphPad Software Inc . ) to calculate the concentration of antibody ( EC50 ) required to inhibit infection by 50% , or the maximum inhibition of infectivity achieved at the highest antibody concentration tested ( ‘% Resistant’ ) . Serum samples were depleted of IgM by treatment with 0 . 1 M of 2-mercaptoethanol in 1X PBS for 1 h at 37°C , as described previously [72 , 78] . Total IgG was purified from 50 μl sera pooled from WT-immune ( n = 5 ) or T198F-immune ( n = 5 ) five-week old WT C57BL/6J mice at day 9 post-infection using the Melon IgG purification kit ( Thermo Scientific ) in a final volume of 500 μl ( 1:10 dilution ) . Purified total IgG was quantified using a human IgG ELISA kit ( Immunology Consultants Laboratory ) for use in neutralization assays as described above . Viruses were diluted to a similar level of infectivity as used in neutralization assays , allowed to equilibrate at the indicated temperature for 1 h ( reference ) and sampled periodically for the next 48–72 h . At each time point , aliquots were collected and stored at –80°C . All frozen samples were thawed simultaneously and used to infect Raji-DC-SIGN-R in triplicate to assess infectivity as described above . Infection was normalized to the level observed at the initial reference time point and fitted with a one-phase exponential decay curve ( GraphPad Prism v 6 . 0g , GraphPad Software Inc . ) to estimate the infectious half-life . The level of prM in RVP preparations was determined by SDS-PAGE and Western blot analysis , as previously described [64 , 123] . Briefly , RVPs were concentrated and partially purified by ultracentrifugation at 4°C ( 32 , 000 rpm for 5 h ) through a 20% sucrose cushion , followed by re-suspension in TNE buffer ( 50 mM Tris , 140 mM NaCl , 5 mM EDTA , pH adjusted to 7 . 4 ) containing 1% Triton-X100 . WNV and DENV1 E proteins were detected by a cross-reactive DII-FL reactive mouse monoclonal antibody , 4G2 ( 1 μg/ml ) . WNV prM-V5 was detected using a 1:5000 dilution of a mouse monoclonal antibody targeting V5 ( Invitrogen ) , while DENV1 prM was detected using mouse monoclonal antibody , prM22 ( 0 . 5 μg/ml ) [124] . IRDye 800CW goat-anti mouse IgG ( LI-COR Biosciences ) diluted 1:2500 was used as a secondary antibody . Protein bands were visualized and quantified using the Odyssey infrared imaging system ( LI-COR Biosciences ) . C57BL/6J mice were purchased from Jackson Laboratories ( Bar Harbor , ME ) and congenic μMT B cell-deficient were bred at Washington University under pathogen-free conditions . Five-week old WT C57BL/6J mice or eight-week old μMT mice were inoculated subcutaneously via footpad injection with 102 focus-forming units ( FFU ) of WNV NY99 WT or T198F , and monitored daily for survival . Where indicated , C57BL/6J mice were injected via an intraperitoneal route with 0 . 5 mg of a mouse monoclonal antibody targeting mouse IFN-α/β receptor ( MAR1-5A3 ) [76] or an isotype control mouse antibody targeting human IFN-γ receptor 1 ( GIR-208 ) one day prior to infection . Purified LPS-free monoclonal antibodies MAR1-5A3 and GIR-208 were purchased from Leinco Technologies . WT and T198F viral stocks were generated by in vitro transcription of an infectious two-plasmid cDNA clone as previously described [125] . The T198F mutation was introduced into plasmid pWN-AB , which consists of the 5'-UTR and structural genes , by site-directed mutagenesis as described above . For immunization studies , five-week old C57BL/6J mice were injected via an intraperitoneal route with 50 μl of WNV WT or T198F RVPs normalized by infectivity and relative E protein content as determined by antigen capture ELISA . Serum from immunized mice collected at days 10 and 21 were analyzed in neutralization studies . High-binding 96-well plates ( Corning ) were coated with 3 μg/ml humanized mAb E16 in 100 μl coating buffer ( 100 mM BupH Carbonate Bicarbonate Buffer , Fisher ) with pH adjusted to 9 . 6 . Plates were washed six times with PBS containing 0 . 05% Tween 20 followed by incubation with 100 μl blocking buffer ( PBS , 3% non-fat dry milk , and 0 . 05% Tween 20 ) . RVPs were serially diluted 2-fold starting at a 1:100 dilution in 100 μl blocking buffer , added to plates , and incubated for 1 h at 37°C . Plates were washed again and were incubated with 100 μl of mouse mAb E16 diluted in blocking buffer ( 2 μg/ml ) for 1 h at 37°C . Following washing , 100 μl of HRP-conjugated goat anti-mouse IgG ( Thermo Scientific ) diluted 1:10 , 000 in blocking buffer were added to plates and incubated for 1 h at 37°C . One-step Ultra TMB-ELISA ( Thermo Scientific ) substrate was added ( 100 μl/well ) and incubated for six minutes at room temperature in the dark . The reaction was stopped by the addition of 100 μl 1N hydrocholoric acid ( Fisher ) and read on a plate reader ( BioTek Synergy H1 ) at a wavelength of 450 nm . On the indicated day post-infection , mice were sacrificed and organs collected following extensive perfusion with PBS . Organs were weighed , homogenized using a bead-beater apparatus , and titrated by plaque assay on BHK-21 cells [126] . Viral burden in serum samples was measured by plaque assay on Vero cells , and viral RNA from serum was isolated using the Viral RNA Mini Kit ( Qiagen ) and measured by quantitative fluorogenic reverse-transcription PCR as described previously [126] . Statistical analyses were performed using GraphPad Prism v 6 . 0g ( GraphPad Software Inc . ) . For results of in vitro experiments , paired t-tests or a one-way ANOVA followed by Dunnett’s multiple comparisons test was used , for two or more comparisons , respectively . For survival analysis , Kaplan-Meier curves were plotted and analyzed by the log rank test . Mouse serum , spleen , and brain viral loads and titers were compared using the Mann-Whitney test . Experiments were approved and performed in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocols were approved by the Institutional Animal Care and Use Committee at the Washington University School of Medicine ( Assurance number A3381-01 ) . | Flaviviruses include emerging pathogens such as WNV , DENV , and ZIKV that threaten global health . Despite causing significant morbidity , effective vaccines or therapeutic agents to protect humans against many flaviviruses are lacking . Because of the importance of antibodies in flavivirus immunity and vaccine protection , much effort is focused on understanding the factors that modulate antibody recognition of flaviviruses . Virus breathing , which allows viruses to sample different conformations at equilibrium , has the potential to transiently expose otherwise inaccessible antibody epitopes . Here , we report the identification a single mutation in the envelope protein that alters the exposure of a poorly accessible epitope and the stability of both WNV and DENV through changes in the ensemble of structures sampled by the virus . For WNV , this change attenuated infection and pathogenesis in mice , suggesting that virus conformational dynamics have relevant consequences in vivo . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"radiochemistry",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"viral",
"structure",
"nuclear",
"decay",
"viruses",
"physiological",
"processes",
"rna",
"viruses",
"sexually",
"transmitted",
"diseases",
"antibodies",
"breathing",
"respiration",
"immune",
"system",
"proteins",
"monoclonal",
"antibodies",
"infectious",
"diseases",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"chemistry",
"virions",
"physics",
"biochemistry",
"west",
"nile",
"virus",
"flaviviruses",
"nuclear",
"physics",
"virology",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms"
]
| 2017 | A single mutation in the envelope protein modulates flavivirus antigenicity, stability, and pathogenesis |
c-Myc ( hereafter called Myc ) belongs to a family of transcription factors that regulates cell growth , cell proliferation , and differentiation . Myc initiates the transcription of a large cast of genes involved in cell growth by stimulating metabolism and protein synthesis . Some of these , like those involved in glycolysis , may be part of the Warburg effect , which is defined as increased glucose uptake and lactate production in the presence of adequate oxygen supply . In this study , we have taken a mouse-genetics approach to challenge the role of select Myc-regulated metabolic enzymes in tumorigenesis in vivo . By breeding λ-Myc transgenic mice , ApcMin mice , and p53 knockout mice with mouse models carrying inactivating alleles of Lactate dehydrogenase A ( Ldha ) , 3-Phosphoglycerate dehydrogenase ( Phgdh ) and Serine hydroxymethyltransferase 1 ( Shmt1 ) , we obtained offspring that were monitored for tumor development . Very surprisingly , we found that these genes are dispensable for tumorigenesis in these genetic settings . However , experiments in fibroblasts and colon carcinoma cells expressing oncogenic Ras show that these cells are sensitive to Ldha knockdown . Our genetic models reveal cell context dependency and a remarkable ability of tumor cells to adapt to alterations in critical metabolic pathways . Thus , to achieve clinical success , it will be of importance to correctly stratify patients and to find synthetic lethal combinations of inhibitors targeting metabolic enzymes .
Activation of one of the three MYC oncogenes is frequently selected for during tumorigenesis . These genes encode the transcription factors c-Myc , N-Myc and L-Myc that regulate a large number of downstream target genes . Although most of the work on MYC oncogenes has involved their role in cell proliferation , it is becoming clear that they may be involved in most aspects of oncogenic transformation [1] . As such , unravelling the mechanisms by which Myc proteins activate genes , and which are the essential genes , is paramount as studies resolving these mechanisms may open up new avenues of targeted intervention against various cancers . Some of Myc's earliest discovered transcriptional targets were genes encoding metabolic enzymes such as Ornithine decarboxylase [Odc] [2] , [3] , Lactate dehydrogenase A [Ldha] [4] and Carbamoyl-phosphate synthase/aspartate carbamoyltransferase/dihydroorotase [Cad] [5] . Later studies using expression profiling identified even more of these genes , indicating that Myc is a master regulator of cellular metabolism and cell growth [6] , [7] . Interestingly , inhibition of polyamine biosynthetic enzymes Odc and Spermidine synthase have shown efficacy in chemoprevention of several cancers in experimental models [8]–[14] and in colon cancer patients [15] . Furthermore , Myc-regulated Ldha , Pyruvate kinase M2 and Glutaminase have also emerged as promising targets based on experimental models of human cancer [16]–[23] , suggesting that targeting various metabolic pathways regulated by Myc may prove beneficial in cancer therapies of patients . To gain in vivo support for this notion we performed genetic ablation experiments in mice to determine the individual contribution to tumorigenesis of three different Myc-regulated metabolic enzymes .
To identify critical Myc-regulated metabolic enzymes , we performed Illumina bead chip arrays on RNA isolated from 4–6 week old wildtype or precancerous , B cell lymphoma-prone λ-Myc transgenic mice , where the human MYC gene is under the control of the immunoglobulin ( Ig ) λ enhancer [24] . Interestingly , when we performed unsupervised Hierarchical clustering on 153 genes ( Table S1 ) encoding metabolic enzymes involved in glycolysis , the Kreb's cycle , oxidative phosphorylation , serine synthesis and one-carbon metabolism , all expression profiles from Myc-transgenic B cells grouped together despite some intra-individual expression level differences that could be due to expression levels of MYC and developmental stage of the B-cell compartment ( Figure 1A ) . Largely , these data are supportive of the Myc target gene database ( http://www . myccancergene . org ) . Many tumor cells use aerobic glycolysis , producing lactate even in the presence of oxygen ( the Warburg effect ) [25] . Most of the glucose from the enhanced glucose uptake is however used to provide metabolites for anabolic processes , such as fat and nucleotide synthesis . Glycolysis and nucleotide metabolism are linked at several steps including the pentose phosphate shunt and via the phosphorylated pathway of serine synthesis ( Figure S1 ) . In the latter , 3-phosphoglycerate dehydrogenase ( Phgdh ) catalyzes the first step and serine hydroxymethyltransferases ( Shmt1 and Shmt2 ) use the final product to produce folate metabolites that are critical for several metabolic pathways including methylation and thymidylate synthesis . Given that Myc regulates genes involved in many metabolic pathways we decided to focus on genes that were induced by Myc and for which there were genetic tools accessible at the start of this project . qRT-PCR analysis confirmed that the selected genes had elevated expression in Myc-transgenic B cells ( Figure 1B ) . Shmt1 and Shmt2 protein levels and their combined activity are elevated in B cells from λ-Myc transgenic mice ( Figure S2A and S2B ) . So to analyze the role of Shmt1 in Myc-induced tumorigenesis , we obtained embryonic stem ( ES ) cells carrying a gene-trapped allele of Shmt1 , which were injected into blastocysts to generate chimeric mice . The offspring from these mice generated wildtype , heterozygous or homozygous Shmt1 knockout mice at the expected Mendelian frequency ( Figure S2C and S2D ) , corroborating a recent publication reporting that Shmt1 is dispensable for mouse development [26] . The homozygous Shmt1 mutant mice did not express any Shmt1 protein in the tissues analyzed ( Figure S2E ) making them suitable for the assessment of the role of this gene for Myc-induced tumorigenesis . To that end , we first back-crossed the Shmt1 knockout mouse for 10 generations to the C57BL/6 strain and then the interbred it with 3 different tumor models where Myc is either the direct cause of transformation ( λ-Myc transgenic mice , Figure 2A [24] ) or constitutes an important circuit ( p53 knockout mice , Figure 2B [27] and ApcMin mice of intestinal tumorigenesis , Figure 2C [28] ) . Surprisingly , we did not observe any major negative impact of Shmt1 loss on tumor initiation and development in these models . The only significant effect was a clear acceleration of disease in the λ-Myc transgenic mice ( Figure 2A ) , suggesting a B-cell specific event . Taken together these data argue against Shmt1 as a target for chemotherapy . Serine and folate metabolites can also be made via pathways involving Shmt2 and Phgdh . However , to assess the role of Shmt2 , we were forced to take a different approach , as Shmt2 gene-trap clones or knockout mice were not available when initiating this project . We hence infected Colon 26 cells , which carry an NMU-induced Kras mutation [29] , with lentiviruses expressing shRNA directed against Shmt2 and Phgdh . Despite achieving potent knockdown levels , we did not observe any effect on viability or ability to form subcutaneous tumors when injected into syngenic Balb/c mice , as compared to cells infected with a control lentivirus ( Figure S3A and S3B ) . To further assess the effect of Phgdh loss in different tumor models , we obtained a Phgdh knockout mouse . Since Phgdh is essential for neurogenesis [30] , Phgdh null embryos die at around embryonic day ( E ) 13 . 5 , which prevented us from analyzing the effects of loss of Phgdh by conventional breeding to our tumor models . We therefore started out by assessing the impact of removal of just one allele of Phgdh on tumorigenesis in λ-Myc transgenic mice and in ApcMin mice . At variance with Odc , which is haploinsufficient for tumor progression in these models , Phgdh heterozygosity did not impact tumorigenesis in these tumor models ( Figure 3A and 3B ) , despite the 50% reduction in Phgdh activity ( Figure 3C ) . As an alternative approach , we crossed λ-Myc; Phgdh+/− mice with Phgdh+/− mice and isolated hematopoietic stem cells from E13 . 5 fetal livers . These cells were then used to reconstitute lethally irradiated syngenic recipients , creating lymphoma-prone mice with varying expression of Phgdh ( Figure 3D ) . Even in this setting , Phgdh was dispensable for Myc-induced tumorigenesis ( Figure 3E ) , suggesting that hematopoiesis and Myc-driven tumorigenesis can occur in the absence of Phgdh . Phgdh is linked to glycolysis and could potentially divert metabolites away from pyruvate usage in the TCA cycle in the mitochondrion . Pyruvate is also kept from entering the TCA cycle via the action of Ldha , encoded by another Myc-regulated gene [23] . Except for RNAi or antisense studies in established tumors , it is not known whether Ldha is needed for the actual transformation event in vivo . To assess this we used a mouse model carrying a procarbazine-induced homozygous germline mutation of Ldha [31] . The mutation has been mapped to an aspartate 223 to histidine exchange which results in a very strong phenotype in erythrocytes causing anemia that is counteracted by extra-medullary hematopoiesis with an associated splenomegaly ( Figure S4A ) . We crossed Ldha mutant mice with λ-Myc mice to generate mice of all relevant genotypes . Some of these mice were sacrificed before they developed tumors to allow analysis of Ldh activity in splenic B cells . Other mice were aged and monitored for tumor development . As seen in Figure S4B , splenic B cells from λ-Myc mice exhibited an elevated level of Ldh activity - consistent with the expression analysis in Figure 1 – whereas the Ldha mutation severely diminished Ldh activity in B cells from both non-transgenic and Myc transgenic mice ( Figure S4B ) . Unexpectedly , the Ldha mutation did not affect Myc-induced B-cell lymphomagenesis . In two independent survival curves generated at Umeå University and Helmholtz Center Munich the median survival time for λ-Myc;Ldhamut/mut was similar to that of λ-Myc;Ldhawt , with no statistical difference ( Figure 4A for the Umeå-generated survival curve; Munich curve is shown in Figure S4C ) . Our unexpected results suggest several possibilities: either Ldha is dispensable for Myc-induced lymphomagenesis; or a compensatory mechanism is occuring; or Ldha deficiency alters the route of transformation . Firstly , a compensation by another Ldh form is improbable since Ldh activity and Ldhb expression were very low or absent in tumors arising in λ-Myc;Ldhamut/mut mice ( Figure S4D and S4E ) . Secondly , Myc-induced lymphomagenesis in the mouse is known to involve spontaneously arising , cooperating oncogenic mutations of tumor suppressors and other oncogenes to block the oncogenic stress response of Myc [32] , [33] . Hence , overexpression of anti-apoptotic proteins such as Bcl-2 or genetic deletion of one tumor suppressor allele such as Arf or p53 dramatically accelerates lymphomagenesis [34]–[36] . To neutralize the genetic heterogeneity in the cooperating oncogenic lesion during lymphomagenesis we interbred the p53 knockout mouse with the Ldha mutant mouse and the λ-Myc mouse . λ-Myc;wildtype , λ-Myc;Ldhamut/wt or λ-Myc;Ldhamut/mut were made heterozygous for p53 by interbreeding . All mice developed disease at an accelerated rate as compared to λ-Myc mice ( Figure 4B ) . The tumors that developed lost the wildtype p53 allele ( data not shown ) . Moreover , the frequency of p53 mutation in the tumors that developed in the first cross ( Figure 4A ) was not different between Ldha genotypes ( data not shown ) . We conclude that Myc-induced lymphomagenesis can occur normally in mice lacking fully functional Ldha . Studies using antisense or RNAi have shown that Ldha is important for breast carcinoma , neuroblastoma , fumarase-deficient renal cell cancers , as well as fibroblast and B-cell tumor cells in vitro [17] , [18] , [20] , [23] , [37] . Although the specific combination of Myc overexpression with loss of p53 was previously unexplored , other explanations to the differences between our findings and those of others , like experimental methods , culture conditions , oxygen supply or oncogenic pathway , could be at play . To test if a dependency of Ldha could be revealed in settings where Myc is downstream rather than the primary oncogenic instigator , we interbred ApcMin mice and p53 knockout mice with the Ldha mutant mouse . As seen in Figure 4C , Ldha deficiency did not impact adenomagenesis in the ApcMin mice . However , we also created p53−/−;Ldhamut/mut or Ldhawt mouse embryo fibroblasts ( MEFs ) which were transduced with an oncogenic H-Ras ( pBabe-HrasG12V-puro ) retrovirus . Interestingly , whereas no growth defect could be observed in vitro , the Ldha mutant Ras-transformed p53 knockout MEFs generated significantly smaller tumors in vivo that appeared less vascularized ( Figure 4D and Figure S5A ) . This effect was not a result of varying expression of oncogenic H-Ras in the different tumors ( Figure S5B ) . Moreover , Colon 26 cells with an endogenous oncogenic Kras allele could not be propagated when transduced with lentiviruses expressing two different Ldha shRNAs despite the fact that these constructs were not lethal in NIH 3T3 cells ( data not shown ) . To confirm the dependency on Ldha we also co-transfected a GFP expressing plasmid with the lentiviral expression constructs expressing Ldha shRNA into Colon 26 . As seen in Figure S3C and S3D , we were able to knockdown expression of Ldha in the cells , which resulted in a progressive loss of cells from 48 h post-transfection . To investigate the ability of Ldha-deficient fibroblasts to proliferate in a hypoxic environment , NIH 3T3 cells infected with Myc ( pWLBlast-c-Myc ) or Ras ( pBabe-HrasG12V-hygro ) retroviruses and control or Ldha shRNA lentiviruses were exposed to hypoxia . As expected , hypoxia resulted in the induction of Ldha and Pdk1 , both downstream targets of Hif1α ( Figure S5C ) . Interestingly , cells infected with the Ldha shRNA incorporated less 3H-thymidine than cells infected with a control lentivirus ( Figure S5D ) . Thus , Ldha is required under defined conditions such as hypoxia and/or in cells with a deregulated Ras pathway . Therefore an Ldha dependency may not be manifested in a Myc-induced lymphomagenesis setting . In agreement with this notion , the λ820 mouse B-cell lymphoma line established from λ-Myc mice [38] succumbed to apoptosis when exposed to hypoxia ( Figure S6A and S6B ) , regardless of whether or not Ldha was knocked down ( Figure S6C ) . In addition , Ldha knockdown did not impact lymphomagenesis in vivo ( Figure S6D ) , although knockdown still left a substantial amount of Ldha transcript and activity in this highly Ldha-expressing cell line ( Figure S6C and S6E ) . Nevertheless , given the sensitivity of λ-Myc lymphoma cells to hypoxia ( Figure S6A and S6B ) , it is unlikely that tumorigenesis in this model contains a hypoxic component and thereby dependency on elevated Ldha activity . Indeed , immunohistochemistry showed that Ldha wildtype or mutant lymphomas from λ-Myc mice exhibited a remarkable sparse expression of angiogenic markers CD34 and SMA ( Figure S7A ) . Despite this there were no signs of obvious necrotic areas , suggesting that nutrients and oxygen can diffuse in these non-solid tumors . The staining results were not due to non-functional antibodies as they readily detected the angiogenic markers in normal spleen and in lymphomas that had disseminated in spleens of λ-Myc mice ( Figure S7B ) . It therefore appears as if lymphomas arising in lymph nodes of λ-Myc mice are neither angiogenic , hypoxic or dependent on Ldha activity .
We are today beginning to appreciate the fact that oncogenes and tumor suppressor genes not only regulate cell proliferation , immortalization , apoptosis , metastasis and angiogenesis [39] but also cellular metabolism . The change in metabolism and the Warburg effect were for a long time believed to be self-evident and secondary to transformation . It is now known that the metabolic changes occur simultaneously and are governed by the same signal transduction pathways as those governing cell proliferation [40] . Since different tumor cells transform in response to variations of oncogenic mutations it is therefore likely that tumor cells can , or even have to , make different metabolic adaptations as well . Some of these adaptations may make the cells dependent on a certain enzyme , whereas others do not . Our study highlights the Ras oncogene as a potential pathway that requires Ldha , illustrated in Ras-transformed fibroblasts and Colon 26 cells , which carry an endogenous Kras mutation . Indeed , it has previously been shown that neu-transformed breast cancer cells are sensitive to Ldha inhibition by RNAi [37] . Since these cells have an activated Ras pathway [41] , this may explain why knockdown of Ldha sensitizes these cells . The potential explanation why Ras-induced fibrosarcomas are sensitive to the Ldha mutation in vivo is that they are not inherently angiogenic , making them sensitive to metabolic perturbation before angiogenesis has occurred . The fact that Myc can stimulate angiogenesis independently of hypoxia-inducible factors [42]–[44] may account for the lack of impact on lymphomagenesis upon mutation of e . g . Ldha . Indeed , we show here that lymphomas from λ-Myc mice are very sensitive to hypoxia , most likely since they are Myc-driven and therefore rely on the TCA cycle and oxidative phosphorylation [45] . Folate biosynthesis has been linked to cancer both from studies on dietary supplements and by the identification of polymorphisms in genes encoding enzymes in folate biosynthesis like SHMT1 , MTHFR and TS [46] . In addition , certain drugs like methotrexate target the folate biosynthetic pathway suggesting that this pathway is of critical importance for tumor cell survival . Interestingly , Shmt2 was first identified as a target in a screen for genes that can rescue the growth defect of Myc null rat fibroblasts [47] . In the same study Shmt1 was also shown to be a Myc transcriptional target gene but it was not further functionally characterized . Herein , we provide evidence that Shmt1 is dispensable for Myc-induced lymphomagenesis and that its deletion even accelerates tumorigenesis . The reason for this acceleration is unknown . It could involve effects on senescence or B-cell development as deletion of genes like Suv39h1 and E2f2 accelerates tumorigenesis by these mechanisms , respectively [48] , [49] . However , our data corroborate other very recent studies . Using an Shmt1 knockout mouse [50] , the Stover group showed that deletion of one allele of Shmt1 promotes adenomagenesis in the ApcMin mice when administered a special diet [26] . In our study the Shmt1 heterozygous ApcMin mice also have the largest mean amount of adenomas , albeit we did not investigate the impact of diet on this model . Interestingly , homozygous deletion of Shmt1 did not impact adenoma formation in ApcMin mice since there was a compensatory increase in thymidylate kinase ( TK1 ) expression [47]: a salvage pathway for thymidine synthesis . We observed a stronger effect on acceleration of Myc-induced lymphomagenesis in homozygous mutant Shmt1 mice , suggesting that the salvage pathway is not completely penetrant . Taken together , we would argue that Shmt1 is a tumor-suppressing modifier in the context of B-cell lymphomas and colorectal adenomas . One of the most important reasons for the systematic analysis of Myc target genes in tumorigenesis is the potential of identifying or validating future drug targets . Our lymphoma and adenoma data cast doubt on the utility of developing targeted interventions against Ldha , Phgdh and Shmt1 . On the other hand , in these models , tumors arise in mice carrying germline mutations of both the oncogenic lesion and the genes encoding the metabolic enzymes . It is thus plausible that adaptations have occurred during development that would not have occurred in cells acutely exposed to an inhibitor . Nevertheless , our data suggest that tumor cells eventually will develop resistance to putative treatments directed against metabolic enzymes since tumor growth undoubtedly can occur in the absence of Shmt1 , Phgdh or Ldha . Therefore , a correct stratification is needed to identify patients whose tumors would be sensitive to inhibitors that are under development , for instance against Ldh [6] , [18] . Such stratification can be performed based on which oncogenic driver mutation the tumor has acquired . As shown here , oncogenic Ras or pathways utilizing this circuit could be a potential parameter . Moreover , two independent studies published while revising this manuscript suggest that PHGDH amplification could be another oncogenic driver mutation in breast cancer and melanoma , which would sensitize cells to inhibition of Phgdh [51] , [52] . As shown here and in these two studies , Phgdh is important in some but not all contexts . To date , very few inhibitors against Ldh have been identified and those known are either poorly bioavailable and/or have other targets . For instance , sodium oxamate is used in high millimolar concentrations but inhibits aspartate aminotransferase at concentrations where lactate production is not even affected when tumor cell growth is [53] . Gossypol , a natural compound from cottonseed first identified as a male contraceptive , also inhibits anti-apoptotic proteins of the Bcl-2 family making interpretation of anti-cancer activity difficult [54] . Even if improved inhibitors are developed the issue whether or not Ldh is a good target is unresolved . Our genetic study shows that cells carrying a defective Ldha are capable of forming tumors , albeit hindered by hypoxia . We and others also show that ablation of Ldha by RNAi can be detrimental for the cell . It is thus possible that either cell context determines sensitivity , or that the ablation of Ldha protein ( RNAi ) is more severe than inhibition of its activity ( D223H mutation ) . This notion would lend support to the idea that Ldha may have other functions , potentially disconnected from its activity [55] . For instance , Ldha can be phosphorylated ( Tyr238 ) and localized to the nucleus [56] and has recently been shown to exist in transcription complexes in ES cells [57] . Future studies should address if glycolysis-independent functions of Ldha , as suggested in transcription [58] , [59] , are the most important functions in some tumors . If so , focus on the development of new therapies should aim at blocking all activities of Ldha .
All animal experiments were performed in accordance with the Regional Animal Ethic Committee Approval no . A6-08 or no . A18-08 . All transgenic mice in the study were on pure C57BL/6 background . The λ-Myc-mice and the Ldhmut/mut mice have been previously described [24] , [31] . The p53 knockout mice and the ApcMin mice were from Jackson Labs , the C57BL/6 and Balb/c mice used as recipients were from Taconics , and the Phgdh knockout mice were from RIKEN BRC , Japan . Shmt1 knockout mice were generated by blastocyst injection of gene-trapped ES cells ( clone AD0236 , Sanger Institute Gene-trap Resources ) at Umeå Transgene Core Facility . After confirmation of germline transmission , mice were backcrossed to C57BL/6 for at least ten generations . Illumina SNP genotyping confirmed that the mice were at least 96% C57BL/6 before starting interbreeding with λ-Myc transgenic mice , p53 knockout mice and ApcMin mice . All mice used in the study were monitored by group members and personnel at the animal facilities ( Umeå Transgene Core Facility or Helmholtz Centre , Munich ) . When showing signs of disease , λ-Myc mice were sacrificed and lymphomas were collected for analyses . Dates of sacrifice were entered into GraphPad Prizm software for the generation of survival curves . Lymphomas were either snap frozen for RNA and protein analyses , or formalin-fixed and embedded in paraffin . Paraffin blocks were sectioned and processed by standard immunohistochemistry methodology using antibodies directed against smooth musle actin ( SMA; Sigma ) or CD34 ( Abcam ) at the Histocenter core facility ( Göteborg , Sweden ) . The ApcMin mice used for adenoma formation studies were sacrificed and analyzed between 120–140 days of age , or when showing signs of disease . The small intestine and colon were dissected out , washed with phosphate-buffered saline and cut length-wise at which point adenomas were counted and tissues were harvested for analyses . Adenomas were counted using dissection microscope as well as by eye by two independent observers . The adenomas were scored irrespective of size and numbers per mouse were entered into GraphPad Prizm software for generation of graphs . The analysis of gene expression changes between magnetically sorted B cells from wildtype or λ-Myc transgenic mice was performed using the Illumina BeadChip system . For in vitro transcription amplification , 200 ng of RNA was used with the Illumina RNA Amplification Kit ( Ambion ) . Amplified RNA ( 1 . 5 µg ) was hybridized to the Sentrix MouseRef-8 Expression Beadchip ( Illumina ) . The primary data were collected from the BeadChips using the manufacturer's BeadArray Reader and analyzed using the supplied scanner software . Data normalization was performed by cubic spline normalization using Illumina's Beadstudio v3 software . Clustering and visualization of genes encoding metabolic enzymes was done using the Spotfire software . MEFs were generated by mechanical disruption and trypsin-digestion of E13 . 5 embryos from which the fetal liver and the head had been removed . The single-cell suspension was grown in DMEM supplemented with 10% fetal bovine serum ( FBS ) , 50 µM β-mercaptoethanol , 1× glutamine , pyruvate , non-essential amino acids and antibiotics ( Invitrogen ) . 293T cells and NIH 3T3 ( from ATCC ) were routinely maintained in DMEM supplemented with 10% FBS , glutamine and antibiotics . Colon 26 cells ( from Cell Line Services ) were cultured in RPMI supplemented with 10% FBS , glutamine and antibiotics . The λ820 cell line was established from λ-Myc transgenic mice and cultured as previously described [38] . Retroviruses and lentiviruses were produced by calcium phosphate-mediated transfection of 293T cells . For retroviruses the following plasmids from Addgene were used: pBABE-HrasG12V-puro , pBABE-HrasG12V-hygro , MSCV-Myc-IRES-GFP , pWZL-Myc-blasticidine , pBABE-puro , pBABE-hygro together with pCL-Eco ( encoding gag , pol and ecotropic envelope ) . Lentiviruses for RNAi were made using pLKO . 1 puro vectors expressing shRNAs ( Sigma Mission RNAi ) , together with packaging plasmids pCMV R8 . 2dvpr and pHCMV-EcoEnv ( both from Addgene ) . Two or three different shRNAs were used per gene ( Table S2 ) and they were compared to a control pLKO vector expressing a control shRNA with no known target in the mouse genome ( non-target vector from Sigma ) . Thirty-six hours post-transfection , the media was harvested four times during an additional 36 h . The virus was filtered and either frozen down in aliquots or applied on target cells in the presence of 4–8 µg/ml polybrene . Following antibiotics selection for 48 h ( or GFP analysis of FACS to confirm at least 90% positive cells ) cells were expanded and used for experiments . The shRNA-containing viruses were always introduced into NIH 3T3 cells after the transduction with control , Myc or Ras retroviruses . MEFs used for sarcoma formation studies were infected with retroviruses encoding oncogenic Hras , whereas Colon 26 cells were infected with lentiviruses expressing shRNA against Shmt2 , Phgdh and Ldha . For MEFs , 1×106 were injected subcutaneously into C57BL/6 recipients , whereas 5×105 Colon 26 cells were mixed with Matrigel ( 1∶1 ) and injected into Balb/c mice . When tumors appeared , the mice were sacrificed and tumors were weighed and material was harvested for analyses . For immunofluorescence , formalin-fixed tumors were embedded in paraffin and sectioned ( 8 µm ) onto glass slides . Following deparaffinization and rehydration , slides were either stained with H&E or subjected to Hoechst and antibody staining using Cy3-conjugated control or anti-smooth muscle actin antibody ( Sigma ) according to standard methodology . Following mounting the sections they were analyzed in a fluorescence microscope . Fetal livers of E13 . 5 embryos were dissected out of embryos from timed pregnancies between λ-Myc; Phgdh+/− males and Phgdh+/− females . Each individual liver was dissociated through a cell strainer and injected via the tail-vein into one lethally gamma-irradiated ( 9 . 25 Gy ) C57BL/6 recipient . Tissue from each embryo was taken for genotyping and the mice positive for the λ-Myc transgene were followed for lymphoma development and treated as previously described in Mouse colonies and Tumor monitoring and analyses . For protein expression analyses by Western blot , cells and tumors were lysed in an appropriate amount of lysis buffer on ice for 30 min . Following sonication , clearing by centrifugation and protein determination , an equal amount of protein per well was loaded on SDS-PAGE gels and separated by electrophoresis . The proteins were transferred to a nitrocellulose membrane , which was subsequently blocked with TBST containing 5% non-fat dried milk . The membranes were then blotted with primary and horseradish peroxidise-conjugated secondary antibodies dissolved in blocking solution . After washing with TBST , the bound proteins were visualized by enhanced chemoluminescence . The primary antibodies used were from BD Biosciences ( H-Ras ) , Cell signalling ( c-Myc and Pdk1 ) , Sigma-Aldrich ( Ldha , Ldhb , Shmt1 , Shmt2 and β-actin ) and Atlas Antibodies ( Shmt1 and Phgdh ) . RNA expression was measured by quantitative reverse transcriptase PCR ( qRT-PCR ) . Briefly , RNA was prepared using the NucleoSpin RNA II kit ( Macherey-Nagel ) . cDNA was prepared using the First strand synthesis kit ( Fermentas ) and the PCR was run using the KAPA mastermix ( Biotools ) on an iQ real-time PCR machine ( Bio-Rad ) . Primer sequences can be found in Table S3 . NIH 3T3 cells expressing either HrasV12 , c-Myc or both were used for parallel infections of lentiviruses encoding shRNAs against Ldha . The same amount of cells were seeded in 24-well format and subsequently infected . 72 hours post infection , each well was split into 3×96 well format in duplicate plates . One set of plates was placed in a hypoxic environment ( using the Modular incubator chamber , Billups-Rothenberg Inc . ) for 40 hours after which the hypoxic treatment was terminated and 3H-thymidine was added to all wells . After two hours , the plates were freeze-thawed and the cells harvested onto glass fibre filters . Microscint scintillation solution was administered to the dried filter , which were subsequently counted on a TopCount scintillation counter . λ820 cells were subjected to hypoxia in 24 well plates or 25 cm2 flasks ( 2×105 cells/ml ) and were harvested for cell counting and apoptosis analyses or RNA analyses , respectively . For apoptosis analyses , cells were stained with Vindelövs reagent ( 10 mM Tris , 10 mM NaCl , 75 µM propidium iodine , 0 . 1% Igepal , and 700 units/liter RNase adjusted to pH 8 . 0 ) and then analyzed with a FACScalibur flow cytometer ( BD Biosciences ) . Apoptosis was determined using DNA histograms and was based on the number of cells that carried less than diploid DNA content ( sub-G1 ) in a logarithmic FL2 channel . Total protein lysates were prepared as described above and the same amount of protein was assayed for LDH activity using the Cytotoxicity detection kit ( Roche Applied Science ) . The reactions were read using the Tecan Infinite200 plate reader at 492 nm . Lysates were also used to assess Shmt activity and Phgdh in accordance with published methods [60] , [61] . | Cancer occurs when cells change their behavior and start to divide in an uncontrolled manner . To achieve this altered behavior , cells need to change their metabolism to be able to grow even when nutrient and oxygen supplies are limiting . Therefore , targeting metabolic pathways could be used to treat patients suffering from cancer . Here we studied a gene called MYC , which can regulate many metabolic pathways . By using genetically modified mice we can show that tumors have a remarkable ability to change their metabolism , even if key enzymes are removed . Taken together , our data suggest that metabolic disturbance by drugs in the clinic may present a future challenge . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"medicine",
"cancers",
"and",
"neoplasms",
"basic",
"cancer",
"research",
"gastrointestinal",
"tumors",
"hematologic",
"cancers",
"and",
"related",
"disorders",
"animal",
"models",
"oncology",
"model",
"organisms",
"molecular",
"genetics",
"lymphomas",
"signaling",
"in",
"cellular",
"processes",
"gene",
"expression",
"biology",
"molecular",
"biology",
"signal",
"transduction",
"genetics",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics",
"signaling",
"cascades"
]
| 2012 | Mouse Genetics Suggests Cell-Context Dependency for Myc-Regulated Metabolic Enzymes during Tumorigenesis |
The presence of 5-methylcytidine ( m5C ) in tRNA and rRNA molecules of a wide variety of organisms was first observed more than 40 years ago . However , detection of this modification was limited to specific , abundant , RNA species , due to the usage of low-throughput methods . To obtain a high resolution , systematic , and comprehensive transcriptome-wide overview of m5C across the three domains of life , we used bisulfite treatment on total RNA from both gram positive ( B . subtilis ) and gram negative ( E . coli ) bacteria , an archaeon ( S . solfataricus ) and a eukaryote ( S . cerevisiae ) , followed by massively parallel sequencing . We were able to recover most previously documented m5C sites on rRNA in the four organisms , and identified several novel sites in yeast and archaeal rRNAs . Our analyses also allowed quantification of methylated m5C positions in 64 tRNAs in yeast and archaea , revealing stoichiometric differences between the methylation patterns of these organisms . Molecules of tRNAs in which m5C was absent were also discovered . Intriguingly , we detected m5C sites within archaeal mRNAs , and identified a consensus motif of AUCGANGU that directs methylation in S . solfataricus . Our results , which were validated using m5C-specific RNA immunoprecipitation , provide the first evidence for mRNA modifications in archaea , suggesting that this mode of post-transcriptional regulation extends beyond the eukaryotic domain .
5-methylcytidine ( m5C ) is a modification that occurs both on DNA and RNA . In eukaryotes , this DNA modification has been extensively studied over the past years , and was found to play a crucial role in genomic imprinting , X-chromosome inactivation , and suppression of repetitive elements [1] , [2] . Less is known about the distribution and role of m5C sites in RNA . In bacteria , m5C positions were described only in rRNA , whereas in archaea and eukaryotes m5C was mapped to both tRNA and rRNA . In tRNA molecules , m5C sites are typically present at the variable region and the anticodon loop , where they have been shown to stabilize the secondary structure of the tRNA and affect codon identification and tRNA aminoacylation [3]–[5] . For example , m5C at position 40 of the yeast tRNA-Phe enables conformational transition of the entire anticodon loop , inducing Mg2+ binding at a distant site and resulting in structural stabilization [6] . Absence of an m5C modification in the anticodon wobble-base ( position 34 ) of tRNA-Leu in yeast was associated with decreased functionality , observed when nonsense suppressor function was estimated in vivo [7] . Recently , it was shown that various modifications on tRNA , including m5C , are dynamically modulated during cellular responses to several stress conditions , suggesting that these modifications may play a role in cellular response to stress , potentially by mediating translation rates [8] , [9] . In rRNA m5C is thought to play a role in translational fidelity [10] . Traditional methods for studying m5C and other RNA modifications include HPLC and mass spectrometry , requiring isolation of the specific RNA molecule being studied to near purity [11] , [12] . As opposed to the highly abundant tRNAs and rRNAs , isolation of specific mRNA molecules to purity is experimentally challenging , and hence studies of m5C modifications on mRNA molecules were limited until recently . Indeed , early studies documenting the presence of m5C in mRNA have been controversial; some studies have failed to detect m5C in eukaryotic mRNA [13]–[16] , whereas others have identified this modification [17]–[19] . Recently , the presence of m5C in human HeLa mRNA was studied in a global manner , identifying thousands of potential m5C sites within human mRNA [20] . This study utilized RNA bisulfite treatment , which selectively converts C residues , but not m5C , into U residues [21] , followed by whole transcriptome sequencing . Such bisulfite treatment is commonly used on DNA sequences when DNA methylation states are studied [22] , [23] . However , its usage for RNA m5C methylation interrogation has been limited until recently . As a result , no study has addressed the presence of m5C sites , or any other RNA modification , in prokaryotes in a transcriptome-wide manner , and no modification to bacterial or archaeal mRNA has been described to date . Here , we set out to generate transcriptome wide maps of m5C RNA modification sites in representative model organisms from all three domains of life . We applied bisulfite treatment on total RNA from these organisms , and generated a computational pipeline that identifies methylated positions after accounting for various artifacts . This approach was highly sensitive , and allowed the identification and quantification of the vast majority of known m5C positions in tRNAs and rRNAs in these organisms , as well as the detection of novel positions in these structural RNAs . Anti-m5C RNA immunoprecipitation confirmed our findings . Intriguingly , we detected m5C sites in several mRNAs in archaea , and identified a sequence motif guiding these RNA methylations . This might be suggestive of an additional layer of regulatory complexity present – potentially having functional consequence – on prokaryotic mRNAs .
In E . coli we identified rRNA modifications at 3 distinct positions . The identified residues included position 967 in the 16S rRNA subunit , and position 1962 in the 23S rRNA subunit ( Figure 2A–B ) . Both these residues have been previously determined to contain m5C modifications [25]–[27] . In addition , position 1402 in the 16S rRNA of E . coli was identified as having a methylation level of 58% . This position is known to harbor a N4 , 2-O-dimethylcytidine modification which also is not amenable to bisulfite conversion [21] , emphasizing the possibility that a fraction of the sites we detected may reflect modifications other than m5C ( see Discussion ) . We identified two modified residues at positions 977 and 1411 of the 16S rRNA of B . subtilis ( Figure 2C ) . While neither one of these residues has been previously directly characterized as methylated in B . subtilis , they are positionally orthologous to the previously characterized positions 967 and 1402 in the E . coli 16S rRNA , respectively ( above ) . Not only were the positions conserved , but also the methylation stoichiometry appeared to be conserved between E . coli and B . subtilis , with methylation levels of 98 . 7% at position 977 compared to 97 . 5% in E . coli , and 63 . 7% at position 1411 compared to 58% in E . coli . In S . solfataricus we identified one m5C site at position 1369 of the 16S rRNA , and another at position 2643 of the 23S rRNA . An additional site with a methylation level of 43% was observed at position 2121 of the 23S rRNA ( Figure 2E–F ) . Position 1369 at the 16S was previously characterized as subjected to an unknown modification [28] , whereas no evidence existed for methylation of the 23S sites . Sanger sequencing confirmed that these sites did not undergo bisulfite conversion , providing further support for their methylation ( Figure 3 ) . The presence of these modifications is well in line with previous chromatography/mass-spectrometry based analysis , which predicted one m5C site in the 16S subunit of S . solfataricus , and 1–2 additional sites in the 23S subunit [29] . In yeast we identified two modified positions within the 26S rRNA subunit . Position 2278 , which was found to have a methylation level of 72% , was previously reported to contain a methylated site [30] . To our knowledge , we are the first to report the presence of a methylation site at position 2870 ( methylation level 98% , Figure 2D ) . Its presence and the nucleotide composition of this detected site ( m5C site preceded by a pyrimidine ) , is consistent with the prediction of an additional m5C site in the 26S unit of yeast rRNA , which was not mapped at the time [30] . Eukaryotic and archaeal tRNA molecules , but not bacterial tRNAs , are known to undergo m5C modifications at specific positions . Consistent with this finding , we identified 46 modified sites in 34 tRNAs in S . solfataricus , and 39 sites in 30 tRNAs in S . cerevisiae . To allow comparative analysis of methylation levels in different tRNAs , we generated a multiple alignment of the archaeal and yeast tRNAs , and color-coded methylated positions along them based on their methylation levels ( Figure 4 ) . Notably , in this analysis we examined significant ( p<0 . 01 ) methylated positions even when methylation levels were below 50% , to allow full comparative inspection of the data . Consistent with previous findings , the methylated positions in both archaea and yeast were clustered at positions 48–49 along the tRNA molecule ( numbers based on E . coli tRNA positions , as in [31] ) . In archaea most ( 30/34 , 88% ) of the tRNAs were methylated at position 48 , while about half of these ( 14/30 ) were also methylated at position 49 . The pattern of m5C modification in yeast was different: usually , either position 48 or position 49 was methylated , with only two tRNAs showing modifications at both positions . Two other tRNAs also contained modifications at position 50 in the variable region . The relatively high coverage of tens and sometimes hundreds of reads in each tRNA allowed us to quantitatively assess similarities and differences in methylation levels at specific positions ( presumably corresponding to the percent of tRNA transcripts modified at those positions ) . In general , levels of methylation in yeast were significantly higher than those in archaea , with mean values of 96% per position in S . cerevisiae as compared to mean values of 82% observed in S . solfataricus ( Mann-Whitney test , p = 3 . 8e-6 ) . This may imply that the typical single , strongly methylated site in yeast may compensate for the presence of two more weakly methylated sites in S . solfataricus . This interpretation , however , needs to be taken with caution , as other factors may account for the observed differences in methylation levels ( see Discussion ) . Surprisingly , a small fraction of the tRNAs were found to be devoid of m5C modifications in the variable region , despite the presence of a cytosine in one of the positions 48–50 . These include tRNA-AlaAGC , TGC and tRNA-ArgCCT in yeast and tRNA-GluTCC , CTC in S . solfataricus . This selective absence of the modification in a subset of tRNAs might provide a potential anchor towards elucidating the role of this modification on tRNAs . Alternatively , other types of RNA modifications , not detectible through bisulfite-conversion , may replace m5C in these tRNAs . Several additional methylated positions were observed in our data , including the previously identified methylation in the anticodon ( position 34 ) of yeast tRNA-Leu [32] , as well as methylations at position 27 of tRNA-Pro and position 56 of tRNA-Leu . Additional putative m5C sites were observed in S . solfataricus tRNA-TrpCCA and tRNA-LeuTAA , but given their relatively low methylation levels , these sites require further verification . Finally , we failed to detect the known methylation at position 40 of tRNA-Phe [6] due to lack of sequence coverage for this particular tRNA . We were next interested in searching for methylated sites within mRNA molecules . Since our initial experiment lacked sufficient sequence coverage to identify such sites within the vast majority of mRNAs due to the high coverage of rRNA and tRNAs , we performed a second bisulfite-seq experiment on the bacterial and the archaeal RNA using size-selected total RNA ( >200 bp ) to deplete tRNAs , combined with rRNA depletion protocols to enrich for mRNAs ( Methods ) . For S . cerevisiae we made use of a dataset of short sequences previously obtained using bisulfite-treatment of polyA-selected RNA [33] . We used the computational pipeline outlined above to analyze the libraries obtained from the mRNA enrichment protocols . Coverage along mRNA in this experiment was substantially improved , with sufficient coverage ( mean coverage ≥5 reads/nucleotide ) in 1 , 394 protein coding genes in B . subtilis , 1 , 110 genes in S . solfataricus , and 4 , 489 genes in yeast . In E . coli , due to very high levels of ribosomal RNA reads ( >99% ) , only 287 protein-coding genes had sufficient coverage ( Table 2 ) . Despite significant coverage of most yeast genes , we were only able to detect a single event of m5C on an mRNA in that organism , occurring on a protein of unknown function . No m5C events were found in the expressed bacterial mRNAs that were covered in our experiment , although the lack of coverage for most E . coli genes prevents us from drawing a strong conclusion as to the possibility of methylated sites in mRNAs for that organism . In S . solfataricus , the mRNAs of two protein coding genes contained a nucleotide methylated to a level of >50% , and 12 additional genes contained nucleotides with methylation levels of 20–50% ( Table 3 ) . Many of the genes harboring the m5C modification were enzymes involved in energy and lipid metabolism ( notably oxidoreductases and dehydrogenases ) , which might imply a possible role for this modification in regulating specific metabolic processes . Interestingly , all 14 m5C positions in S . solfataricus mRNAs were located within a strong sequence motif of AUCGANGU ( Figure 5C ) . This same motif is found at the m5C sites we recorded at positions 2121 and 2643 of the S . solfataricus 23S rRNA ( Figure 2 ) . These results suggest that the same machinery ( probably a methyltransferase ) is responsible for m5C modifications both on the 23S rRNA and on the mRNAs . Moreover , our results imply that this machinery recognizes a consensus sequence of AUCGANGU . The identification of a strong consensus motif reinforces the validity of our bisulfite-sequencing approach in identifying real modified sites on RNA bases , and supports the existence of a single predominant methylation machinery on mRNAs in S . solfataricus . To verify that the m5C modifications we detected reproducibly appear in S . solfataricus mRNAs we sequenced a second bisulfite-treated biological replicate sample of total RNA from S . solfataricus . Indeed , all m5C sites identified in the first experiment were also detected in the second experiment with similar methylation levels but with higher coverage ( Table 3 ) . To rule out the possibility that the consensus motif we identified is intrinsically resistant to bisulfite conversion , we mixed the second S . solfataricus RNA sample , prior to bisulfite conversion , with two synthesized 200 nt RNA fragments each harboring a common representation of the consensus motif ( either ATCGAGGT or ATCGAAGG; Methods ) . Despite very deep coverage obtained for the two artificial consensus-bearing RNAs ( 130 , 000 and 2 . 8 million reads , respectively ) , no evidence for m5C was observed within these RNAs , and the percentage of cytosines that were not converted into uracils within the two synthesized consensus motifs was 0 . 6% , similar to non-methylated residues ( Table 2 ) . These results show that the methylation sites we observed in S . solfataricus mRNAs do not represent a motif-dependent artifact . To determine if the presence of this consensus sequence is sufficient to drive methylation , we searched for the motif within all S . solfataricus protein coding genes . We identified 75 sites that were covered by at least 5 reads ( regardless of p-value score ) . Of these , 31 ( 41% ) were methylated at levels >10% , and 13 ( 17% ) were methylated at levels higher than 20% . These levels are far higher than expected by chance alone: examining all sites ( n = 22 , 665 ) covered by at least 5 reads , but significantly differing from the consensus sequence , only 5 . 9% had methylation levels of >10% , and 0 . 9% of the sites had methylation levels exceeding 20% . These results suggest that the presence of this motif is to a large extent sufficient for recognition and methylation by the putative methyltransferase . However , the varying levels of methylation suggest that additional factors other than the immediate sequence environment – such as RNA availability or more distant motifs – may also play a role in determining methylation levels . We next examined the relative position of the modified sites within the genes they reside . Methylated bases tended to be localized towards the beginning of the gene ( Fig . S1 , P = 0 . 07 ) . The apparently non-uniform distribution of modified sites may provide an angle for future elucidation of the function of this RNA modification . Several modifications other than m5C may result in an inability to convert C->U following bisulfite treatment , including 3-methylcytidine , N4-methylcytidine , N4 , 2′-O-dimethylcytidine and N4-acetylated variants [20] , [21] . To further examine the nature of the novel modifications we observed in S . solfataricus RNAs , we set out to conduct direct immunoprecipitation of modified RNA . For this , we used a monoclonal antibody that specifically binds 5-methylcytosine . This antibody is broadly used to specifically detect m5C modifications in DNA ( e . g . [34] ) , but since it was raised against 5-methylcytosine nucleotide conjugated to ovalbumin without the ribose or deoxyribose sugar , it is blind to the DNA/RNA context of the modification and hence binds the RNA form of m5C as well . As a control we used a second antibody that specifically binds 5-hydroxymethylcytosine ( hm5C ) but not m5C . The hm5C modification is similar to m5C but carries an additional hydroxyl group on top of the added methyl group . Total RNA of S . solfataricus was sheared into ∼100 nt-long fragments ( the “input” ) and immunoprecipitated using the anti-m5C or anti-hm5C antibodies . Libraries were prepared from immunoprecipitated as well as non-immunoprecipitated input control RNA fragments , and subjected to massively parallel sequencing on the Illumina platform . Mapping the resulting reads to the S . solfataricus genome resulted in a highly non-uniform coverage ( Fig . 6 ) . Peaks in the coverage were clearly observed in the rRNA at the exact locations identified by the bisulfite sequencing as m5C modified ( positions 1369 in the 16S and 2121 and 2643 in the 23S ) . These peaks represented enrichment of 18–30 fold over the rest of the rRNA molecule , but no enrichment was observed in these positions in the non-immunoprecipitated ( input ) sample or the sample immunoprecipitated by the hm5C antibody . Most ( 10/14 , 71% ) of the sites we identified in the mRNAs of S . solfataricus ( Table 3 ) were also found to be enriched in the immunoprecipitated library ( Fig . 6C–E , enrichment of 10–35 fold ) . These results provide strong independent verification that the sites we identified in the rRNAs and mRNAs of S . solfataricus indeed correspond to m5C modifications .
We have employed bisulfite treatment of RNA , combined with high-throughput sequencing , to generate a sensitive map of methyl-modified cytosines in four model organisms across the microbial tree of life . Our map not only recovers the vast majority of known sites within tRNAs and rRNAs in these organisms , but also provides a measurement of methylation efficiency for each position , and reveals novel sites within rRNAs and mRNAs . Several modifications other than m5C may result in a lack of C→U conversion following bisulfite treatment [20] , [21] . Indeed , and in agreement with previously published data [21] , position 1402 in E . coli identified in this study is known to harbor an N4 , 2′-O-dimethylcytidine modification , which is known to be resistant to bisulfite conversion . Therefore , the novel modified positions we report in this study might correspond to modifications other than m5C . To verify that the novel modifications in mRNAs we identified in S . solfataricus indeed represent m5C , we utilized m5C-specific antibodies that are usually used in pulldown experiments of modified DNA molecules that are an epigenetic marker for gene silencing in mammals . Since these antibodies also bind the RNA form of m5C , they formed an ideal tool for independent validation of our bisulfite-based findings . Each of the approaches has its strengths and weaknesses: while the bisulfite conversion provides a single-nucleotide resolution positioning of the modified site , it may also report other types of modifications . The antibody-based approach provides specificity to a single type of modification but with lower resolution of about 100 bp [35] . Therefore , the combination of bisulfite-conversion and RNA-immunoprecipitation provides synergistic results . In eukaryotes , the chemical modification of mRNAs is emerging as an important factor in the regulation of gene expression . These modifications have been shown to affect mRNA translation , splicing , stability and transport [20] , [35] , [36] . To our knowledge , our report presents the first evidence for mRNA modification in archaea , opening a window into a possible additional layer of gene regulation in archaea . Further studies are needed in order to elucidate the possible function of these modifications . In particular , profiling of the methylation status across different cellular states , or in response to external stimuli , will allow investigation of the extent to which this modification is dynamically regulated . Knockout studies can provide better understanding of the enzymes involved in mediating this modification across different organisms and will be crucial for understanding its function . Our identification of a defined sequence motif associated with all mRNA methylated positions in S . solfataricus strongly suggests that these modifications indeed occur in vivo and are not an artifact of the detection method . This motif , along with the recently reported sequence motif associated with m6A modifications on mammalian mRNAs [35] , are the only reported cases of clearly defined linear sequence motifs directing multiple RNA methylation events . Indeed , studies on various tRNA and rRNA modifying enzymes had suggested preference for local RNA structures rather than for a specific sequence [25] , [37] , [38] . In this respect , it is possible that there is a fundamental difference between tRNA and rRNA molecules , in which modifications are often sequentially added according to folding and maturation levels , therefore demanding structural dependency , and mRNAs , in which the linear sequence may be the most important determinant . The motif we identified in S . solfataricus , AUCGANGU , is also found at the modified positions C2121 and C2643 in the 23S rRNA . Therefore , it is likely that the same methyltransferase responsible for modification of these rRNA residues also methylates the positions we detected in mRNAs . It is possible that the modifications we find on mRNAs might reflect a “spillover” of this enzymes' activity onto non-specific mRNA substrates . However , no methylated site on S . solfataricus mRNAs resembles the modified sequence of the 16S rRNA in that organism , suggesting a specific activity for the 23S methylase , but not for the 16S methylase , on mRNA positions . Interestingly , recent work by Squires and coworkers similarly found that m5C modifications on human mRNAs are mediated by the RNA methyltransferase NSUN2 , previously known to act only on human tRNAs [39] . Our approach not only detects methylated positions in RNAs , but also provides a quantitative measure of the fraction of transcripts harboring this methylation ( to which we refer as “methylation level” ) . The close similarities in rRNA methylation levels of orthologous positions between the gram positive bacterium B . subtilis and the gram negative E . coli reinforces the validity of this quantitative measure . Nevertheless , this measure should be taken with caution , as factors such as miniscule DNA contamination ( which is non-methylated ) , for example , could contribute noise to this measurement . In addition , bulky modifications ( which are prevalent in tRNAs ) may hinder reverse transcriptase processivity , and hence , completely modified tRNA molecules may be under-represented in the sequenced data . Therefore , the stoichiometric measurements we recorded might be biased . Methodologies that are not based on reverse transcription should be applied to decisively determine methylation stoichiometry . An additional limitation of our approach is that it will fail to detect positions methylated at low levels , given the conservative cutoff we set of filtering out rarely methylated positions . Indeed , apart from the positions we described with high methylation levels , there were thousands of additional positions with significant p values , but with much lower methylation levels . About 95% of these sites had methylation levels lower than 10% , including many putative positions in rRNAs and tRNAs . As most of these positions have not been previously reported , despite the fact that tRNAs and rRNAs have been extensively characterized , it is highly probable that the majority of these sites represent artifacts from various sources . This notwithstanding , this group contains a single site in the E . coli 16S rRNA ( position 1407 ) , which is , indeed , known to be modified . This suggests that there may be additional real methylated nucleotides within this group . However , our approach is currently unable to differentiate between rarely methylated positions and experimental artifacts . The emergence of ultra-high throughput sequence interrogation technologies is revolutionizing research on RNA modifications . Application of such technologies , including RNA-seq and RIP-seq , to study such modifications in eukaryotes has recently revealed that mRNAs carry a plethora of conserved modifications , such as A-to-I [40] , m6A [35] and m5C [20] . Nevertheless , the field of mRNA modifications is still in its infancy , and even in eukaryotes , where several modifications have already been extensively characterized , their functional consequences are still poorly understood . Our report that mRNAs of archaea also carry such modifications raises the intriguing possibility that mRNA modifications in prokaryotes form an additional layer of gene regulation that has not yet been addressed . Whether the m5C modifications are functionally relevant , and whether other RNA modifications also exist on prokaryotic mRNAs , remains to be determined .
Escherichia coli ( MG1655 ) cells were grown in LB medium to log phase at 37°C . Bacillus subtilis str . 168 cells were grown in LB medium to mid log phase at 37°C . S . solfataricus P2 ( DSMZ 1617 ) cells were grown in defined modified Brock's mineral medium with final pH 3 . 5 , and S . cerevisiae ( BY4741 ) cells were grown in YPD medium to log phase at 30°C . All cells pellets were suspended in RNAlater ( Ambion , AM7022 ) for 30 min at room temperature . Pre-treatments for RNA isolation included Lysozyme digestion for E . coli cells , glass beads vortex for B . subtilis and Lyticase digestion for S . cerevisiae . Total RNA was then extracted using Tri-Reagent ( Molecular Research Center Inc . ) according to manufacturer's instructions . RNA samples were treated with Turbo DNA-free kit ( Ambion ) and rRNA was removed from the E . coli and B . subtilis samples using MICROBExpress mRNA enrichment kit ( Ambion ) . Bisulfite conversion was performed based on the protocol by Schaefer et al . [21] . Briefly , one microgram of each RNA sample was incubated in DNA protect buffer and bisulfite mix , EpiTect Bisulfite Kit ( Qiagen ) , through 6 cycles of denaturation step at 70°C , followed by a deamination reaction step of 1 hour at 60°C . The RNA was purified from the bisulfite reaction mix using Micro Bio-Spin 6 columns ( Bio-Rad ) and treated with 0 . 5M Tris–HCl , pH 9 at 37°C for 1 hour . RNA was then washed on YM-10 microcon ( Millipore ) 5 times with 0 . 5 ml ultra-pure water . RNA samples were analyzed using the Bioanalyzer ( Agilent ) to assess degradation status . For all 4 organisms , untreated RNA and bisulfite-treated RNA were used as templates for cDNA synthesis using SuperScript II RT ( Invitrogen ) and random hexamers according to manufacturer's protocol . cDNA was amplified by PCR ( ABgene ) using regular primers ( for untreated RNAs ) and primers specific to cytosine-deaminated sequences ( for bisulfite treated RNAs ) . In order to improve sequencing results , generic M13 sequences ( containing all 4 bases ) were combined with the 5′ end of all the deaminated primer sequences . PCR products were separated on agarose gel and extracted using gel extraction kit ( Qiagen ) . Sanger sequencing of the PCR product was conducted from the M13 primer sequence . Equal amounts ( 50 ng ) of each of the four organism's bisulfite-treated total RNA were mixed together . The mixed bisulfite-treated RNA sample was used as a template for cDNA library preparation according to the mRNA-seq Illumina protocol , omitting the polyA-based mRNA purification step . In brief , RNA was first fragmented by divalent cations at 94°C for 5 min . Double stranded cDNA was generated using SuperScriptII and random primers . cDNA was then end-repaired , adenylated and end-ligated to adapters . Following gel separation , a ∼200 bp fragment was gel-purified . The cDNA library was further amplified and sequenced using 40 single-read cycles on a Genome Analyser II ( Illumina ) . 40-nt long reads were aligned using Novoalign ( Novocraft Technologies Sdn Bhd , http://www . novocraft . com ) to bisulfite-converted genomes of Bacillus subtilis ( NC_000964 ) , Escherichia coli ( NC_000913 ) , Sulfolobus solfataricus ( NC_002754 ) and Saccharomyces cerevisiae ( NC_001133-NC001148 ) downloaded from the NCBI website . Reads that did not align to the reference sequence at their original length , were iteratively trimmed by two base-pairs from the end of the read and then realigned , as long as their length exceeded 35 nt . To allow identification of methylated positions in tRNAs and rRNAs occurring in multiple copies in the genome , reads mapping to multiple regions in the genome were randomly assigned to one such region . Parameters used for the indexing step were “–b” , and for alignment were “-t 60 -h 120 -b 4 -l 35 -s 2 -F STDFQ -r Random -u 6” . For the transcriptome-based study of rRNAs and tRNAs , reads were aligned against a database consisting of unique copies of all fully processed tRNAs obtained from the tRNAdb [41] , and against a single representative of each of the 5S , 16S and 23S rRNA genes in each of the four organisms . Identical alignment parameters were set as above . The rational for this approach was based on two considerations: First , to prevent dilution of reads to multiple , identical copies of the same tRNAs and rRNAs , and second , to prevent loss of reads due to differences between the transcriptome and the genome , such as in cases of tRNA introns that are excised during maturation of the tRNA molecule . For each cytosine in the genome in both orientations , the number of reads in which the cytosine underwent conversion to uridine ( suggesting that it was not methylated ) , and those in which it was not converted ( suggesting methylation ) were counted . To eliminate artifacts of various sources , the following filters were applied: ( 1 ) identical reads were considered a single read , to eliminate PCR amplification artifacts , ( 2 ) reads with ≥3 unconverted cytosines were eliminated , as they might reflect transcripts that for did not obtain sufficient exposure to bisulfite , ( 3 ) only positions with a sequencing quality >20 were counted , to eliminate low-quality positions . In the transcriptome based analysis we did not apply the first filter , since for the highly expressed rRNA and tRNA molecules identical reads generally do not reflect PCR artifacts , but merely reflect the high coverage of these genes . Each position was then assigned a methylation level , equivalent to the proportion of reads in which it was not converted , and a p-value determining the significance of the number of converted and non-converted reads was assigned using Fisher's exact test against the null hypothesis that all reads were converted . Identified methylation sites that were within 10 nt of an additional site were discarded . Since many genes are present in the genome in multiple copies , following identification of all significantly methylated positions ( p<0 . 01 ) , all positions sharing an identical 31-nt sequence surrounding the identified position were collapsed together into a single sequence . Curation of the data confirmed that this efficiently collapsed together sequences from identical genes . The joint methylation level for each collapsed position was recalculated based on the cumulative number of converted and non-converted reads in its ‘parents’ . For the alignment of tRNAs from tRNAdb [41] , a multiple sequence alignment of tRNAs was generated using mafft v6 . 850b , with the parameters “–maxiterate 1000 –localpair” . The motif AUCG-A/U-G/UG-U/G was searched in all protein coding genes in which the potentially methylated ‘C’ in the center of this motif was covered by at least 3 reads , and methylation ratios were recorded . As a negative control the analysis was repeated taking only motifs with at least three mismatches compared to the consensus ( specifically , demanding a “C” at position 0 , and that position −2 was not an “A” , −1 not a “T” , and +1 not a “G” ) . For the analysis presented in Fig . S1 , we assembled a dataset of 38 sites harboring a consensus and with a methylation P value<0 . 05 . Notably , to gain more statistical power for the downstream analysis , here we did not set a threshold on the minimal methylation level . For each site , its relative position within the gene was calculated , from a scale of 0 to 1 ( representing the 5′ and 3′ ends of a gene , respectively ) . We compared the distributions of these methylation sites to 22 , 665 controls lacking a consensus site , and obtained marginally significant results ( t-test , P = 0 . 07 ) . Two ∼200 bp sequences from the E . coli genome , containing the consensus sequence found in Sulfolobus solfataricus , were amplified via PCR from the genome of E . coli MG1655 ( primers: p1_fwd TAATACGACTCACTATAGGGTCATGCACGGTGTCGTTATT; p1_rev AAAGGTTTCCATGTCGAACG; p2_fwd TAATACGACTCACTATAGGGGCTGTGGTGATCAGTGTGCT; p2_rev TGGCGTTGATAAAACTGACG ) . These amplified sequences were used as template for an in-vitro transcription reaction using MaxiScript kit ( Ambion , AM1344 ) to produce RNA transcripts . These RNA transcripts were spiked into Sulfolobus solfataricus total RNA prior to the bisulfite conversion treatment . Reads obtained from this library were aligned separately against the S . solfataricus genomes , and against a synthetic genome comprising the two E . coli templates . Non-converted cytosines were quantified using the pipeline and filters described above . Sulfolobus solfataricus total RNA was chemically fragmented ( Ambion , AM8740 ) to an average size of ∼100 bp and was subjected to immunoprecipitation with an anti m5C monoclonal antibody ( Diagenode , MAb-081-010 ) or anti-hm5C polyclonal antibody ( Diagenode , pAb-HMC-050 ) based on the protocol described in ref [35] , using two rounds of IP . The precipitated RNA was used for dsDNA Illumina library construction . Illumina MiSeq sequencing of 30 bp paired-end was used to characterize rRNA modifications . Then , Illumina HiSeq run was conducted on the same library with single-end 50 bp reads to gain enough coverage to analyze the mRNA positions . | Ribonucleic acids are universally used to express genetic information in the form of gene transcripts . Although we envision RNA as a mere copy of the DNA four-base code , modification of specific RNA bases can expand the information code . Such modifications are abundant in transfer RNA ( tRNA ) and ribosomal RNA ( rRNA ) , where they contribute to translation fidelity and ribosome assembly . Recent studies in eukaryotes have shown that mRNA modifications such as RNA-editing ( conversion of an adenosine base to inosine ) , N6-adenine methylation ( m6A ) , and 5-methylcytidine ( m5C ) can change the coding sequence , alter splicing patterns , or change RNA stability . However , no mRNA modifications in bacteria or archaea have been documented to date . We have used an approach that enables mapping of the m5C modifications across all expressed genes in a given organism . Applying this approach on model bacterial , archaeal , and fungal microorganisms enabled us to reveal the modified RNA bases in these organisms , and to provide an accurate and sensitive map of these modifications . In archaea , we documented multiple genes whose mRNAs are subject to RNA modification , suggesting that similar to eukaryotes , these organisms may utilize mRNA modifications as a mechanism for gene regulation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"bacteriology",
"bacterial",
"biochemistry",
"escherichia",
"coli",
"genome",
"complexity",
"prokaryotic",
"models",
"model",
"organisms",
"gene",
"regulation",
"bacillus",
"subtilis",
"molecular",
"genetics",
"biology",
"genomics",
"microbiology",
"computational",
"biology"
]
| 2013 | Transcriptome-Wide Mapping of 5-methylcytidine RNA Modifications in Bacteria, Archaea, and Yeast Reveals m5C within Archaeal mRNAs |
Aphids are amongst the most devastating sap-feeding insects of plants . Like most plant parasites , aphids require intimate associations with their host plants to gain access to nutrients . Aphid feeding induces responses such as clogging of phloem sieve elements and callose formation , which are suppressed by unknown molecules , probably proteins , in aphid saliva . Therefore , it is likely that aphids , like plant pathogens , deliver proteins ( effectors ) inside their hosts to modulate host cell processes , suppress plant defenses , and promote infestation . We exploited publicly available aphid salivary gland expressed sequence tags ( ESTs ) to apply a functional genomics approach for identification of candidate effectors from Myzus persicae ( green peach aphid ) , based on common features of plant pathogen effectors . A total of 48 effector candidates were identified , cloned , and subjected to transient overexpression in Nicotiana benthamiana to assay for elicitation of a phenotype , suppression of the Pathogen-Associated Molecular Pattern ( PAMP ) –mediated oxidative burst , and effects on aphid reproductive performance . We identified one candidate effector , Mp10 , which specifically induced chlorosis and local cell death in N . benthamiana and conferred avirulence to recombinant Potato virus X ( PVX ) expressing Mp10 , PVX-Mp10 , in N . tabacum , indicating that this protein may trigger plant defenses . The ubiquitin-ligase associated protein SGT1 was required for the Mp10-mediated chlorosis response in N . benthamiana . Mp10 also suppressed the oxidative burst induced by flg22 , but not by chitin . Aphid fecundity assays revealed that in planta overexpression of Mp10 and Mp42 reduced aphid fecundity , whereas another effector candidate , MpC002 , enhanced aphid fecundity . Thus , these results suggest that , although Mp10 suppresses flg22-triggered immunity , it triggers a defense response , resulting in an overall decrease in aphid performance in the fecundity assays . Overall , we identified aphid salivary proteins that share features with plant pathogen effectors and therefore may function as aphid effectors by perturbing host cellular processes .
Like most plant parasites , aphids require intimate associations with their host plants to gain access to nutrients . Aphids predominantly feed from the plant phloem sieve elements , and use their stylets to navigate between the cells of different layers of leaf tissue during which plant defenses may be triggered . Indeed , aphid feeding induces responses such as clogging of phloem sieve elements and callose formation , which are suppressed by the aphid in successful interactions with plant hosts [1] . In addition , some aphid species can alter host plant phenotypes , by for example inducing the formation of galls or causing leaf curling [2] indicating that there is an active interplay between host and aphid at the molecular level . During probing and feeding , aphids secrete two types of saliva: gelling saliva , which is thought to protect stylets during penetration , and watery saliva , which is secreted into various plant host cell types and the phloem [3] . The secretion of aphid saliva directly into the host-stylet interface [4] , suggests that molecules present in the saliva may perturb plant cellular processes while aphids progress through different feeding stages . Interestingly , the knock-down of the C002 salivary gene in Acyrthosiphon pisum ( pea aphid ) negatively impacts survival rates of this aphid on plant hosts [5] , [6] . Furthermore , proteomics studies based on artificial aphid diets showed the presence of secreted proteins , including C002 , in aphid saliva indicating that these proteins are delivered inside the host plant during feeding [7] , [8] . However , whether and how these aphid salivary proteins function in the plant host remains elusive . Suppression of host defenses and altering host plant phenotypes is common in plant-pathogen interactions and involves secretion of molecules ( effectors ) that modulate host cell processes [9] , [10] . Therefore it is likely that aphids , similar to plant pathogens , deliver effectors inside their hosts to manipulate host cell process enabling successful infestation of plants [9] . Effector-mediated suppression of plant defenses , such as Pathogen-Associated Molecular Pattern ( PAMP ) -triggered immunity ( PTI ) , generally involves the targeting of a plant virulence target , or operative target [11] . However , plant pathogen effectors that are deployed to suppress host defenses are recognized by plant disease resistance ( R ) proteins in particular host genotypes , resulting in effector-triggered immunity ( ETI ) [12] . Interestingly , the R proteins that recognize plant pathogens and those that confer resistance to aphids , such as Mi-1 . 2 and Vat , share a similar structure , and contain a nucleotide binding site ( NBS ) domain and leucine rich repeat ( LRR ) regions [13]–[15] . The Mi-1 . 2 resistance gene confers resistance in tomato to certain clones of Macrosiphum euphorbiae ( potato aphid ) , two whitefly biotypes , a psyllid , and three nematode species [16]–[19] , indicating that there is significant overlap in plant pathogen and aphid recognition in plants . In addition , aphid resistance conferred by several resistance genes was shown to be race-specific [16] , [20] . This suggests that depending on their genotype , certain aphid clones may be able to avoid and/or suppress plant defenses and fits with the gene-for-gene model in plant-pathogen interactions [21] . Therefore , it is likely that not only plant pathogens , but also aphids , secrete effectors that in addition to targeting host cell processes may trigger ETI depending on the host genotype . Plant pathogen effectors generally share the common feature of modulating host cell processes [22] . Various assays have been developed to identify the functions of effectors from bacterial and eukaryotic filamentous plant pathogens [22]–[24] . One important and common function of plant pathogen effectors is the suppression of PTI . This activity is especially common among type III secretion system ( T3SS ) effectors . For example , the large majority of Pseudomonas syringae DC3000 effectors can suppress PTI responses , including the oxidative burst [25] . However , effectors from eukaryotic filamentous plant pathogens can also suppress PTI , as demonstrated for the AVR3a effector from Phytophthora infestans , which suppresses cell death induced by the PAMP-like elicitor INF1 [26] , [27] . Another activity of plant pathogen effectors is the induction of phenotypes in plants . For example , several effectors , including CRN2 and INF1 , from the oomycete plant pathogen P . infestans induce cell death upon overexpression in planta [28] , [29] , whereas other effectors , like AvrB from P . syringae DC3000 induce chlorosis [30] . Also , overexpression of effector proteins from plant pathogenic nematodes in host plants can affect plant phenotypes , as shown for the Heterodera glycines CLE protein Hg-SYV46 that alters host cell differentiation [31] . As effectors exhibit functions important for pathogenicity , their deletion can have detrimental effects on pathogen virulence . However , due to redundancy , the knock-down or deletion of single effectors does not always impact virulence . On the other hand , overexpression of plant pathogen effectors can enhance pathogen virulence , as shown for active AvrPtoB , which enhances virulence to P . syringae DC3000 in Arabidopsis [32] , and for the H . schachtii effector 10A06 that , in addition to altering host plant morphology , increases nematode susceptibility in Arabidopsis [33] . We exploited publicly available aphid salivary gland sequences to develop a functional genomics approach for the identification of candidate aphid effector proteins from the aphid species Myzus persicae ( green peach aphid ) based on common features of plant pathogen effectors . Data mining of salivary gland expressed sequences tags ( ESTs ) identified 46 M . persicae predicted secreted proteins . Functional analyses showed that one of these proteins , Mp10 , induced chlorosis and weak cell death in Nicotiana benthamiana , and suppressed the oxidative burst induced by the bacterial PAMP flg22 . In addition , we developed a medium-throughput assay , based on transient overexpression in N . benthamiana , that allows screening for effects of aphid candidate effectors on aphid performance . Using this screen , we identified two candidate effectors , Mp10 and Mp42 , that reduced aphid performance and one effector candidate , MpC002 , that enhanced aphid performance . In summary , we found aphid secreted salivary proteins that share features with plant pathogen effectors and therefore may function as aphid effectors by perturbing host cellular processes .
We developed a functional genomics approach to identify candidate effectors from M . persicae using 3233 publicly available aphid salivary gland ESTs [34] . We hypothesized that aphid effectors are most likely secreted proteins that are delivered into the saliva through the classical eukaryotic endoplasmic reticulum ( ER ) -Golgi pathway of the salivary glands . A feature of proteins secreted through this pathway is the presence of an N-terminal signal peptide . Therefore , we used the SignalP v3 . 0 program [35] to predict the presence of signal peptides in the amino acid sequences encoded by the open reading frames ( ORFs ) found in salivary gland ESTs . Out of 5919 amino acid sequences corresponding to predicted ORFs , we identified 134 nonredundant sequences with signal peptide ( Figure 1A ) . Out of these 134 proteins , 19 were predicted to contain a transmembrane domain in addition to the signal peptide , and are therefore likely to remain in the salivary gland membrane upon secretion . Hence , 115 predicted secreted proteins remained . In order to investigate the M . persicae candidate effector protein in functional assays , we started with the cloning of 46 candidates that corresponded to full-length sequences within the set of 115 candidates . Effectors are subject to diversifying selection because of the co-evolutionary arms race between host and pathogen proteins [36] , [37] . Therefore , we used the presence of amino acid polymorphisms among alignments of deduced protein sequences of M . persicae and A . pisum ESTs as an additional criterion . Three candidates did not fulfill this criterion and were removed from our candidate set bringing the total to 43 candidates . We applied a similar data mining approach as described above to 4517 publicly available salivary gland ESTs from A . pisum , thereby identifying 24 candidates ( Table S1 ) . In the A . pisum salivary gland ESTs we predicted only 1751 ORFs , explaining the relatively low number of A . pisum candidates . A total of three candidates were found in both M . persicae and A . pisum datasets ( combinations Mp1/Ap1 , Mp5/Ap7 and Mp16/Ap4 ) . The remaining 21 non-overlapping A . pisum candidates were subjected to BLAST searches ( E value<10−15 ) against all available M . persicae ESTs to identify putative M . persicae homologs . This led to the identification of three M . persicae sequences ( Mp3 , Mp54 and MpC002 ) that were added to the M . persicae candidate effector dataset bringing the total to 46 ( Figure 1A , Table S2 ) . Interestingly , for two candidates , Mp39 and Mp49 , no similar sequences were present in the publicly available aphid sequence datasets , including the A . pisum genome sequence ( Table S2 ) . Also , no homologs of these proteins were identified by BLAST searches against GenBank nucleotide and protein databases ( E value<10−5 ) . This suggests these proteins may be specific to M . persicae . A total of 11 candidates were shared between the independent salivary gland EST datasets from M . persicae and A . pisum but were not present in gut ESTs from M . persicae ( Table S2 ) providing support that the corresponding proteins may share a similar function in both these aphid species . For four candidates matches were found in gut ESTs from M . persicae , suggesting these proteins may be derived from salivary gland contaminants in dissected gut tissues and not function uniquely in the salivary gland or saliva . Indeed , gene expression analysis of Mp51 in various aphid tissues dissected from aphids fed on N . benthamiana confirmed that this gene is specifically expressed in the aphid gut ( Figure S1 ) . In contrast , candidate effector genes Mp1 , Mp2 , Mp10 , Mp30 , Mp42 , Mp47 , Mp50 and MpCOO2 , were expressed in aphid heads and salivary glands but not in aphid guts ( Figure S1 ) , suggesting that their corresponding proteins are indeed produced in the salivary glands . Furthermore , Mp1 and MpCOO2 were previously identified in saliva of M . persicae using a proteomics-based approach [7] confirming that these two proteins are secreted into aphid saliva . To investigate the functions of the 46 effector candidates , we amplified the corresponding sequences encoding the mature proteins , without the signal peptide encoding sequences , from M . persicae cDNA for cloning ( Figure 1B ) . To preserve the authentic sequence in the 3′ end of the ORF , we designed reverse primers in the 3′ untranslated regions ( UTRs ) based on EST sequences when possible . Amplicons were cloned in a 35S cassette and corresponding constructs were transformed directly into Agrobacterium tumefaciens followed by sequencing ( Figure 1B ) . Two out of the 46 candidates , Mp7 and Mp38 could not be amplified from M . persicae cDNA . Of the remaining 44 candidates , four ( Mp6 , Mp17 , Mp33 and Mp35 ) were represented by two polymorphic forms , with polymorphisms within the mature protein portion . Except for one of the polymorphic Mp6 sequences , all sequences were identical to those in the M . persicae EST databases . To rule out that the polymorphism in Mp6 was due to PCR errors , we repeated the Mp6 PCR and sequencing several times on individual aphids with similar results . Both forms of the four polymorphic candidates were cloned resulting in a total of 48 cloned M . persicae effector candidates . Functional assays were performed based on transient over-expression in N . benthamiana to assess whether the M . persicae candidate effectors 1 ) induce a phenotype in planta , 2 ) suppress PAMP-triggered immunity and 3 ) affect the ability of M . persicae aphids to reproduce ( fecundity ) ( Figure 1B ) . We assessed fecundity of M . persicae lineage RRes ( genotype O ) , which can utilize N . benthamiana as a host . Several plant pathogen effectors induce a phenotype upon overexpression in planta , which may reflect their virulence activity [22] . Hence , we performed transient overexpression of the effector candidates in N . benthamiana by agroinfiltration to screen for the induction of phenotypes . Out of the 48 , one candidate effector , Mp10 , induced chlorosis starting from 2 days post inoculation ( dpi ) ( Figure 2A ) . In addition , we observed local cell death in a low number of infiltration sites ( Figure S2A , S2B , S2C , S2D ) . The phenotype was not affected by co-expression with the silencing suppressor p19 ( Figure S2E ) . To independently confirm the phenotype , we expressed Mp10 in N . benthamiana using a Potato virus X ( PVX ) -based vector ( PVX-Mp10 ) . Systemic PVX-based overexpression of Mp10 induced systemic chlorosis in N . benthamiana starting at 10 dpi ( Figure 2B ) . This also suggests that the Mp10 response is not dependent on the presence of Agrobacterium . To determine whether the response to Mp10 was specific to N . benthamiana , we infected N . tabacum , Solanum lycopersicum ( tomato ) and N . benthamiana plants with PVX-Mp10 in parallel . Starting at around 10 dpi , systemic chlorosis was observed in N . benthamiana expressing PVX-Mp10 , but not in control PVX-infected plants ( Figure 2B ) . Whereas mosaic symptoms were observed in S . lycopersicum , indicative of PVX infection , no Mp10-induced chlorosis was observed ( Figure 2C; Figure S3A , S3B ) . Mp10 expression was confirmed by semi-quantitative RT-PCR in systemically PVX-Mp10 infected leaves of S . lycopersicum suggesting that the lack of symptoms is not due to a loss of the Mp10 sequence from PVX-Mp10 ( Figure 2E ) . In contrast , N . tabacum plants infected with PVX-Mp10 did not show mosaic symptoms indicative of virus infection , while N . tabacum inoculated with PVX alone did ( Figure 2D; Figure S2B ) . No Mp10 expression could be detected in leaves of N . tabacum plants inoculated with PVX-Mp10 , whereas expression of the viral coat protein was detected , indicating that PVX itself did systemically spread in N . tabacum ( Figure 2E ) . In contrast , PVX-Mp42 did spread systemically in N . benthamiana , N . tabacum and S . lycopersicum , indicating that this aphid protein can be systemically expressed in these plant species using PVX ( Figure S4 ) . It is possible that PVX-Mp10 may evoke an avirulence response in N . tabacum causing the selection of PVX without the Mp10 insert . Loss of foreign gene fragments from the PVX genome has been reported previously and is most likely due to selection pressures forcing virus recombination [38] . The lack of mosaic symptoms in PVX-Mp10-inoculated N . tabacum plants is possibly due to the initially low abundance of recombined PVX-virus as compared to the vector control . The SGT1 protein , an ubiquitin-ligase associated protein , is required for plant cell death responses , including those involved in plant resistance [39] . To investigate whether SGT1 is required for the Mp10 chlorosis response , we generated SGT1-silenced N . benthamiana plants using Tobacco rattle virus ( TRV ) -based virus-induced gene silencing ( VIGS ) . Silenced plants ( treated with TRV-SGT1 ) and control plants ( treated with TRV ) ( Figure 2H ) were infiltrated with Agrobacterium strains expressing Mp10 or the positive control INF1 , an elicitin from P . infestans that induces cell death in control plants , but not in SGT1-silenced plants [40] . Both the Mp10-induced chlorosis and the INF1-induced cell death were pronouncedly reduced in the SGT1-silenced plants , but not in the TRV-treated control plants ( Figure 2F and 2G ) , indicating SGT1 is required for these chlorosis and cell death responses . Suppression of PTI induced by PAMPs like flg22 and chitin is a common feature of plant pathogen effectors . To determine whether aphid candidate effectors can suppress PTI , we assessed whether any of our 48 candidates suppressed the oxidative burst response induced by the bacterial PAMP flg22 . We decided to screen for suppression of the oxidative burst induced by flg22 only , as this PAMP gives a strong and consistent oxidative burst response in N . benthamiana , which is convenient for use in large screens . N . benthamiana leaf discs overexpressing the effector candidate genes under control of the 35S promoter were challenged with the flg22 elicitor and the production of reactive oxygen species ( ROS ) was measured using a luminol-based assay [41] . The bacterial effector AvrPtoB , a suppressor of the flg22-mediated oxidative burst response [42] , was included as a positive control . We found that Mp10 suppresses the flg22-induced oxidative burst in leaf discs harvested 2 days post agroinfiltration ( three replicated experiments ) ( Figure 3A ) , whereas other candidate effectors did not ( data not shown ) . Although the level of suppression by Mp10 was significant compared to that of the empty vector control , it was not as effective as AvrPtoB . We tested whether Mp10 also suppressed the oxidative burst induced by a fungal PAMP , chitin , and found that while Mp10 suppressed the flg22 response , no suppression of the chitin-induced oxidative burst was observed ( Figure 3B ) . Thus , Mp10 specifically suppresses the oxidative burst induced by the PAMP flg22 . We developed a medium-throughput 24-well plate assay to assess M . persicae fecundity on N . benthamiana leaves transiently overexpressing the 48 candidate effectors ( Figure 4A ) . Leaf discs were harvested from infiltrated leaves one day after agroinfiltration and placed upside down on water agar in 24-well plates . Four first-instar nymphs were placed on each leaf disc and the plate was incubated up-side-down under a light source . Aphids were moved every 6 days to plates with freshly infiltrated leaf discs , as expression levels of green fluorescent protein ( GFP ) in leaf discs were constant during 6 days and then tapered off ( Figure S5 ) . The aphids placed initially on the leaf discs generally started producing nymphs after about 10–11 days . Nymph production ( fecundity ) was assessed on day 12 , 14 and 17 by counting and removing newly produced nymphs on each leaf disc . The total nymph production per adult was calculated and compared among the treatments and GFP and vector controls . In our initial screens , in which candidate effector constructs were infiltrated on different leaves and not always side-by-side with the vector control , we identified 14 candidates that either enhanced or reduced aphid fecundity by one time the standard error compared to the empty vector ( EV ) control ( Figure S6 ) . To confirm the effect on aphid fecundity of these 14 candidates , we conducted additional assays in which the candidates were infiltrated side-by-side with the vector control ( EV ) on the same leaves . Two candidates , Mp10 and Mp42 , reduced aphid fecundity in three repeated confirmation assays compared to the vector control ( Figure 4B ) . In addition , one candidate , MpC002 , enhanced aphid fecundity in three repeated confirmation assays compared to the vector control ( Figure 4B ) . Transient overexpression of Mp10 did not induce chlorosis in leaf discs ( Figure S7 ) or leaves that were detached from the plant 24hrs after infiltration ( data not shown ) . Thus , leaves need to be attached to the plant for chlorosis to occur and the chlorosis itself was therefore not likely responsible for the observed reduction in aphid performance . In summary , we have developed a novel assay to screen for effects of in planta expressed aphid salivary proteins on aphid performance and thereby identified three candidates that potentially function as effectors by eliciting plant defenses or promoting aphid infestation of host plants . To determine whether the candidates that alter aphid fecundity , ( i . e . Mp10 , Mp42 , and MpC002 ) share similarity to proteins of known or predicted function , we performed BLAST searches against the GenBank non-redundant ( nr ) protein database ( E value<10−5 ) . One of the three candidates , Mp10 showed homology to an insect protein of predicted function , the olfactory segment D2-like protein ( OS-D2-like protein ) . The OS-D2-like protein is a member of a family of chemosensory proteins in aphids that contain the conserved cysteine pattern CX6CX18CX2C [43] . Mp10 also shows similarity to chemosensory proteins ( CSPs ) from other insects ( E value<10−5 ) , including the CSP5 protein from the mosquito Anopheles gambiae ( Figure 5A ) . The four cysteines in Mp10 are conserved among different members of the CSP family [44] , [45] ( Figure 5A ) . Among the aphid sequences similar to Mp10 , polymorphisms are predominantly present after the predicted signal peptide sequence , in the mature protein region . For Mp42 and MpC002 , similar sequences were identified in the genome sequence of the aphid species A . pisum only , but these proteins have no similarities to proteins with known functions . Alignment of Mp42 to a putative A . pisum homolog shows strong sequence divergence especially in the mature protein regions ( Figure 5B ) . Finally , alignment of MpC002 to A . pisum C002 shows sequence divergence consisting of both amino acid polymorphisms and a 45 amino acid gap in A . pisum C002 after the predicted signal peptide sequence ( Figure 5C ) . The presence of polymorphisms mainly in the mature protein regions may reflect that the functional domains of these proteins have diversified due to distinct selective pressures .
Aphids , like other plant parasites , deliver repertoires of proteins inside their hosts that function as effectors to modulate host cell processes . These insects most likely secrete effectors into their saliva while progressing through the different plant cell layers during probing and feeding . The identification and characterization of these proteins will reveal new insights into the molecular basis of plant-insect interactions . Here , we have described a functional genomics pipeline to identify M . persicae effector candidates as well as various assays to determine whether the candidates share features with plant pathogen effectors . Using this approach , we identified three candidate effectors , Mp10 and Mp42 , MpC002 that modulate host cell processes and affect aphid performance . The induction of chlorosis and local cell death by Mp10 can reflect a genuine effector activity of this aphid salivary protein . Ectopic expression of bacterial TTSS as well as filamentous plant pathogen effectors can affect host immunity and induce a variety of phenotypes in plants , ranging from chlorosis to necrosis [22] , [28] . Both the P . syringae type III effectors AvrB [30] and HOPQ-1 [46] induce chlorosis and for AvrB this activity is plant genotype specific [47] . No Mp10 induced chlorosis was observed in tomato despite expression levels of PVX-Mp10 that were comparable to N . benthamiana . This suggests that the Mp10 response was specific for N . benthamiana . Interestingly , PVX-Mp10 was unable to infect N . tabacum , suggesting this protein may induce an unknown defense mechanism that is effective against PVX-Mp10 . There are several possibilities that may explain the Mp10 phenotype in a biologically relevant context . The first possibility is that the artificially high expression of Mp10 could lead to the induction of the chlorosis/local cell death phenotype and therefore this response could be an artifact of the Agrobacterium-mediated overexpression assay . However , in this case we would expect that the induction of chlorosis and local cell death by Mp10 would be more widespread in various plant species , and would also be observed in N . benthamiana leaf discs or detached leaves . Another possibility is that the high expression of Mp10 could lead to excessive targeting of the operative target as well as other host proteins leading to an exaggeration of the true virulence activity [22] . Finally , the induction of chlorosis and local cell death could reflect avirulence activity of Mp10 . Feeding of M . persicae is known to induce chlorosis and premature leaf senescence in plants , and this response is related to PAD4-mediated defense responses [48] . Therefore , Mp10 may exhibit an avirulence activity specifically in Nicotiana spp resulting in chlorosis and local cell death . The induction of chlorosis in N . benthamiana by P . syringae effector AvrB is thought to be due to weak activation of TAO1 , an NBS-LRR protein , and requires the plant-signaling component Rar1 [49] . We found that chlorosis induction by Mp10 requires the co-chaperone SGT1 , which is required for activation of NBS-LRR proteins and plant resistance responses [39] . Therefore , Mp10 may activate an NBS-LRR resistance protein resulting in ETI ( further discussed below ) . We also found that Mp10 suppressed the ROS response induced by flg22 , suggesting that suppression of PTI may be a feature shared by plant pathogens and insects . Possibly , the flg22-induced signaling pathway may not be specific to bacteria as other ( non-bacterial ) PAMPs can induce this pathway . Also , plants may have a PTI pathway ( s ) that is induced by an unknown insect PAMP ( s ) and partially overlaps with the signaling pathway induced by flg22 . To date the role of perception of PAMP-like molecules in plant-insect interactions remains elusive . However , chitin is a major structural component of the insect cuticle . Degradation of chitin by plant chitinases generates fragments that induce PTI [50] . Whether the chitin in the insect cuticle is degraded to induce plant defenses during interaction with host plants remains to be investigated . It has been hypothesized that sheath saliva protects the insect stylets , which mainly consist of chitin , from triggering plant defenses [51]–[53] , potentially including PTI . Recent studies showed that insect saliva of both chewing insects [54] and aphids [55] contains elicitors that induce defense responses in host plants . The nature of these elicitors and their role in triggering PTI are unknown . Despite the lack of an understanding of the role in perception of PAMP-like molecules in plant-insect interactions , our data suggest that an aphid salivary protein , Mp10 , can interfere with a specific PAMP response in a M . persicae host plant . It is therefore possible that Mp10 is a genuine suppressor of PTI . Alternatively , the overexpression of Mp10 may perturb a signaling component in the PTI pathway that is required for recognition of flg22 . As Mp10 induces weak chlorosis starting from 2 dpi , it is possible that this response itself is responsible for loss of the oxidative burst response to PAMPs . However , the Mp10 chlorosis response does not interfere with the oxidative burst triggered by chitin . This suggests that the induction of chlorosis itself may not be sufficient to block the oxidative burst induced by flg22 , but that Mp10 specifically interferes with the flg22-triggered signaling cascade . Despite the suppression of the flg22-mediated oxidative burst by Mp10 , its overexpression in N . benthamiana reduced aphid fecundity . A plausible explanation for this contradictory observation is that Mp10 may activate an NBS-LRR resistance protein resulting in ETI , thereby reducing aphid performance . Thus , the recognition of Mp10 , potentially through ETI , in Nicotiana spp may mask the true virulence activity of this protein . If true , this recognition may be suppressed by other effectors during plant-aphid interactions so that Mp10 can exhibit its virulence function . The leaf disc assay allowed us to generate vast amounts of functional data and directly implicated three effector candidates in plant-aphid interactions . The differences in aphid fecundity observed in our screens were quite variable , requiring replication of experiments . Despite the variation , Mp10 , Mp42 , and MpC002 showed consistent effects on aphid fecundity throughout the individual replicates ( data not show ) . The fecundity was affected by Mp10 , Mp42 , and MpC002 by around 1–1 . 5 nymph produced per adult over a nymph production period of about 6 days . Although these differences may seem small , they are expected to have a large impact on the population size of aphids . Furthermore , M . persicae does not perform as well on N . benthamiana as it does on other hosts , such as Arabidopsis thaliana . Despite the low reproduction level on N . benthamiana , the fecundity differences found in our screens are similar to those observed over a 2-day period on A . thaliana in a study by Pegadaraju et al . [56] which shows that overexpression of PAD4 reduced aphid fecundity by about 1 . 5 nymphs per adult . The number of candidate effectors with an effect on aphid fecundity identified in this study may have been limited by our approach . For example , when the amount of an effector secreted by the aphid is sufficient to modulate host cell processes to promote feeding , in planta overexpression may not necessarily further enhance this effect . Also , there could be differences in plant responses to aphids in leaf discs versus whole plants as certain plant responses to aphids may require an intact plant transport system . Despite these limitations , the development of a novel leaf disc-based assay allowed us to identify three effector candidates from the aphid species M . persicae . Out of the three candidates that affect aphid fecundity in the leaf-disc assays , only Mp10 shows homology to a protein of predicted function , namely OS-D2 , a member of a family of predicted chemosensory proteins . Insect chemosensory proteins ( CSP ) are thought to be involved in olfaction and gustation . Indeed , several CSPs have been specifically found in chemosensory organs and are predicted to function in chemoperception [43] , [57] , [58] . However , for some members of this large protein family functions have been identified in insect development [59] and leg regeneration [60] , suggesting that CSPs may have divergent functions . This is further supported by gene expression studies , which show that some CSPs are specifically expressed in antenna [61] or mouthparts [62] , whereas others are expressed throughout the insect [63] . CSPs are thought to bind small molecules , such as fatty acids , and for some members of this protein family there is evidence that they bind to pheromones [64] , [65] . In the aphid species Megoura viciae a Mp10 homolog was found to be expressed in aphid heads without antenna , indicating that it is not an antenna specific CSP [43] . Interestingly , in mosquitos , members of a family of odorant binding-related proteins , also with predicted functions in olfaction and gustation , are secreted into host cells to manipulate host physiology by for example scavenging host amines [66] . Counteracting host amines has evolved in various blood-feeding insects independently through adaptation of members of the lipocalin or odorant-binding protein families [66] . It is possible that also in plant feeding insects , proteins predicted to be involved in chemosensing are actually involved in early plant host recognition and plant host cell manipulation . For Mp42 and MpC002 no homology was found to proteins of known or predicted function . This is not surprising as most plant pathogen effectors described to date do not show similarity to proteins of known function based on amino acid alignments . The reduction in aphid performance upon overexpression of Mp42 could reflect that Mp42 induces defense responses against aphids in the plant . In contrast , the enhancement of aphid fecundity by MpC002 suggests that this protein may exhibit an effector activity to promote aphid infestation . Indeed , the A . pisum homolog of MpC002 , ApC002 , has been implicated in aphid feeding [5] . Interestingly , ApC002 is secreted into plant tissues during aphid feeding and silencing of ApC002 gene expression reduces aphid survival on plants , but does not affect when aphids feed from diet [67] . However , whether A . pisum performs better upon overexpression of C002 in planta is not known . Our data suggest that the MpC002 homolog may exhibit a similar role in M . persicae , and that this protein is important during plant-aphid interactions . Future studies will be aimed at further characterizing these candidates to identify their plant targets and the molecular processes they perturb .
We downloaded the following datasets in November 2008 from GenBank for bioinformatics analyses . A total of 3233 M . persicae salivary gland ESTs , 27868 M . persicae ESTs ( all available ESTs ) , and 2558 M . persicae gut ESTs [34] , as well as 4517 A . pisum salivary gland ESTs ( GenBank accessions DV747494-DV752010 ) . For similarity searches against the A . pisum genome sequence , we obtained the whole shotgun genome sequence scaffolds from GenBank ( accessions EQ110773-EQ133570 ) in May 2010 . The pipeline for the identification of M . persicae candidate effectors was developed as follows . The 3233 salivary gland ESTs from M . persicae were subjected to ORF calling . More specifically , we performed translations of all possible ORFs of 70+ amino acids , defined by an ATG to stop or an ATG to the end of a sequence , from both strands of the cDNA . We then applied the SignalP v 3 . 0 program [35] to predict the presence of signal peptides in the amino acid sequences with an HMM score cut-off value of >0 . 9 and a predicted cleavage site within the amino acid region 1–30 . As some predicted secreted proteins were represented multiple times within the M . persicae salivary gland EST dataset , we used BLASTP searches to remove redundant sequences . Alignments were inspected manually and sequences that showed >95% identity throughout most of the alignment with an E value<10−10 were classified as being redundant . To remove sequences that in addition the signal peptide also contained a transmembrane domain we used TMHMM v . 2 . 0 . The remaining sequences were searched using TBLASTN ( E value<10−5 ) against all M . persicae and A . pisum ESTs in our datasets as well as the A . pisum genome sequence to assess whether they encoded full-length proteins . Criteria for selecting full-length sequences were: 1 ) the presence of a conserved start and stop site in ESTs within the alignments; 2 ) the absence of a methionine within the alignments upstream of the methionine predicted to be the start of the ORF; 3 ) similarity to a predicted full-length A . pisum protein , when available . The remaining predicted secreted protein sequences were then assessed for the presence of polymorphisms within the alignments described above . Sequences not showing any sequence variation in alignments with M . persicae sequences and that contained up to one amino acid difference in alignments of the mature protein regions with A . pisum sequences were removed from the candidate list . The 4517 salivary gland ESTs from A . pisum were analyzed with the same procedures except that no analyses was performed for the presence of polymorphisms . The amino acid sequences of the predicted secreted proteins ( Table S1 ) were searched using BLASTP ( E value of <10−5 ) against the amino acid sequences of the M . persicae candidates to identify overlap in the datasets . A . pisum candidates without a hit were then searched using TBLASTN against all available M . persicae ESTs ( E value of <10−5 ) to identify M . persicae predicted secreted proteins with sequence similarity . The M . persicae candidates identified using our pipeline and subjected to cloning were designated MpC002 , Mp1-12 , Mp14-17 , Mp19-24 , Mp28-33 , Mp35-37 , Mp39-47 , Mp49-51 , Mp53-54 , wherein Mp stands for M . persicae ( Table S2 ) . The M . persicae colony of lineage RRes ( genotype O ) was maintained in cages on N . tabacum plants . Cages were located in a contained growth room at 18°C under 16 hours of light . A . tumefaciens strain GV3101 was used in molecular cloning and agroinfiltration experiments and were routinely cultured at 28°C in Luria-Bertani ( LB ) media using appropriate antibiotics [68] . All bacterial DNA transformations were conducted by electroporation using standard protocols [68] . Primers were designed for amplification of sequences corresponding to the ORFs encoding the mature proteins ( after the signal peptide encoding sequences ) ( Table S3 ) . To confirm the 3′ end of the ORFs , we designed , where possible , the 3′-primer in the 3′UTR sequence . Sequences were amplified from M . persicae cDNA using Phusion polymerase ( Finnzymes ) and ligated into SpeI/BamHI , SpeI/BglII or BglII/BamHI digested pCB302-3 vector [69] to generate 35S-constructs . To assess whether sequences were polymorphic within the M . persicae clonal lineage used in our studies , we performed sequence analyses of 4 clones per construct . To generate constructs for PVX-based expression , we amplified sequences encoding mature ORFs and ligated these into ClaI/NotI digested pGR106 vector . The PTV vectors used in this study have been described previously [40] . Aphids were dissected in PBS and tissues stored in RNA later . We collected 25 salivary glands , 10 guts , 5 heads and 5 whole aphids . RNA extractions were performed with the NucleoSpin RNA XS kit ( Macherey-Nagel , Germany ) . cDNA was synthesized from 80 ng total RNA per sample using expand reverse transcriptase ( Roche Diagnostics Ltd ) . RT-PCR was performed with gene specific primers for each effector candidate indicated in Table S3 . MpActin primers were used as a control for equal cDNA template amounts . For RT-PCR on plant tissues , 50 mg leaf tissue was ground in liquid nitrogen and RNA was extracted with the RNeasy Plant minikit ( Qiagen ) . cDNA was synthesized from 500ng DNase treated RNA and subjected to PCR reactions with primer pairs Mp10-pvx-F/R and Mp42-pvx-F/R ( Table S3 ) for amplification of Mp10 and Mp42 expressed in PVX , respectively . For amplification of the PVX coat protein we used primer pair PVX-CP-F/R and for amplification of plant tubulin we used the primer pair Tub-F/R ( Table S3 ) . Primers used for RT-PCR on RNA extracted from SGT- and HSP90-silenced plants were described elsewhere [26] . Recombinant A . tumefaciens strains were grown as described elsewhere [70] except that the culturing steps were performed in LB media supplemented with 50 µg/mL of kanamycin . Agroinfiltration experiments were performed on 4–6 week-old N . benthamiana plants . Plants were grown and maintained throughout the experiments in a growth chamber with an ambient temperature of 22°–25°C and high light intensity . For transient overexpression of candidate effectors by agroinfiltration , leaves of N . benthamiana were infiltrated with A . tumefaciens strain GV3101 carrying the respective constructs at a final OD600 of 0 . 3 in induction buffer ( 10mM MES , 10mM MgCl2 , 150 µM acetosyringone , pH = 5 . 6 ) . For agroinfection assays , cotelydons of N . benthamiana , N . tabacum ( cv Petite Gerard ) or S . lycopersicum ( MoneyMaker ) were wound-inoculated with candidate effector clones using P200 pipette tips . Each strain was assayed on 2 replicated plants . As a control , plants were wound-inoculated with A . tumefaciens strains carrying pGR106-Δgfp [26] . Systemic PVX symptoms were scored 14 days post inoculation . We performed gene silencing as described elsewhere [40] . A . tumefaciens suspensions expressing the binary TRV-RNA 1 construct , pBINTRA6 , and the TRV-RNA2 vector , PTV00 or PTV-SGT1 were mixed in 1∶1 ratio ( RNA1- RNA2 ) in induction buffer ( final OD600 is 0 . 6 ) . Leaves were challenged with Agrobacterium strains carrying 35S-Mp10 and 35S-INF1 or the 35S vector . We developed a medium-throughput 24-well assay to test whether overexpression in planta of effector candidates affects aphid nymph production rates . For this purpose , we overexpressed the candidates ( 35S-constructs ) by agroinfiltration in N . benthamiana at a final OD600 of 0 . 3 . One day after infiltration , leaf discs were collected using a cork borer ( No . 7 ) from the infiltration sites and placed upside-down on top of 1ml water agar in 24-well plates . A total of 6 infiltration sites , from 6 different leaves , were used per construct and a total of 4 different constructs per 24-well plate . In initial screens , we infiltrated multiple sets of 4 candidate effectors at the same time , with one set including the vector and GFP controls ( two candidate effectors plus the two controls ) . The 4 candidates within a set were infiltrated side-by-side on the same 6 leaves . Leaf discs from each set of candidates were placed in one 24-well plate ( 6 discs times 4 candidates ) . For the confirmation assays , we performed infiltrations of each candidate effector with the vector control side-by-side on the same 6 leaves , and leaf discs were placed in one 24-wells plate . On each leaf disc , we placed 4 M . persicae first-instar nymphs . The wells in the plate were individually sealed off using a cap of a 5ml BD Falcon round bottomed test tub with the top of the cap removed and covered with mesh . After 6 days , the nymphs were moved to a new 24-wells plate with fresh leaf discs infiltrated with the candidate effector constructs . Another 6 days later , the now adult aphids were again moved to a new 24-well plate with freshly infiltrated leaf discs . The numbers of adults ( initially first-instar nymphs ) were counted 6 , 12 , 14 and 17 days after setting up the first 24-wells plate and the number of newly produced nymphs were counted on day 12 , 14 and 17 . The newly produced nymphs were removed from the wells during counting . Wells wherein all 4 aphids that were initially placed on the discs died were taken out of the analyses . To calculate the production of nymphs per adult aphid , we calculated the average number of nymphs produced per adult by combining the average production rates throughout the experiment . These average production rates were obtained by dividing the number of nymphs on day 12 by the number of adults on day 6 ( calculated per well ) , dividing the number of nymphs on day 14 by the number of adults on day 12 , and dividing the number of nymphs on day 17 by the number of adults on day 14 . To obtain the total average production rate , we calculated the sum of the average production rates for days 12 , 14 and 17 . N . benthamiana leaf discs transiently overexpressing the effector candidates were subjected to a luminescence-based assay [41] . Leaf discs were floated overnight in 200ul water in a 96-well plate . The production of ROS was measured after replacing the water with a solution of luminol ( 20uM ) and horseradish peroxidase ( 1ug ) supplemented with either flg22 peptide ( 100nM ) or chitin ( 100 µg/ml ) . Luminescence was measured using a Varioskan Flash plate reader . A total of 8 discs per construct , obtained from 4 different infiltration sites , were used per replicate . Assays with flg22 to screen the 48 candidates for suppression activity were repeated two times . The assays with chitin and flg22 were repeated three times . All statistical analyses were conducted using Genstat 11 . ROS assay was analysed using a two-sample t-test . Leaf discs fecundity assays were analysed using one-way ANOVA with “construct” as the treatment and “repeat” as the block . Data was checked for approximate normal distribution by visualising the residuals . | Aphids are insects that can induce feeding damage , achieve high population densities , and most importantly , transmit economically important plant diseases worldwide . To develop durable approaches to control aphids , it is critical to understand how aphids interact with plants at the molecular level . Aphid feeding induces plant defenses , which can be suppressed by aphid saliva . Thus , aphids can alter plant cellular processes to promote infestation of plants . Suppression of plant defenses is common in plant pathogens and involves secretion of effector proteins that modulate host cell processes . Evidence suggests that aphids , like plant pathogens , deliver effectors inside their host cells to promote infestation . However , the identity of these effectors and their functions remain elusive . Here , we report a novel approach based on a combination of bioinformatics and functional assays to identify candidate effectors from the aphid species Myzus persicae . Using this approach , we identified three candidate effectors that affect plant defense responses and/or aphid reproductive performance . Further characterization of these candidates promises to reveal new insights into the plant cellular processes targeted by aphids . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"plant",
"biology/plant-biotic",
"interactions",
"genetics",
"and",
"genomics/bioinformatics",
"genetics",
"and",
"genomics/functional",
"genomics"
]
| 2010 | A Functional Genomics Approach Identifies Candidate Effectors from the Aphid Species Myzus persicae (Green Peach Aphid) |
The neurotrophic tyrosine kinase receptor type 2 ( Ntrk2 , also known as TrkB ) and its ligands brain derived neurotrophic factor ( Bdnf ) , neurotrophin-4 ( NT-4/5 ) , and neurotrophin-3 ( NT-3 ) are known primarily for their multiple effects on neuronal differentiation and survival . Here , we provide evidence that Ntrk2 plays a role in the pathologic remodeling of the spleen that accompanies chronic infection . We show that in Leishmania donovani-infected mice , Ntrk2 is aberrantly expressed on splenic endothelial cells and that new maturing blood vessels within the white pulp are intimately associated with F4/80hiCD11bloCD11c+ macrophages that express Bdnf and NT-4/5 and have pro-angiogenic potential in vitro . Furthermore , administration of the small molecule Ntrk2 antagonist ANA-12 to infected mice significantly inhibited white pulp neovascularization but had no effect on red pulp vascular remodeling . We believe this to be the first evidence of the Ntrk2/neurotrophin pathway driving pathogen-induced vascular remodeling in lymphoid tissue . These studies highlight the therapeutic potential of modulating this pathway to inhibit pathological angiogenesis .
Disruption in lymphoid tissue organisation is evident in a wide array of chronic inflammatory disorders , including cancer , infectious disease , psoriasis , liver disorders , autoimmune and metabolic diseases , and the persistence of vascular remodelling has been shown to impair immune responses and promote the chronicity of inflammation [1–4] . It is perhaps no surprise , therefore , that anti-angiogenic drugs are becoming an important therapeutic option for an array of chronic diseases and disorders , including those of infectious origin . HIV , malaria , schistosomiasis and leishmaniasis all promote chronic inflammation and remodelling of lymphoid organs [5–8] , and murine models of these diseases have been used extensively to study the mechanistic basis of stromal cell and vascular remodelling and their consequences in terms of immune dysfunction and disease progression [7 , 9–11] . Splenomegaly is a defining characteristic of the parasitic disease visceral leishmaniasis ( VL ) and like many examples of splenomegaly , VL is associated with a remodeling of splenic architecture [4 , 12] . In murine experimental VL , enlargement of red pulp vessels and neovascularization of the white pulp are prominent . Previously , we demonstrated that administration of the broad spectrum receptor tyrosine kinase inhibitor ( RTKi ) sunitinib maleate to Leishmania donovani-infected mice halted progressive vascular remodeling , reduced mononuclear phagocyte ( MP ) number and improved the efficacy of immune-dependent drugs [4 , 13 , 14] . Compartment-specific remodeling of the red pulp vasculature was subsequently identified as a function of Ly6C+ inflammatory monocytes [14] , whereas these cells played no apparent role in regulating white pulp neovascularisation or the loss of follicular dendritic cells ( FDC ) and fibroblastic reticular cells ( FRC ) , suggesting alternate and independent mechanisms exist to regulate these processes . Neurotrophins and their receptors are novel targets for angiogenic therapies [15 , 16] . For example , Bdnf is recognised as an angiogenic factor , being critical for the establishment of cardiac vasculature during development [17 , 18] , and for promoting Ntrk2-dependent skeletal muscle neoangiogenesis in adult ischemic hind limbs [19] . However , it is the limited tissue expression of the neutrophin receptors , such as Ntrk2 , which makes them so desirable as targets for therapy , with the potential for aberrant expression in pathological conditions in other tissues being largely overlooked . MPs are recognized as playing a significant role in inflammation-induced angiogenesis [2 , 20] and MP are known to be a source of Bdnf and NT-4/5 with their production being increased upon inflammatory stimulation [21 , 22] . Chronic parasitic and bacterial infections have also been associated with elevated levels of Bdnf/BDNF [23–25] . For example , serum levels of BDNF were significantly higher in patients suffering from chronic Chagas disease compared to healthy controls [25] . However , the cellular source ( s ) of neurotrophin were not determined in these studies . A causal link , therefore , between neurotrophin expression and infection-induced , macrophage-directed angiogenesis has not been previously reported . Here , we report that Ntrk2 is aberrantly expressed on endothelial cells that form newly emerging vessels within the white pulp during L . donovani infection and that expression of Bdnf is up regulated in a sub-population of MPs with phenotypic similarity to resident tissue macrophages . Together , these observations have important implications for understanding the pathogenesis of VL and illustrate the potential for aberrant expression of angiogenic receptors and their ligands during inflammation .
To determine whether MPs were involved in the process of white pulp remodeling , we first characterized MPs in the spleen of sunitinib-treated and control mice after the onset of splenomegaly ( d28 post infection;[4] ) . Three distinct populations of CD11c+MHCII+ MPs were identified in infected mice based on CD11b , F4/80 and forward/side scatter profile and morphology: F4/80hiCD11blo cells ( large , macrophage-like morphology; ∼80% containing parasites; Fig . 1A ) ; F4/80loCD11bhi cells ( smaller , classic macrophage morphology , <5% infected; Fig . 1B ) ; and F4/80loCD11blo cells ( small , dendritic cell morphology , no parasites; Fig . 1C ) . These MP populations increased in number 9–14 fold in infected compared to naïve mice . Although administration of sunitinib reduced the number of all populations , this was most marked for the F4/80hiCD11blo cells ( Fig . 1A-C , right panels ) . As the sensitivity of F4/80hiCD11blo MPs to sunitinib treatment correlated with inhibition of white pulp neovascularisation , we further characterized these cells in both untreated infected and sunitinib-treated infected mice . Phenotypically , F4/80hiCD11blo MPs from both groups of mice were CD68+Ly6G/C- CD80+ SIGNR1loCD115+/- ( Fig . 2A-C ) , suggesting that these MPs might be resident rather than inflammatory monocytes / macrophages . To further characterize these cells , we used an in-house MP-targeted oligoarray ( consisting of >500 genes representing multiple GO pathways; S1 Table ) to identify genes differentially expressed ( DE ) in F4/80hiCD11blo MPs vs . a reference population of F4/80hiCD11blo peritoneal MPs . The top DE gene was Slc40a1 , an iron export protein expressed at high levels by tissue resident red pulp macrophages ( Table 1 ) , with many other DE genes being associated with pro-inflammatory and IFNγ-regulated responses , e . g . Cxcl9 , Il18 , Fpr2 and Ccl4 ( Table 2 and S2 Table ) . F4/80hi macrophages were described 30 years ago immediately adjacent to arterioles in the periarteriolar lymphoid sheath [26 , 27] . We localized F4/80hiCD11blo MPs in situ , taking advantage of their expression of SIGNR1 , a receptor for the uptake of FITC-dextran [28] , and the absence of SIGNR1hi marginal zone macrophages in L . donovani infected mice [11] . 74% of cells labeled intra-vitally with FITC-dextran were CD11c+F4/80hiCD11blo ( gating strategy in S1 Fig . ) . In situ , F4/80hiCD11blo FITC-dextran+ MPs were located in either the white pulp region of the spleen or adjacent to the MZ ( Fig . 3A , B ) . F4/80hiCD11blo MPs were most commonly ( 87 . 5%±7 . 2 ) found in association with Meca32+ vessels adjacent to the MZ or no more than a distance of two cell nuclei away ( 8 . 3% ± 8 ) , and located predominantly at vessel junctions ( Fig . 3C , D ) . x-y-z- reconstructions confirmed that they were closely associated with smooth muscle actin ( SMA ) -positive cells ( S1 Video ) . Finally , 3D-rendering of z-plane images confirmed the presence of F4/80hiCD11blo MPs tightly associated with vessel junctions and vasculature that protruded into the white pulp from the marginal sinus ( Fig . 3E-F and S2 Video ) . The close association of F4/80hiCD11blo MPs with vessels penetrating the white pulp suggested that there maybe a causal relationship with vascular remodeling . To determine the angiogenic potential of F4/80hiCD11blo MPs , we used SVEC4–10 mouse endothelial cells in an in vitro tube formation assay . SVEC4–10 cells cultured in the presence of an optimized cocktail of growth factors ( EGM ) migrated , directionally align and formed tube networks to a greater extent than cells cultured in in basal media ( EBM ) ( Fig . 3G ) . Addition of F4/80loCD11bhi and F4/80loCD11blo cells had negligible tube promoting activity ( <20% of maximal activity; difference in tube length of 2 . 2±2 . 3% and 2 . 1±2 . 7% , respectively vs . EBM control ) , whereas F4/80hiCD11blo MPs significantly enhanced mean loop area and tube length ( 5 . 7±3 . 0% and 8 . 6±3 . 1% for F4/80hi CD11blo MPs and EGM respectively , compared to EBM control; Fig . 3G-I ) . Hence , of the splenic MPs tested , only F4/80hiCD11blo MPs displayed angiogenic activity in vitro . To identify factors associated with the angiogenic activity of these MPs , we studied the expression of 84 angiogenesis-related genes using PCR array . 69% ( 58/84 ) of genes were DE ( 2-fold cut-off ) in F4/80hiCD11blo MPs relative to F4/80loCD11bhi MPs ( 46 up , 12 down ) , 50% ( 42/84 ) were DE relative to F4/80loCD11blo MPs ( 28 up , 14 down ) and 55% ( 46/84 ) were DE relative to non-adherent CD11c- splenocytes ( 28 up , 18 down ) . 21% ( 18/84; 14 up , 4 down ) were DE relative to all groups ( Table 2 ) . These included several members of the Tgf family and known ligands for sunitinib , including Pdgf , Kitl Artn , Pgf and Csf1 . Neurotrophic factors Fgf11 , Bdnf and Ntf5 ( neurotrophin 4/5 ) were all up regulated in F4/80hiCD11blo MPs ( Table 1 ) . F4/80hiCD11blo MPs also expressed higher levels of Bdnf protein compared to other MP populations studied ( ΔMFI = 811 vs ΔMFI = 247 and ΔMFI = 139; Fig . 4A ) . Although we cannot rule out a causal link between intracellular infection of F4/80hiCD11blo MPs by L . donovani amastigotes and their Bdnf expression , Bdnf gene expression was not enhanced following L . donovani infection of a variety of other MPs in vitro [29] and in vivo [30] . In summary , these data demonstrate that F4/80hiCD11blo MPs in infected mice: i ) are appropriately positioned in the tissue; ii ) express neurotrophins with known angiogenic activity; and iii ) are pro-angiogenic in vitro . Spleen tissue sections were analyzed for the expression of Ntrk2 , the receptor for Bdnf to investigate a possible role for this pathway in vascular remodeling in vivo . In the naïve spleen , Ntrk2 expression was restricted to a population of F4/80+ red pulp MPs , corresponding to previous reports [31] ( Fig . 4B ) . Expression of Ntrk2 was not altered at d1 post infection , suggesting that induction of Ntrk2 was not the result of inflammatory responses that immediately follow infection . Induced expression was seen in the marginal zone area at day 7 post infection and staining within the white pulp area was beginning to become apparent at day 14 ( S2 Fig . ) . In spleens from chronically infected mice , Nrtk2 was aberrantly expressed on stromal cells , with white pulp vessels showing the highest intensity of staining for Ntrk2 ( Fig . 4C ) . Ntrk2+ vessels were found to be in extremely close association with CD11c+F4/80+ MPs ( Fig . 4C ) . Ntrk2 expression has been reported on endothelial cells and smooth muscle cells in developing and adult cardiac tissue [18 , 32] . Co-staining with Meca32 indicated that Ntrk2 was expressed on splenic endothelial cells , but vessels associated with high levels of SMA staining did not express detectable levels of Ntrk2 ( Fig . 4D ) . As SMA acquisition is a marker of vessel maturity [33] , these data suggest that Ntrk2 is up regulated only on developing or immature blood vessels . Finally , to determine whether this pathway played a functional role in vascular remodeling , we inhibited Ntrk2 signaling in vivo using the small molecule antagonist ANA-12 [34] . Mice were treated for 7 days i . p . with sunitinib or ANA-12 commencing on day 21 p . i . [4] ( Fig . 5A ) . Quantitative image analysis indicated that sunitinib [4] but not ANA-12 inhibited red pulp vascular remodeling ( Fig . 5B and C ) . In contrast , both sunitinib and ANA-12 significantly inhibited white pulp neovascularization ( 50%±10 and 34%±8 reduction vs . vehicle control , respectively; Fig . 4B and D ) . ANA-12 did not alter splenic parasite burden , as was also the case for sunitinib ( Fig . 5E ) . Unlike sunitinib , however , ANA-12 treatment had no significant effect on spleen size ( Fig . 5F ) suggesting that splenomegaly and red pulp vascular remodeling are linked . Similarly , MP frequency was unaltered following ANA-12 treatment ( 1 . 15±0 . 1% vs 1 . 0±0 . 1% for F4/80hiCD11blo MPs , 5 . 83±0 . 5% vs 5 . 82±0 . 3% for F4/80loCD11bhi MPs and 5 . 25±0 . 5% vs 5 . 3±1 . 5% for F4/80loCD11blo MPs , comparing ANA-12 to vehicle ) . Collectively , these data demonstrate that targeting Ntrk2 signaling in vivo selectively inhibits ongoing white pulp neovascularization , without affecting MP infiltration and/or proliferation or splenomegaly and red pulp vascular remodeling ( Fig . 6 ) .
Chronic inflammation results from sustained immune mediated inflammatory responses leading to significant tissue destruction and / or remodeling , but the cells and mechanisms involved in these processes , especially in the context of infectious disease , are not fully understood . Our data provides the first evidence that the neurotrophic receptor Ntrk2 and its ligands play a role in mediating pathological vascular remodeling . First , we show that Ntrk2 is aberrantly expressed on lymphoid tissue ( white pulp ) vasculature in the spleen of mice with chronic infection-associated inflammation . This contrasts with the largely CNS-restricted expression of Ntrk2 expression observed in healthy adult tissues . Second , we have identified a population of splenic F4/80hi CD11blo MPs that express Ntrk2 ligands and possess all the characteristics needed to drive white pulp vascular remodeling . Third , we demonstrate that a selective antagonist of Ntrk2 inhibits ongoing white pulp neovascularization . Angiogenesis and inflammation have long been thought of as co-dependent processes and there is increasing evidence that disruption to vasculature prolongs and intensifies the inflammatory response [35] . It has been shown that MPs are recruited to sites of neoangiogenesis and support neovascularization by releasing angiogenic factors , including VEGF , PDGF , FGF and metalloproteinases [36–38] . Recently , evidence has suggested that neuronal factors such as neurotrophic factors , ephrins , bone morphogenetic proteins ( BMPs ) and their receptors [19 , 39 , 40] , can also play critical roles as angiogenic regulators . A study by Kermani et al highlighted that Bdnf promoted neovascularization in ischemic adult limbs and this correlated with the expression of Ntrk2 on endothelial cells . In agreement with that study , we found enrichment of Bdnf in F4/80hi CD11blo MPs and could directly show that these cells were closely associated with endothelial cells expressing Ntrk2 . The expression of Ntrk2 by splenic endothelial cells was only observed in chronic infection and was not evident in the steady-state . ANA-12 is a low molecular weight heterocyclic compound that was identified using a novel structure-based in silico screening technique as a selective Ntrk2 inhibitor [34] . In mice , ANA-12 has been shown to have anti-depressant and anti-anxiolytic activity [34] and it alters cocaine-mediated behaviour in rats [41] . Ntrk2 is also targeted , less selectively by other RTKi’s in the clinic or in early stage clinical development e . g . Lestaurtinib ( for neuroblastoma;[42] ) and PLX7486 ( pancreatic adenocarcinoma; ClinicalTrials . gov identifier: NCT01804530 ) . If aberrant expression of Ntrk2 is confirmed in other settings of chronic inflammation associated with pathogenic angiogenesis , Ntrk2-selective drugs may provide a new therapeutic option for treatment of these conditions . Importantly , the potential for aberrant expression and function of this neurotrophic receptor , as shown here , also highlights the need to consider inflammation when designing organ-specific therapeutics . Developmentally , tissue macrophages are known to have functions in matrix remodeling [43] , epithelial proliferation and outgrowth [44] , angiogenesis and tissue organization [45 , 46] . It has been suggested that in chronic disease , the developmental activities of macrophages are dysregulated and that these activities contribute to disease pathology [47] . This is evident in some models of cancer where depletion of macrophages or inhibition of macrophage access to the tumor site inhibits tumor growth and angiogenesis[48–51] . Gene expression signatures of tumour-associated macrophages revealed enrichment in genes associated with developmental functions including matrix remodeling , and angiogenesis [52] . Consistent with these findings , F4/80hi CD11blo MPs isolated from the spleen of L . donovani-infected mice revealed enrichment in expression of genes functionally related to cancer and cell survival . These cells displayed angiogenic properties in vitro and given their spatial relationship with emerging blood vessels , it is highly likely that these cells are playing a role in neovascularisation in vivo . Splenic macrophages are themselves a heterogeneous population , of which the immunological phenotype and immunological function of all populations has still not been extensively explored , particularly during chronic inflammation . The CD11c+MHCII+F4/80hi CD11blo MP population identified here that is expanded during L . donovani infection is rare in the spleen of uninfected mice , making it hard to draw direct comparisons with any function of these cells in the steady-state . F4/80+ macrophages found in the white pulp of the spleen were originally described 30 years ago in adult mice to be immediately adjacent to arterioles in the periarteriolar lymphoid sheath [26] , and later studies defined these MPs as having high expression of the F4/80 antigen [27] . Although the low expression of SIGNR1 suggests that these MPs may have some relationship to marginal zone macrophages , cell-tracking studies of MZM in the spleens of infected mice suggest that most bone fide MZM are lost from the spleen during infection [11] . A formal identification of the relationship of the CD11c+MHCII+F4/80hi CD11blo MP population described here to different splenic MP populations present in the steady state would require lineage tracking studies that are beyond the scope of the current manuscript . Finally , the impact of ANA-12 on tissue remodeling was highly selective compared to the impact we have previously observed using a broader spectrum RTKi , sunitinib maleate . Following sunitinib administration , the ongoing remodeling of red and white pulp vasculature is halted and follicular dendritic cell and fibroblastic reticular cell networks are largely restored [4] . In contrast , the impact of ANA-12 administration was selective for white pulp neovascularization . Likewise , whereas red pulp vasculature is remodeled largely by Ly6C+ inflammatory monocytes , white pulp neovascularization appears to reflect the more local behavior of MPs with a more resident-like phenotype . Further studies will be required to determine whether selective blockade of white pulp neovascularization using ANA-12 has any selective effects on immune function than might support its use in combination therapy for disease such as leishmaniasis , where lymphoid tissue remodeling is so prominent . However , in common with broader spectrum RTKi , placental expression of Ntrk2 and the influence of maternal energetic status on placental Ntrk2 expression[53] would likely be an impediment to development of this and similar drugs for the treatment of women of childbearing age . In conclusion , we have shown that selected features of pathogenic angiogenesis induced by chronic infection are mediated through the interaction of a neurotrophic receptor with ligands produced by local MPs . Our data provide new insights into the role of the neurotrophins and their receptors in inflammation , and further characterize the exquisite compartment specific nature of tissue remodeling processes . The role of macrophage expression of Bdnf in other chronic infectious and inflammatory disease settings clearly warrants further study .
All animal care and experimental procedures were regulated under the Animals ( Scientific Procedures ) Act 1986 ( revised under European Directive 2010/63/EU ) and were performed under UK Home Office License ( Ref # PPL 60/4377 ) and with approval from the Animal Procedures and Ethics Committee of the Department of Biology , University of York . Female C57BL6 CD45 . 1 and CD45 . 2 mice were obtained from Charles River UK , housed under specific pathogen-free conditions and used at 6–10 weeks of age . The Ethiopian strain of L . donovani ( LV9 ) was maintained by passage in RAG-2-/- mice . Mice were infected by injecting 3x107 amastigotes i . v . via the lateral tail vein ( total n = 20 for each treatment experiment ) . Animals were then allocated to treatment groups . Sunitinib maleate ( 35mg/kg; Sequoia Research Products Ltd ) ( n = 5 ) or ANA-12 ( 1mg/kg; Sigma ) ( n = 5 ) was administered daily via oral gavage or intraperitoneally for 7d . Vehicle-treated mice received citrate-buffered saline , pH 3 . 5 ( Sm ) or PBS/1% DMSO ( ANA-12 ) ( n = 5 ) . A sample size of 5 for each experiment was required to detect a 30% change in vascularization at 80% power based on previous data [4] . For FITC-dextran uptake , 70 , 000 MW FITC-dextran was injected i . v . at 10mg/ml one hour before animals were sacrificed . Murine endothelial cells ( ATCC ) were grown in DMEM media supplemented with 10% heat inactivated FCS in an atmosphere of 95% air and 5% CO2 at 37°C in plastic flasks . At confluence , the cells were subcultured at a 1:3 ratio and used at passage numbers three through to ten . Cells were tested for mycoplasma contamination using MycolAlert mycoplasma detection kit ( Lonza , Cat No: LT07–418 ) Custom microarrays with 568 gene probes , including 32 control probes were printed in-house . The 536 test probes were chosen to provide insight into the various functions and signalling pathways of macrophages . Total RNA was isolated from splenic macrophages from L . donovani-infected mice ( experimental samples ) and from naive pooled peritoneal macrophages ( reference sample ) . Amplified reference and experimental RNA fluorescently labelled with Cy3 or Cy5 were combined in equal amounts on custom printed microarray slides and hybridised overnight at 42°C . The average experimental: reference ratios were compared by t-tests . Microarray data was analysed using Ingenuity Pathways Analysis . The web-based pathways analysis tool IPA ( Ingenuity Systems , www . ingenuity . com ) was used to identify biological and molecular networks . Knowledge coming from published , peer-reviewed scientific publications is stored in the Ingenuity Pathways Knowledge Base ( IPKB ) and is continuously updated , this web-based tool then allows for the mapping of gene expression data into relevant pathways based on their functional annotation and known molecular interactions . The genes considered to have been differentially regulated to a significant extent when comparing F4/80hiCD11blo cells with conventional peritoneal macrophages were uploaded into IPA along with the gene identifiers and corresponding fold change values . In the network analysis , networks of these genes are then algorithmically generated based on their connectivity . The functional analysis of a network identified the biological functions and/or diseases that were most significant to the genes in the network , and the functional analysis of the entire data set identified the biological functions and/or diseases that were most significant to the data set . Mononuclear cells were prepared from the spleens of C57BL6 L . donovani infected mice . Isolated cells were labelled with F4/80 ( clone: BM8 ) , CD11c ( clone: N418 ) , CD80 ( clone: 16–10A1 ) , CD86 ( clone: GL1 ) , CD40 ( clone: IC10 ) , CD115 ( clone: AFS98 ) , MHCII ( clone: M5/114 ) , CD11b ( clone: M1/70 ) , Gr-1 ( Ly6C/G clone: RB6–8C5 ) all purchased from eBioscience and CD68 ( clone FA-11 ) and CD169 ( clone MOMA-1 ) were purchased from Serotec . Intracellular cytokine staining was performed following surface staining on fixed cell ( 2% paraformaldehyde ) and then permeabilised with 0 . 5% saponin . Cells were analyzed using a cyAn flow cytometer and analyzed using Summit software ( Beckman Coulter ) . Cells were sorted based on forward and side scatter and expression of PE or PE-Cy7 conjugated anti-CD11c , Alexa Fluor 450 anti-MHCII , Alexa Fluor 488 conjugated anti-CD11b and Alexa Fluor 647 anti-F480 on a MoFlo cell sorter ( Beckman Coulter ) . Sorted cells were then used for tube formation assays , analysed for mRNA expression or morphological analysis was carried out with approximately 3000 sorted cells spun onto glass slides , fixed in methanol and stained with Giemsa . Light microscopy was conducted on a Zeiss Axioplan and imaged with an Optronics CCD camera using MagnaFire software ( Optronics ) . Phenotypic analysis of MPs was also carried out on splenic frozen sections . Sections ( 6–10μm ) from FITC-dextran injected , infected or uninfected mice were acetone fixed and labeled with biotin conjugated Gr-1 or biotin conjugated F4/80 and Alexa fluor 647 SIGNR1 ( eBioscience ) . Thicker sections ( 10–30μm ) were used to assess the location of MPs within the spleen . Sections were labeled with a purified rat anti mouse pan endothelial cell antigen antibody ( Biolegend , clone Meca-32 ) and/or a Cy3 labelled monoclonal antibody against α-Smooth Muscle Actin ( α-SMA ) ( Sigma , Clone 1A4 ) were used to identify endothelial cells or vessels . Rat anti-mouse FDC antibody ( FDCMI ) ( BD Pharmingen ) was used to detect follicular dendritic cells . Fluorochrome conjugated goat anti-rat antibody ( Invitrogen ) were used for detection of purified antibodies . Sections were counterstained with DAPI and mounted in Pro-long Gold anti-fade ( Invitrogen ) and visualized using a Carl Zeiss upright LSM META 510 , inverted LSM META 710 confocal microscope or a Carl Zeiss Axio Scan . Z1 digital slide scanner with brightfield and 4-colour fluorescence . Angiogenesis assays were carried out as described previously [54] . Isolated splenocytes from chronically infected mice were first enriched for MPs by plastic adherence for 1hr at 37oC in RPMI ( 10% FCS ) before cell sorting . Proliferating SVEC4–10 cells ( 1 . 5 x104 cells ) were added to sorted populations of MPs or non-adherent cells ( 2 x 104 ) in endothelial basal media-2 , with no FCS , ( EBM-2; Lonza ) and then plated onto growth factor-reduced Cultrex basement membrane ( BME; Trevigen ) coated 96 well plates . Negative controls of SVEC4–10 cells in EBM-2 and positive controls of SVEC4–10 cells in endothelial cell growth media-2 ( EGM-2; Lonza ) containing FCS and growth supplements ( SingleQuot kit; Lonza ) were included . Tube formation was assayed 4 hours after plating . Images were taken using a Zeiss inverted fluorescent microscope with x5 objective . All conditions were set up in each experiment in triplicate and imaged . Images were then uploaded onto http://www . wimasis . com/ for independent blind analysis using WimTube , web-based image analysis software , which quantitatively evaluated the generation of new vessels . Assays were analyzed by WimTube Quantitative Image Analysis . Total RNA was extracted from the sorted cells using the RNeasy RNA isolation kit ( Qiagen ) , and then quantified using a Nanodrop ND-100 . For the PCR array , 200ng of total RNA was reverse transcribed using RT2 First Stand Kit ( Qiagen ) , and cDNA was directly added to PCR Master mix containing SYBR green . The mixtures were then aliquoted into 96-well PCR array plates , to profile the expression of 84 growth factor pathway-related genes using a mouse growth factor signalling pathway RT2 Profiler PCR array ( PAMM-041Z , Qiagen ) according to the manufacturer’s instructions . The array also included 6 housekeeping genes and 3 RNA as internal controls . Arrays were run on an ABI 7300HT qPCR instrument equipped with SDS 2 . 3 software , using RT2 SYBR Green/ROX qPCR master mix ( Qiagen ) . Data analysis was done by the 2-ΔΔCt method on the manufacturer’s Web portal http://www . SABiosciences . com/pcrarraydataanalysis . php Data are expressed values as means ± SEM . Comparison was performed using the unpaired Student’s t-test ( for data following a Gaussian distribution ) and the Mann-Whitney test ( for data that did not assume Gaussian distribution ) . D’Agostino and Pearson omnibus normality test was used to test for Gaussian distribution . A probability of less than 5% ( P < 0 . 05 ) was considered to be statistically significant . All statistical analyses were performed with Prism v5 . 01 ( GraphPad , Inc . ) software . | Visceral leishmaniasis ( VL ) , a globally important parasitic disease responsible for over 40 , 000 deaths p . a . , results in pronounced changes in splenic organisation associated with impaired immune function and persistent parasite infection . We have previously shown that receptor tyrosine kinase ( RTKi ) inhibitors can restore splenic architecture and improve immunocompetence , and that mononuclear phagocytes ( MPs ) are involved in this process . Here , we provide evidence that neurotrophin receptor Ntrk2 ( also known as TrkB ) plays a role in the pathologic remodeling of the spleen that accompanies experimental Leishmania donovani-infection . We show that following infection of mice with L . donovani , Ntrk2 is expressed on splenic endothelial cells that are closely associated with F4/80hiCD11bloCD11c+ macrophages expressing Ntrk2 ligands . Administration of the Ntrk2 antagonist ANA-12 to infected mice significantly inhibited compartment-specific vascular remodeling of the spleen . This study expands our understanding of the pathogenesis of experimental VL and also demonstrates the potential of Ntrk2/Bdnf as targets for treatment of infection-induced vascular remodeling . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | The Neurotrophic Receptor Ntrk2 Directs Lymphoid Tissue Neovascularization during Leishmania donovani Infection |
Organ and tissue formation requires a finely tuned temporal and spatial regulation of differentiation programmes . This is necessary to balance sufficient plasticity to undergo morphogenesis with the acquisition of the mature traits needed for physiological activity . Here we addressed this issue by analysing the deposition of the chitinous extracellular matrix of Drosophila , an essential element of the cuticle ( skin ) and respiratory system ( tracheae ) in this insect . Chitin deposition requires the activity of the chitin synthase Krotzkopf verkehrt ( Kkv ) . Our data demonstrate that this process equally requires the activity of two other genes , namely expansion ( exp ) and rebuf ( reb ) . We found that Exp and Reb have interchangeable functions , and in their absence no chitin is produced , in spite of the presence of Kkv . Conversely , when Kkv and Exp/Reb are co-expressed in the ectoderm , they promote chitin deposition , even in tissues normally devoid of this polysaccharide . Therefore , our results indicate that both functions are not only required but also sufficient to trigger chitin accumulation . We show that this mechanism is highly regulated in time and space , ensuring chitin accumulation in the correct tissues and developmental stages . Accordingly , we observed that unregulated chitin deposition disturbs morphogenesis , thus highlighting the need for tight regulation of this process . In summary , here we identify the genetic programme that triggers the timely and spatially regulated deposition of chitin and thus provide new insights into the extracellular matrix maturation required for physiological activity .
Organ formation requires a finely tuned temporal and spatial control of events . Once cells have acquired the organ cell fate , they undergo a series of consecutive morphogenetic steps until they reach the mature and physiological state , which is then maintained by homeostasis . Many examples in the literature illustrate the failure of organ formation when cells cannot reach their final differentiated state . However , the premature acquisition of mature traits may also lead to deleterious effects . A general feature of the maturation of many organs and tissues is the deposition of an extracellular matrix ( ECM ) . The ECM provides biochemical and structural support , participates in cell adhesion , segregates and protects tissues , regulates cell-cell communication , and senses and transduces mechanical signals [1 , 2] . Insect epithelial cells ( in particular epidermal , tracheal , foregut , and hindgut cells ) deposit a specialised ECM at the end of embryogenesis known as the cuticle , which is renewed during moulting and metamorphosis . The cuticle serves as an exoskeleton and provides protection against dehydration , predators , and pathogens [3] . A major component of the cuticle is the polysaccharide chitin , a polymer of UDP-N-acetylglucosamine ( GlcNAc ) monomers synthesised by the Leloir pathway [4 , 5 , 6 , 7 , 8 , 9] . Chitin is deposited in a highly organised arrangement at the apical surface of epidermal and tracheal cells to form the cuticle [10] . Independently , and before the deposition of this apical tracheal cuticle , a matrix that contains a chitin filament and chitin-binding proteins assembles transiently inside the lumen of the tracheal tubes in Drosophila melanogaster . This chitinous matrix plays a key role in the regulation of tracheal tube size and shape [5 , 11 , 12 , 13 , 14 , 15 , 16] . Chitin is produced by glycosyltransferase chitin synthases ( CHS ) , which polymerise the GlcNAc monomers , thus forming linear polymers [17 , 18] . CHS reside in the apical membrane and form a pore through which the nascent polymerised chitin fibers are extruded . However , the exact mechanism by which CHS polymerise and extrude chitin is not fully understood . Here we report the mechanism involved in the timely and spatially regulated chitin deposition in Drosophila . Our results demonstrate that chitin deposition requires two functions , one exerted by the already known class A chitin synthase Krotzkopf verkehrt ( Kkv ) and a second by two MH2-containing proteins , Expansion and Rebuf ( Exp and Reb ) . Exp/Reb perform the same function and are an absolute requirement for chitin deposition . In their absence , the luminal chitin filament is not assembled and the tracheal and epidermal cuticles are chitin-less , an identical phenotype to that of kkv mutants . In agreement with the absolute requirement of both functions , we found that the pattern of expression of these genes fully accounts for the regulated chitin deposition . When exp/reb genes are over- or misexpressed , they bring about early and increased chitin deposition in places where kkv is normally expressed . Strikingly , the simultaneous misexpression of kkv and exp/reb promotes chitin deposition in ectopic ectodermally-derived tissues . This observation demonstrates that together both activities are not only required but are also sufficient to promote chitin deposition . Our analysis shows that unregulated chitin deposition impairs morphogenesis , thus highlighting the need of a finely tuned control of deposition . At the cellular level , we found that Exp/Reb accumulate strongly at the apical membrane , colocalising with Kkv in an independent manner , and that this subcellular localisation correlates with chitin deposition . Our results suggest that Exp/Reb could be involved in the translocation of the Kkv-synthesized chitin polymers across the membrane and/or their release into the extracellular domain to form microfibrils . In summary , here we unveil a highly regulated developmental mechanism that exquisitely ensures the coordinated acquisition of a mature trait during organ formation . Furthermore , we provide a clear case in which the premature acquisition of a mature trait leads to morphogenetic defects . Finally , our results may also provide new targets for the control of insect plagues through the regulation of chitin deposition , as putative orthologs of these genes are found in the ecdysozoa clade .
In the course of a microarray analysis , we identified CG13188 ( named expansion ( exp ) in a recent independent publication [19] ) as a target of Ttk [20] . BDGP reported expression of this gene in the tracheal system and in the epidermis at late embryonic stages . We raised an antibody against the protein isoform B , which is the one expressed in the embryo [19] . Antibody stainings confirmed the expression and allowed us to refine the temporal pattern in the trachea: Exp protein was first detected at late stage 12-early stage 13 in the Visceral Branch ( VB ) , Transverse Connective ( TC ) , and Lateral Trunk ( LT ) region ( Fig . 1A ) . This pattern extended first to the Dorsal Trunk ( DT ) ( Fig . 1B ) and later to the Dorsal Branches ( DBs ) ( Fig . 1C ) during stage 14–15 . A search for similar genes identified the gene CG13183 ( rebuf , reb ) ( 56% aa similarity ) , which lies next to CG13188 ( Fig . 1G ) . BDGP reported the expression of reb in the tracheal DT , and our in situ hybridisation experiments confirmed this pattern ( Fig . 1F ) . We also raised antibodies against Reb , which confirmed expression exclusively in the DT from early stage 13 , with a stronger accumulation in the DT fusion region ( Fig . 1D , E ) . The spatiotemporal tracheal pattern of Exp and Reb indicated branch-specific regulation . We found that the transcription factor Spalt ( Sal ) , which is first restricted to the dorsal part of the trachea and later to the DT [21] , negatively regulates the initial pattern of Exp in the dorsal part ( Fig . 1H , I ) . In contrast , Sal positively regulates Reb expression in the DT ( Fig . 1J , K ) . The molecular analysis of Exp and Reb proteins identified a single recognisable SMAD/FHA domain ( also called MH2 ) . MH2 domains are typically found in members of the Smad family [22 , 23] , which mediate the TGFβ signal , thus raising the possibility that these two genes participate in the TGFβ pathway . However , our functional characterisation showed that Exp and Reb do not transduce the TGFβ signal in the trachea but perform a different activity ( see below and Beich-Frandsen et al . in preparation ) . A similar conclusion has recently been published [19] . Homology searches with CG13188 and CG13183 revealed the presence of orthologous sequences only in invertebrates , including arthropods and nematodes . No homologous sequence was found in fungi that also produce extracellular chitin . A subset of the retrieved sequences from insects and non-insects was used to generate an evolutionary tree ( Fig . 1L , S1 Text; Beich-Frandsen under review for further details ) . The data suggested that CG13188 represents the insect ancestral protein in Drosophila . In Drosophilids but not other Dipterans like Anopheles gambiae , CG13188 duplicated to give rise to CG13183 . Orthologs of CG13188 were found in distant species like Apis mellifera ( honeybee ) and Tribolium castaneum ( red flour beetle ) . Interestingly , treatment of T . castaneum larvae or pupae with dsRNA against the orthologue TC010825 causes lethality ( http://ibeetle-base . uni-goettingen . de/details/iB_01740 ) , indicating the functional requirement of the gene . We tested the requirements for exp by expressing RNAi lines in the trachea . Tracheal down-regulation of the gene ( around 70% decrease by qPCR , S1A-C Fig . ) produced no detectable defects in the pattern of migration ( Fig . 2A , E ) , organisation ( S1D-E Fig . ) , or diversification of tracheal cells ( S1F-G Fig . ) . However , we detected a clear defect in chitin deposition when we used a marker for chitin ( chitin binding probe , CBP ) . In the wild type , a chitin filament is deposited transiently inside the lumen during the tube expansion period [15] ( Fig . 2B ) . exp tracheal down-regulation prevented luminal chitin accumulation in dorsal and ventral branches ( i . e . the DB , LT and VB ) , remaining only in the DT and part of the TC ( Fig . 2F , compare to 2B ) . In the wild type , several proteins , such as Gasp ( visualised with 2A12 ) and Vermiform ( Verm ) [12 , 16 , 24] , accumulate in the lumen with the chitinous matrix ( Fig . 2C , S1H ) . In animals with reduced exp function , these proteins did not accumulate in the lumen of dorsal and ventral branches , but instead remained in the cytoplasm ( Fig . 2G , S1I ) , further indicating that the chitin filament was not properly formed . In the wild type , tracheal branches become physiologically functional and fill with gas at the end of embryogenesis ( Fig . 2D ) . Only the branches that accumulated chitin ( i . e . DT ) filled with gas in exp down-regulation , possibly causing larval death by asphyxia . We also noted the presence of apical expansions ( S1J-K Fig . ) , as recently reported [19] . Using RNAi , we tested the tracheal requirement for reb . RNAi expression produced no detectable defects in the pattern , migration , or diversification of tracheal cells , or in gas filling ( Fig . 2I-L and not shown ) . Nor did RNAi prevent chitin accumulation in the trachea ( Fig . 2J ) . We tested the effects of the absence of reb by using a combination of deficiencies ( BSC329/BSC879 ) that removes reb and three other genes ( Fig . 1G ) , excluding exp . Mild defects in chitin deposition were detected . We observed that in the wild type , chitin deposition starts in the DT region at stage 13 . By early stage 14 , deposition expands first to the VB and then to the TC and LT . From late stage 14 , chitin accumulates in all the branches , including the DB , and very strongly in the DT ( S2A-H Fig . ) . In the mutants , chitin accumulation in the DT at stage 13 was delayed ( Fig . 2M , N ) , and later chitin levels in the DT were slightly lower than in the wild type ( Fig . 2O-P ) . This result indicates that reb is involved in chitin deposition . The mild defects in chitin deposition in the absence of reb could be due to the presence of exp , which is expressed in all tracheal cells , including the DT . To test this hypothesis , we down-regulated exp in the absence of reb . Embryos showed a normal branching pattern ( S2I Fig . ) but were devoid of the chitin filament ( Fig . 2Q , S2I’ ) . Branches did not fill with air at the end of embryogenesis , and the embryos died ( S2J Fig . ) . For this reason , we named the gene rebuf , which in Catalan means “huff and buff” . These results show that reb is also required for chitin deposition in the DT . When reb is absent , chitin is still deposited in this trunk due to the presence of exp . Similarly , when exp is down-regulated , chitin is deposited in the DT as a result of the presence of reb . Thus , reb is largely dispensable when exp is present but is an absolute requirement for luminal tracheal chitin deposition in the absence of exp . All together , these observations indicate that exp and reb exert redundant functions on chitin deposition . We next studied the effects of the absence of both genes . For this purpose , we used a combination of two deficiencies ( BSC879/ED2247 ) that uncover exp and reb and 12 other genes ( Fig . 1G ) . The transheterozygous embryos did not accumulate the tracheal chitin filament ( Fig . 3A , F ) , and the trachea remained uninflated ( Fig . 3D ) . Chitin filament formation was fully rescued by adding back either exp or reb in the tracheal cells ( Fig . 3B , C ) , thereby indicating that the defects are caused exclusively by the absence of these genes . Thus , we used this deficiency combination as a null condition for both exp and reb ( hereafter exp reb mutants ) . We examined the tracheal defects of exp reb mutants , observing a normal pattern of migration and tracheal cell specification ( Fig . 3A ) ; however , branches completely lacked luminal and apical-cuticular chitin ( Fig . 3A , E-H ) . Staining with the apical determinant Crumbs ( Crb ) and the Adherens Junctions marker E-Cadherin ( E-Cadh ) indicated that the general apicobasal polarity and adhesion , respectively , were not affected ( Fig . 3I-L ) . However , these and other apical markers ( S4A-C Fig . ) also revealed a cystic appearance of the DT , with many dilations and constrictions , and the presence of apical expansions , usually at the joints between branches . E-Cadh stainings also revealed abnormal cell shapes in the DT . In addition to the trachea , chitin is also deposited in the procuticle layer of the epidermal cuticle . Exp is expressed in the epidermis at late embryonic stages ( http://insitu . fruitfly . org ) , thus raising the possibility that it is also required for epidermal chitin deposition . To test this , we down-regulated exp by RNAi using widely expressed Gal4 drivers ( 69B or tubulinGal4 , tubGal4 ) . We detected a clear phenotype attributable to chitin defects: the embryos appeared inflated and were shorter ( probably due to the inflation ) ( S1L-M Fig . ) , and showed defects in the head skeleton . Occasionally , we detected a defect of mouth inversion or rotation ( in 20% of embryos , n = 20 ) ( S1N-O Fig . ) . These defects were more severe in the absence of exp and reb , where we observed a fragmented and granular head skeleton and an inflated and misshapen larval shape ( Fig . 3M-P ) . The defects observed were identical to those of kkv mutants ( see following chapters ) . To further demonstrate a chitin deposition defect , we examined cuticle deposition by transmission electron microscopy ( TEM ) . The tracheal cuticle forms a spiral structure that constitutes the inner wall of the tube . It is composed of a chitinous procuticle , a thin chitin-less epicuticle , and an envelope that covers the whole structure . A thicker procuticle accounts for the spiral ridges , the so-called taenidiae . In exp reb mutants , the tracheal envelope and epicuticle detached from the epithelial cells , while the space between these structures and the epithelial cells were devoid of material ( Fig . 3S , T ) . This phenotype is indistinguishable from the phenotype of kkv mutants [25] . We also studied the cuticle of the integument . The wild-type cuticle consists of three histologically distinct horizontal layers: the outermost envelope , the middle epicuticle , and the innermost procuticle ( Fig . 3Q ) . Chitin microfilaments are arranged in parallel sheets ( laminae ) that are stacked helicoidally within the procuticle , resulting in a crystalline organisation of chitin . The envelope and the epicuticle of exp reb mutants showed a decrease in thickness . In addition , the procuticle in exp reb mutants was amorph ( Fig . 3R ) , resembling the procuticle of kkv mutants ( see below ) , which do not deposit chitin [25 , 26] . Taken together , our results show that exp and reb are required for general chitin deposition in the embryo , which is absolutely required for the tracheal luminal filament and cuticle organisation . We addressed whether the restricted expression of reb was functionally relevant by performing over- and mis-expression experiments . Tracheal expression of GS15050 and LA00773 ( two independent P-UAS lines inserted in front of reb , Fig . 1G ) produced a high overexpression of the gene in all tracheal cells from early stage 11 ( Fig . 4A-C ) . This expression produced clear effects on the deposition of chitin , as it started to accumulate strongly at early stage 13 not only in the DT but in all tracheal branches ( Fig . 4D , G ) . This generalised increase in chitin accumulation was maintained throughout development ( Fig . 4H , I ) . By the end of embryogenesis , we observed a fully penetrant strong phenotype: branches were straight and apparently shrank ( particularly the DT ) ( Fig . 4F , I ) , and the LT displayed lack of cell intercalation ( Fig . 4J ) , as cells remained connected by intercellular junctions [27] . In addition , the trachea did not fill with air , indicating that it was not physiologically functional ( Fig . 4K ) . We attribute these morphogenetic defects to the premature and excessive chitin deposition . This phenotype was completely rescued by simultaneous expression of a rebRNAi construct ( Fig . 4L ) ( but not by overexpression of other UAS-lines , S3E-F Fig . ) . This result validates rebRNAi as a functional line and indicates that the defects produced by GS15050/LA00773 tracheal expression are due exclusively to the overexpression of reb . The overexpression of exp produced similar , although milder defects ( S3A-D Fig . ) . Taken together , these results show that Reb overexpression brings about advanced and increased chitin deposition and that this leads to defects in tracheal tube morphogenesis . These findings thus highlight the relevance of the tight regulation of reb . kkv encodes the chitin synthase required for luminal and cuticular chitin deposition in the trachea and epidermis [5 , 15 , 25] . We therefore compared the phenotypes of kkv and exp reb mutants . The absence of kkv produced similar defects to those caused by the depletion of Exp/Reb . In particular , in kkv and exp reb mutants the cuticular and luminal chitin of the tracheae was absent ( Fig . 5A-C ) and there were dilations and constrictions in the DT and apical expansions in other branches ( Fig . 5D-F , S4A-C ) . Both mutants displayed defects in the luminal accumulation of several markers , such as Gasp , ANFGFP ( used as a secretion marker [28] ) or Verm ( S4D-L Fig . ) , but not of other markers like Pio-pio ( S4M-O Fig . ) . In addition , the defects in cuticle formation were comparable in both mutants ( Fig 5G-L , S4P-R ) . These similarities raised the possibility that these genes act in the same genetic pathway required for chitin deposition . To test this hypothesis , we performed epistatic experiments . The tracheal expression of kkvGFP rescued the defects of a kkv mutant and restored the luminal accumulation of chitin ( S4S Fig . ) . However , this protein was unable to restore chitin accumulation produced by the absence of exp reb or by the down-regulation of exp ( Fig . 5M , N , S4T ) . On the other hand , while exp and reb rescued luminal chitin in exp reb mutants ( Fig . 3B , C ) , they did not restore chitin accumulation in a kkv mutant background ( Fig . 5O ) . These results indicate that kkv is not required for exp/reb expression and that exp/reb are not required for kkv expression . But most importantly , these epistatic experiments also showed that both functions are required in parallel for chitin deposition and that the presence of one cannot substitute the other when this is absent . To find evidence of a functional interaction between Kkv and Exp/Reb , we analysed the subcellular localisation of these proteins . Exp immunostainings showed that the protein first accumulated in the cytoplasm of tracheal cells ( in the temporal pattern described in Fig . 1A-C ) , but from stage 13–14 it also accumulated apically , lining the lumen , and later this apical accumulation was conspicuous ( Fig . 5P ) . Reb protein was also enriched apically in the DT ( Fig . 1E , 5Q ) and in the rest of branches when misexpressed ( Fig . 4B , C , G , H ) . To analyse the accumulation of Kkv , we expressed a UASkkvGFP line in the trachea . We observed cytoplasmic accumulation of KkvGFP , but also a clear apical accumulation . KkvGFP colocalised with Exp and Reb in the apical region ( Fig . 5P , Q ) . As described above , the misexpression of reb brings about early and increased chitin accumulation in the trachea . In contrast , we found that kkvGFP overexpression does not induce , increase , or advance chitin deposition in the trachea or other tissues ( S5A-C Fig . ) . We simultaneously overexpressed the two genes in the trachea . Chitin accumulated even earlier ( already at stage 11 ) and more strongly than when Reb was overexpressed alone ( Fig . 6A , B ) . The increased chitin accumulation correlated perfectly with dramatic morphogenetic defects , as the tracheal branches became short , very straight , and generally unfused ( Fig . 6C-D ) . Staining with E-Cadh revealed defects in tracheal cell shape organisation and cell intercalation ( DB , LT and VB , which in the wild type form autocellular junctions [27] , remained non-intercalated with intercellular junctions ) ( Fig . 6D ) . These results show that together these two genes have the capacity to trigger chitin deposition and that the early and excess deposition of chitin blocks tracheal morphogenesis . More strikingly , we found that concomitant expression of kkvGFP and reb triggers chitin deposition in ectodermally-derived tissues that normally do not hold this polysaccharide . For instance , when using the breathlessGal4 ( btlGal4 ) driver , chitin deposited in tissues in which it does not normally accumulate , such as the midline and the proventriculus , sites in which the driver is expressed ( Fig . 6E-F ) . Chitin also highly accumulated in salivary glands ( SGs ) when using a forkheadGal4 ( fkhGal4 ) line . SGs accumulating luminal chitin did not properly invaginate or undergo extension ( Fig . 6H-I ) . Expression in the dorsal part of imaginal discs ( apterous ( ap ) domain ) promoted apical accumulation of chitin between the peripodial membrane and the epithelia ( Fig . 6G ) , which led to strong morphogenetic defects in the adult wing and notum ( S5D-E Fig . ) . Similar but milder effects were observed when expressing kkvGFP and exp ( S3G-H Fig . ) . These results show that the simultaneous activities of Kkv and Exp/Reb are sufficient to promote chitin deposition in the ectoderm , even in tissues in which chitin does not normally accumulate , thus blocking morphogenesis . We evaluated the capacity of kkv and reb to trigger chitin synthesis when concomitantly misexpressed in mesodermal or endodermal tissues . We found no extracellular chitin deposited in those tissues ( Fig . 6J-L ) . A detailed inspection indicated that , in contrast to what happens in ectodermal tissues , Reb did not localise apically in these non-ectodermal tissues and instead accumulated in the cytoplasm . KkvGFP was enriched in the entire cortical region . This result indicates that the apical accumulation of Reb requires factors present only in ectodermal tissues or the general ectodermal apicobasal polarity . It also indicates a clear correlation between the subcellular localisation of these proteins and chitin deposition . We also evaluated the activity of kkv and reb/exp in cell culture in vitro assays ( S6 Fig . ) . We transfected S2 cells and found that the presence of KkvGFP leads to the formation of small chitin-containing particles intracellularly . The pattern of these particles was not changed when the cells transfected with kkvGFP were also cotransfected with exp , reb , or exp+reb , and we never detected fibrilar chitin deposited extracellularly . Interestingly , we found that Kkv and Exp/Reb did not localise in the cell membrane region . The results in cell culture are in line with our in vivo experiments in non-ectodermal tissues , and further support a correlation between the subcellular localisation of these proteins and their activity in chitin deposition . We examined the participation of Exp/Reb in chitin deposition . For this purpose , we first analysed whether the sugar precursors ( UDP-GlcNAc ) were present in the exp reb mutants . The lectin wheat germ agglutinin ( WGA ) recognises the terminal GlcNAc residues of chitin [29] . In the wild-type trachea , WGA strongly accumulated in the lumen and apical surface ( Fig . 6M ) . In mutants , WGA was largely absent from the lumen but still accumulated in the apical region ( Fig . 6N ) . This observation strongly suggests that , as occurs in kkv mutants [4] , exp/reb are not required for the synthesis or apical accumulation of GlcNAcs . Exp/Reb and KkvGFP colocalisation results suggested an interdependent apical accumulation . However , KkvGFP still accumulated apically in the absence of exp reb ( Fig . 6O-P ) , and Exp and Reb were also apical in kkv mutants ( Fig . 6Q-T ) . These findings indicate that Kkv and Exp/Reb accumulate apically , lining the lumen in an independent manner . The ectopic expression of kkv and reb in the SGs ( which comprise large , single-layered epithelial cells that form a tube ) gave some clues about the possible mechanism . As indicated , when we simultaneously expressed kkvGFP and reb , fibrilar chitin was deposited extracellularly in the lumen of the SGs , while Kkv and Reb colocalised in the apical membrane ( Fig . 6U ) . In contrast , when we expressed only kkvGFP at high levels in the SGs , chitin was not deposited in the lumen . However , we observed the presence of small chitin particles ( marked with CBP ) highly enriched in the apical and/or membrane region ( Fig . 6V , S5F-G ) . These chitin particles did not colocalise with cytoplasmic kkvGFP vesicle-like particles ( S5G Fig . ) . Occasionally , we also detected the presence of chitin particles in the apical region of tracheal cells overexpressing kkvGFP but mutant for exp/reb ( S5H Fig . ) . It has been proposed that chitin deposition requires first the formation of polymers , their translocation across the membrane to the extracellular space , and finally their assembly into microfibrils [17 , 18 , 30] . In light of this , a possible interpretation of our results is that , in the presence of Kkv , small chitin polymers are synthesized but cannot be extruded extracellularly and/or further organised into microfibrils without the activity of Exp/Reb .
Both Exp and Reb are required for chitin deposition . While no chitin is deposited in the trachea in the absence of both genes , the presence of one or the other can fully rescue this phenotype . Thus these two genes can perform the same function and they are interchangeable . The pattern of expression and functional requirements of each of these genes in normal conditions illustrate an elegant mechanism of activity . The phenotype of exp down-regulation ( in all branches except the DT ) is complementary to the pattern of reb ( expressed only in the DT from stage 13 ) . Our results show that Reb allows chitin deposition in the DT in the absence of Exp . We also reveal that the removal of reb causes only a delay in DT chitin deposition , due to the presence of Exp in the DT from stage 14 onwards . Thus exp and reb are redundant when expressed in the same tissue . What is the functional relevance of having two genes with interchangeable roles in chitin deposition and that show partially overlapping expression ? An unequivocally answer is difficult , but several lines of evidence are worth considering . On the one hand , we found that reb expression is restricted to the DT during embryogenesis , suggesting that it is required only to ensure early and strong chitin deposition in this region . This is consistent with our results , showing that in the absence of reb , DT chitin deposition is delayed and decreased , while the rest of branches and the cuticle are not affected . On the other hand our findings suggest that Reb is more efficient than Exp in performing the same function . Indeed , strong over or misexpression of reb alone or in combination with kkv generated stronger effects than when strongly overexpressing exp . In addition , in normal conditions , chitin is first deposited in branches that express reb ( i . e . the DT ) , in spite the fact that exp is also expressed at the same time in other branches ( see Fig . 7A ) . In contrast to reb , exp is expressed in all tracheal cells and in the rest of chitin-synthesising tissues and is required for general chitin deposition . Hence , we hypothesise that Exp is more general but less efficient at promoting chitin deposition , while Reb is more restricted but more efficient . Various explanations could be given regarding the differences in efficiency , ranging from differences at the functional level to differences in the subcellular localisation ( we note that Reb apical localisation is more conspicuous than that of Exp , both in normal and overexpression conditions ) . In summary , Reb appears to be required only when a rapid and strong accumulation of chitin is needed . The deposition of chitin first in the DT region may represent an advantage , particularly considering that the DT does not undergo cell intercalation [27] , a process that is impaired when excess chitin accumulates . Thus , the earlier and stronger accumulation of chitin in the DT may be a safety mechanism which serves to prevent cell intercalation , thereby allowing normal morphogenesis . Chitin deposition depends on CHS enzymes present in all chitin-synthesising organisms . In spite of the relevance of CHSs , their exact mechanism of activity remains obscure [17 , 18] . Kkv encodes the epidermal and tracheal CHS in Drosophila [5 , 15 , 25] . Kkv is transcriptionally expressed in tissues that normally deposit chitin , i . e . the tracheal system and the epidermis . In the trachea , Kkv is present in all tracheal cells from late stage 11-early stage 12 until late embryonic stages [5] and BDGP in situ homepage ( http://insitu . fruitfly . org ) . kkv is an early target of Trachealess ( Trh ) activity [37] . In spite of the general and early kkv tracheal expression , the deposition of chitin does not start until stage 13 , when it occurs in a spatio-temporal restricted manner ( Fig . 7B , S2A-H ) . This observation indicates that Kkv activity is subjected to post-transcriptional regulation . We conclude that the regulated spatiotemporal expression of exp and reb perfectly accounts for this post-transcriptional regulation of Kkv activity . This conclusion is based on two findings . First , Sal-dependent exp/reb expression in the trachea is consistent with the temporal pattern of chitin deposition by Kkv , whose expression is independent of Sal; and second , Kkv is only sufficient to promote chitin deposition ( both in chitin-synthesising and chitinless tissues ) in the presence of Exp/Reb . Therefore , chitin is deposited only in those cells that concomitantly express kkv and exp/reb ( Fig . 7A-B ) . The unexpected incapacity of Kkv to trigger chitin deposition on its own may reflect the need of a mechanism that ensures chitin accumulation only when and where it is required . It has been proposed that chitin deposition by the midgut-specific chitin synthase-2 ( CHS2 ) in the midgut of the tobacco hornworm Manudica sexta depends on a chymotrypsin-like protease . It has been hypothesized that the proteolytic regulation of CHS-2 activity , which promotes the synthesis of chitin needed to protect the midgut epithelium against damage , relies on the nutritional state [38 , 39] . As expected , the situation is different in the epidermis and the trachea . If Kkv were sufficient to promote chitin deposition , any temporal or spatial misregulation of its expression would lead to ectopic chitin deposition , which we show causes serious morphogenetic defects . Interestingly , we found that kkv and reb/exp expression patterns are regulated in distinct manners ( at least in the trachea ) ( Fig . 7B ) . This differential regulation may represent a way to restrict chitin accumulation in a coordinated manner , providing a complex control mechanism for final organ maturation . What is the molecular mechanism of chitin deposition ? Kkv and Exp/Reb accumulate in the apical membrane in tissues that normally deposit chitin , which have an ectodermal origin . Our experiments indicate that when Kkv and Reb are present , but are not apical ( e . g . when misexpressed in the endoderm , or in S2 cells ) , no extracellular chitin deposition is detected . These observations show that only ectodermally-derived tissues contain the molecular machinery or the adequate apicobasal polarity to properly localise or maintain these proteins apically localised . In addition , the correlation between the subcellular localisation of these proteins and extracellular chitin deposition strongly suggests that the function of Kkv and Exp/Reb is required in the apical membrane . Thus , the subcellular localisation of these proteins could represent an extra level in the regulation of chitin deposition . In line with this hypothesis , we speculate that the inability of Reb and Kkv to accumulate apically in the membrane prevents extracellular chitin deposition in non-ectodermal tissues . What is the role of Exp/Reb in chitin deposition ? Chitin formation has been proposed to follow three steps . In the first step , the catalytic domain of the CHS that faces the cytoplasm forms polymers . In the second , the nascent polymers are translocated through the membrane to the extracellular space . Finally , in the third step the polymers spontaneously assemble to form crystalline microfibrils [17 , 18 , 30] . When we strongly expressed kkv in the absence of exp/reb , we observed an apical/membrane enrichment of small chitin-containing particles that appear to be unable to be deposited extracellularly or to form microfibrils . When we added exp/reb to this background , chitin was deposited extracellularly in a fibrillar organisation . These results suggest that Exp/Reb may participate in the steps of polymer translocation across the membrane and/or in microfibril formation . Exp/Reb do not hold canonical transmembrane domains , suggesting that they could interact or recruit , through their MH2 domain , other proteins directly involved in the translocation . It has been proposed that the carboxy-terminal region of CHS is involved in membrane translocation of polymers [30] , thus raising the possibility that Exp/Reb interact with this domain . Alternatively , Exp/Reb may be required to directly or indirectly ( by promoting the formation of a complex ) activate CHS posttranscriptionally . Several postranscriptional modifications have been proposed for CHS , such as oligomerisation , phosphorylation , proteolytic cleavage ( of a soluble factor that activates chitin synthesis ) , and the release of the nascent polymers to form microfibrils [17 , 18 , 38 , 40] . Exp/Reb may participate in these modifications . In summary , altogether our results indicate that chitin deposition needs to be highly regulated in time and space , and that the finely tuned regulation of chitin deposition relies on a spatiotemporal control of the activity of Exp/Reb controlling Kkv-dependent chitin deposition . Thus , here we have identified the genetic programme required for timely and
The fly strains used are described in FlyBase: Df ( 2R ) BSC329 , Df ( 2R ) ED2247 , Df ( 2R ) BSC879 , Df ( 2L ) 32FP-5 ( removes sal-m and sal-r ) , kkvIB22 , and kkv63-20 . The transgenes used were: P{TRiP . HMS01445}attP2 , P{TRiP . HMS01444}attP2 , P{KK111583}VIE-260B , P{GD7952}v17126 , P ( GSV6 ) GS15050 , P ( Mae-UAS . 6 . 11 ) LA00773 , and UAS-ANFGFP . For overexpression experiments , we used the following Gal4 drivers: btlGal4 ( in all tracheal cells ) , fkhGal4 ( in salivary glands ) , 69B ( generally epidermal expression ) , tubGal4 ( general expression ) , apTomato-Gal4; Gal80ts ( in the dorsal part of the imaginal wing disc ) , 48YGal4 ( in the intestinal tract ) and twiGal4 ( in the mesoderm ) . To maximise the expression of the transgenes crosses were kept at 29°C . To visualise the “tracheal pattern” , the embryos carrying btlGal4 UAS-srcGFP ( cell membrane staining ) were stained for GFP . The bltGal4 in this combination also drives other UAS transgenes Transgenic flies carrying UAS-CG13188 , UAS-CG13188-HA , UAS-CG13183 , and UAS-kkvGFP were generated ( see S1 Text for details ) Fully developed embryos were dechorionated in bleach , devitellinized by shaking in 100% methanol , and incubated over night at 65°C in Hoyer’s medium mixed with lactic acid ( 1:1 ) . Embryos were analysed by light microscopy using a Nikon Eclipse 80i microscope . To evaluate gas filling in the tracheae , we followed the procedure described in [41] For ultrastructural analyses by transmission electron microscopy ( TEM ) , wild-type and exp reb mutant embryos were immobilised by high-pressure freezing , fixed by freeze substitution , embedded in Epon , and sectioned as described previously [26] . Images were taken on a CM10 electron microscope . We followed standard protocols for immunostainings and in situ hybridisations . Embryos were staged as described [42] . Imaginal discs were obtained by dissecting third instar larvae . The following primary antibodies and dilutions were used: mouse anti-2A12 ( recognises Gasp , 1:10 ) , rat anti-DEcad ( 1:100 ) , and mouse anti-Crb ( 1:20 ) from Developmental Studies Hybridoma Bank , DSHB; rbb anti-Verm ( 1:300 ) from S . Luschnig; goat anti-GFP ( 1:600 ) Molecular Probes and Roche; ck anti-ßGal ( 1:500 ) abCAM; GP anti-Uif ( 1:400 ) from R . Ward; and rbb anti-Pio ( 1:100 ) from M . Affolter . CBP ( chitin-binding probe ) conjugated with Cy3 , Cy2 and Cy5 was used at 1:300 ( generated by N . Martin ) . WGA conjugated with Alexa-555 , -488 , and -647 was used at 1:300 ( Molecular Probes ) . Cy3- , Cy2- and Cy5-conjugated secondary antibodies ( Jackson ImmunoResearch ) were used at 1:300 . A reb riboprobe was generated using the following primers: Forward: 5′- AACTGTGCCTCGGCGCTAGTC Reverse: 5′- AGCAGTCGAAACACGCAGCTT Confocal images were acquired with a Leica TCS-SPE system . Images were post-processed with ImageJ and Adobe Photoshop and assembled using Adobe Illustrator . Polyclonal antibodies against CG13188 and CG13183 were generated ( see S1 Text for details ) . Purified recombinant proteins were used as antigens to immunise rats and rabbits following standard protocols . Drosophila S2 cells were transiently transfected using Cellfectin II Reagent ( Invitrogen ) . Cells were cultured in Schneider’s Insect Medium ( Sigma ) enriched with 10% of FBS ( Fetal Bovine Serum , Gibco ) and were used for immunostaining assays after 3 days of expression . Reb and Exp cDNAs were obtained by PCR from RE66796 and RE28239 clones , respectively ( DGRC , Bloomington , IN ) . Kkv ( with or without GFP ) cDNA was obtained by PCR from UAS-GFPkkv flies ( B . Moussian ) . In experiments without KkvGFP , DNA constructs were co-transfected with pAc5 . 1-GFP ( a gift from J . Bernués ) to visualise expressing cells . The fragments were cloned into pMT or pAC5 . 1/V5-His A ( Invitrogen ) expression vectors . Total RNA from control and exp RNAi embryos at stages 11 to 16 was used to synthesise complementary cDNA by random hexamer priming ( RevertAid H Minus First Strand cDNA Synthesis FERMENTAS Kit ) . A LightCycler 480 Real-Time PCR System and the SYBR Green PCR Master Mix ( Roche ) were used to amplify cDNAs . CG13167 , a mitochondrial ATPase with stable expression , was used to normalise relative quantities . Samples were analysed using the LightCycler 480 Real-Time PCR System software ( Roche ) ( see S1 Text for details ) . | In this work we studied the maturation of the extracellular matrix during Drosophila embryogenesis . Drosophila deposit a chitin-rich extracellular matrix with key physiological functions , such as the control of organ size and shape , and cuticle formation . Chitin synthesis depends on chitin synthases , and in Drosophila the gene krotzkopf verkehrt ( kkv ) encodes the main enzyme of this family . Our observations indicate that Kkv alone is not sufficient to induce chitin formation . We have identified another function ( which is exerted by the activity of two genes encoding MH2-domain proteins ) that are equally required for chitin deposition . The most striking result of our analysis is that the presence of Kkv and the newly identified function is sufficient to trigger chitin deposition in ectodermally-derived tissues , even if they are normally devoid of this polysaccharide . Importantly , we also demonstrate that unregulated chitin deposition ( absent , advanced , or ectopic ) leads to severe defects in morphogenesis . We show that the temporal and spatial pattern of kkv and the other two genes perfectly recapitulates the deposition of chitin , thereby unveiling a highly co-ordinated mechanism for the acquisition of mature traits . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Deciphering the Genetic Programme Triggering Timely and Spatially-Regulated Chitin Deposition |
Genetic redundancy , whereby two genes carry out seemingly overlapping functions , may in large part be attributable to the intricacy and robustness of genetic networks that control many developmental processes . We have previously described a complex set of genetic interactions underlying foregut development in the nematode Caenorhabditis elegans . Specifically , LIN-35/Rb , a tumor suppressor ortholog , in conjunction with UBC-18–ARI-1 , a conserved E2/E3 complex , and PHA-1 , a novel protein , coordinately regulates an early step of pharyngeal morphogenesis involving cellular re-orientation . Functional redundancy is indicated by the observation that lin-35; ubc-18 double mutants , as well as certain allelic combinations of pha-1 with either lin-35 or ubc-18 , display defects in pharyngeal development , whereas single mutants do not . Using a combination of genetic and molecular analyses , we show that sup-35 , a strong recessive suppressor of pha-1–associated lethality , also reverts the synthetic lethality of lin-35; ubc-18 , lin-35; pha-1 , and ubc-18 pha-1 double mutants . SUP-35 , which contains C2H2-type Zn-finger domains as well as a conserved RMD-like motif , showed a dynamic pattern of subcellular localization during embryogenesis . We find that mutations in sup-35 specifically suppress hypomorphic alleles of pha-1 and that SUP-35 , acting genetically upstream of SUP-36 and SUP-37 , negatively regulates pha-1 transcription . We further demonstrate that LIN-35 , a transcriptional repressor , and UBC-18–ARI-1 , a complex involved in ubiquitin-mediated proteolysis , negatively regulate SUP-35 abundance through distinct mechanisms . We also show that HCF-1 , a C . elegans homolog of host cell factor 1 , functionally antagonizes LIN-35 in the regulation of sup-35 . Our cumulative findings piece together the components of a novel regulatory network that includes LIN-35/Rb , which functions to control organ morphogenesis . Our results also shed light on general mechanisms that may underlie developmental genetic redundancies as well as principles that may govern complex disease traits .
Genetic redundancy describes the phenomenon in which the combined inactivation of two distinct genes produces a phenotype that is not observed in either single mutant . One of the current challenges facing geneticists and developmental biologists alike is to understand the underlying bases of genetic redundancy at the molecular level . This may in many cases prove to be a difficult undertaking given the complexity of regulatory networks and the many difficulties associated with establishing clear connections between seemingly disparate genes . Nonetheless , redundancy is an issue of great biological importance , as evidenced in C . elegans , where most genes fail to show obvious or highly penetrant phenotypes following inhibition or inactivation [1]–[3] . To date , the most intensively studied case of genetic redundancy in C . elegans involves the Synthetic Multivulval ( SynMuv ) genes ( for a review , see [4] . The SynMuv genes can in most cases be divided into two principal non-overlapping groups , termed class A and class B [5] . Inhibition of individual class A or class B genes does not typically alter normal patterns of vulval cell induction in hermaphrodites . In contrast , the combined loss in activity of any class A–class B gene pair leads to the ectopic induction of vulval tissue ( the Muv phenotype ) . In addition , a class C group of SynMuv genes has recently been identified; mutations in class C genes are synthetic with mutations in both class A and class B SynMuv genes [6] . Extensive work has shed considerable light on the role of SynMuv genes in vulval development . Namely , most class A and B genes act within the hypodermis , a multi-nucleate epidermal tissue that lies adjacent to the developing vulval precursor cells ( VPCs ) , where they redundantly inhibit the expression of the EGF-like ligand , LIN-3 [7] . Secreted LIN-3 induces vulval cell development through activation of a conserved EGFR–Ras–Map kinase pathway in the VPCs [8] . Thus , in the absence of both class A and class B SynMuv activity , abnormally high levels of LIN-3 , secreted by the hypodermis , leads to the hyperinduction of vulval cell fates . Based on studies in C . elegans , Drosophila , and mammals , the large majority of proteins encoded by the class B SynMuv gene family function within a conserved set of structurally related transcriptional repressor complexes that include DRM ( Dp , Rb and MuvB ) and NuRD ( nucleosome remodeling and histone deacetylase; reviewed by [4] , [9] . Among the components that are common to these complexes are LIN-35 , the sole C . elegans Retinoblastoma protein ( pRb ) family ortholog , and EFL-1 , a member of the E2F family of transcription factors [10]–[12] . Similar to its role in other systems , LIN-35 acts in large part to mediate the transcriptional repression of E2F target genes [13] . Nevertheless , the precise means by which class A and B SynMuv genes influence the expression of LIN-3 in the hypodermis is currently unclear . Furthermore , the precise molecular functions of class A genes are presently unknown , although a role in transcription has been proposed [4] . We have previously described a forward genetic screen for identifying mutations that show strong synthetic genetic interactions in conjunction with the loss of lin-35 [14] . This and other work has led to the identification of a diverse array of redundant functions for LIN-35 including roles in cell cycle control [14] , [15] , cell fate specification [16] , asymmetric cell division [17] , larval growth [18] , [19] , fertility [16] , [20] , organogenesis [20] , [21] , and organ function [22] . In addition , LIN-35 , along with a number of other class B SynMuv genes , has been shown to function non-redundantly in the control of transgene expression [23] , RNAi [24] , [25] , germline and somatic sex-linked apoptosis [26] , [27] , ribosome biogenesis [28] , and the somatic silencing of germline gene expression [13] , [25] . In our current work , we have sought to understand the mechanistic basis for the synthetic genetic interactions observed between lin-35 and two mutations previously identified by our screen , ubc-18 and pha-1 [21] , [29] . Both lin-35; ubc-18 and lin-35; pha-1 double mutants arrest predominantly as L1 larvae and display severe defects in pharyngeal morphogenesis . Furthermore , ubc-18 pha-1 double mutants are also synthetically lethal , indicating that the functions of these three genes are interconnected [29] . Notably , the genetic interactions between pha-1 and lin-35 or ubc-18 can be observed only under conditions in which pha-1 activity is weakly compromised . This is because strong loss-of-function mutations in pha-1 are themselves lethal , and arrested pha-1 mutant animals display defects in pharyngeal and body morphogenesis [30] . Through an analysis of the suppressor mutation sup-35 , we demonstrate that SUP-35 acts as an inhibitor of pha-1 transcription . Furthermore , we show that LIN-35 and UBC-18 act through distinct mechanisms to negatively regulate SUP-35 expression . Thus , the simultaneous loss of lin-35 and ubc-18 leads to increased levels of SUP-35 , which in turn trigger a reduction in the levels of PHA-1 . These findings provide a straightforward explanation for the observed genetic interactions between these genes and more generally provide further insight into the nature of mechanisms that can underlie genetic redundancies .
As described in the Introduction , lin-35 mutations are strongly synthetic with hypomorphic mutations that affect the pha-1 locus , leading to strong pharyngeal morphogenesis defects [29] . In addition , recessive mutations in three genetic loci ( sup-35 , sup-36 , and sup-37 ) were demonstrated to strongly suppress the embryonic- and larval-lethal phenotype of strong loss-of-function pha-1 mutants [31] . We have previously shown that mutations in sup-36 and sup-37 efficiently suppress the synthetic lethality of lin-35; pha-1 and lin-35; ubc-18 double mutants [29] . As described below , these and other related synthetic genotypes were also suppressed by mutations in sup-35 . Thus , to learn more about the interplay between these various factors and their roles in pharyngeal development , we sought to identify the sup-35 locus . Previous mapping data had placed sup-35 on LGIII , ∼0 . 1 cM to the left of the pha-1 locus [31] . To identify the gene encoding sup-35 , we carried out RNAi feeding of 384 clones corresponding to genes in the region proximal to pha-1 . Two clones , which target the highly related genes Y48A6C . 1 and Y48A6C . 3 , were identified that strongly suppress the embryonic lethality of pha-1 ( e2123ts ) mutants ( referred to hereafter as pha-1 ( ts ) ) at the non-permissive temperature of 25°C ( Table 1 ) . These RNAi clones also suppress the less severe L1 larval-arrest phenotype of pha-1 ( ts ) mutants at intermediate temperature of 20°C ( data not shown ) . Because Y48A6C . 1 and Y48A6C . 3 share extensive sequence homology ( an 878-bp segment present in both genes is 99% identical ) , each RNAi construct is expected to inhibit both gene products through off-target effects; no additional off targets for these RNAi constructs are predicted . These results suggest that sup-35 may be encoded by either Y48A6C . 1 or Y48A6C . 3 . However , an additional RNAi construct that is expected to target Y48A6C . 1 , but not Y48A6C . 3 , failed to suppress pha-1 ( ts ) mutants at 25°C , suggesting that Y48A6C . 3 is the relevant locus ( data not shown ) . Additional support for Y48A6C . 3 as the affected locus was provided by sequencing both Y48A6C . 1 and Y48A6C . 3 in sup-35 ( e2223 ) pha-1 ( ts ) double mutants . We detected a T-to-A transversion at nucleotide position 19 of the Y48A6C . 3 open reading frame , resulting in the conversion of a cysteine to a serine at amino acid position seven . In contrast , we failed to identify any differences in the Y48A6C . 1 locus between the published wild-type ( N2 ) and sup-35 ( e2223 ) mutant sequences . Furthermore , we identified sequence alterations in Y48A6C . 3 in five previously isolated alleles of sup-35 [31] as well as in 14 additional alleles identified by our laboratory . A summary of our sequence analysis is shown in Figure 1A . Two alleles , fd35 and fd42 , contained a single nucleotide insertion and deletion , respectively , leading to frameshifts within exon 5 of Y48A6C . 3 . All other sup-35 alleles contained either large deletions or insertions within Y48A6C . 3 , and are presumed to be null alleles ( Figure 1A and data not shown ) . Taken together , our findings strongly indicate that sup-35 is encoded by Y48A6C . 3 . Furthermore , given that the majority of these alleles were identified as spontaneous revertants ( [31] and this work ) , the sup-35 genomic region would appear to be unusually unstable and subject to recombination events that lead to gross alterations of the locus . Based on the WormBase predicted gene model , as well as an ORFeome-generated full-length cDNA , sup-35 encodes a 332-amino-acid protein containing two N-terminal C2H2-type Zn-finger domains along with two tetratrico peptide repeats ( TPR ) at its C terminus . The molecular lesion identified in sup-35 ( e2223 ) is predicted to disrupt the first Zn finger , indicating that this domain is likely to be essential for SUP-35 function . The presence of the Zn-finger motifs suggests a potential role for SUP-35 in transcriptional regulation . Alternatively , the Zn-fingers may be involved in protein-RNA , protein-protein , or protein-lipid interactions . Interestingly , other than its close paralog Y48A6C . 1 , SUP-35 is most similar to an evolutionarily conserved family of RMD ( regulators of microtubule dynamics ) proteins ( Figure 1B; [32] . Of the six RMD family members in C . elegans , SUP-35 is most similar to RMD-2; the C-terminal 215 amino acids of SUP-35 are 52% identical to a corresponding region in RMD-2 , which in turn shares greater homology with SUP-35 than with other C . elegans RMD proteins ( Figure 1B and data not shown ) . Interestingly , RMD-2 , along with RMD-1 and RMD-3 , can physically associate with microtubles in vitro [32] . Consistent with the RMD-like domain of SUP-35 having an important functional role is the observation that two alleles of sup-35 , fd35 and fd42 , may specifically affect this region of the protein . Nevertheless , SUP-35 differs from other C . elegans RMD family members , as well as RMD proteins in other organisms , by the presence of its unique N-terminal Zn-finger domains . The TPR domains in SUP-35 suggest a possible role in protein-protein interactions [33] . To assess the pattern of SUP-35 expression during development , multiple independent transgenic strains were generated expressing full-length SUP-35 fused to GFP under the control of the native sup-35 promoter/enhancer region ( also see Materials and Methods ) . For reasons described below , the SUP-35::GFP expression analysis was performed in sup-36 and sup-37 mutant backgrounds , both of which gave identical results . SUP-35::GFP expression was first observed in embryos at around the 50- to 100-cell stage . Expression of SUP-35::GFP was ubiquitous throughout the proliferative phase of embryogenesis and was strongly enriched in the cytoplasm ( Figure 2A and 2B ) . Commensurate with the onset of visible morphogenesis ( ∼400 minutes ) , SUP-35::GFP localization became pronounced in nuclei , most notably in cells comprising the pharyngeal primordium ( Figure 2C and 2D ) . Pharyngeal cells also maintained nuclear SUP-35::GFP expression throughout larval stages and into adulthood ( data not shown ) . In addition , weaker SUP-35::GFP could be detected in the nuclei of several non-pharyngeal cells in the posterior . Mutations in either sup-36 or sup-37 are capable of suppressing all pair-wise combinations of mutations in lin-35 , ubc-18 , and pha-1 [29] . Consistent with this , the same constellation of synthetic-lethal mutations was efficiently suppressed by loss of sup-35 ( Table 1 ) . This includes suppression by the canonical allele of sup-35 , e2223; a consortium-generated deletion allele , tm1810; and by sup-35 ( RNAi ) . Suppression by sup-35 ( tm1810 ) also further confirms the molecular identity of this locus . Previous studies from our laboratory have implicated the RING finger–domain protein , ARI-1 , as the primary co-partner of UBC-18 in the regulation of pharyngeal development [34] . Consistent with this , a consortium-generated deletion allele of ari-1 , tm2549 , showed strong synthetic interactions with pha-1 ( ts ) , and this lethality was suppressed by sup-35 ( RNAi ) ( Table 1 ) . Taken together , these findings suggest that sup-35 functions within a regulatory network that includes pha-1 , lin-35 , ubc-18 , and ari-1 to control pharyngeal development . Extragenic suppression in C . elegans arises through a number of distinct mechanisms [35] . Such mechanisms can , in some cases , be distinguished based on whether or not suppression occurs in the presence of a null allele . For this reason , we first sought to determine whether the strongest characterized allele of pha-1 , e2123ts , retains activity at the non-permissive temperature of 25°C; e2123ts is a missense mutation that leads to a conversion of cysteine to tyrosine at amino acid position 169 of PHA-1 [36] . We thus generated high-copy extrachromosomal arrays carrying the pha-1 ( ts ) variant in mutant animals that were already chromosomally homozygous for the pha-1 ( ts ) mutation . We then assayed for the ability of pha-1 ( ts ) high-copy overexpression to rescue the lethal phenotype of pha-1 ( ts ) mutants at 25°C . If the protein product of pha-1 ( ts ) were to retain residual activity at 25°C , we would expect to see some suppression of pha-1 ( ts ) temperature sensitivity . As shown in Table 2 , overexpression of pha-1 ( ts ) efficiently rescued defects associated with genomic pha-1 ( ts ) loss of function , indicating that , at 25°C , pha-1 ( e2123ts ) does not behave as a null allele . Given the absence of a well-characterized null allele of pha-1 , we decided to make use of a regional deficiency on chromosome III , tDf2 , which removes both the pha-1 and sup-35 loci , as well as 46–72 additional genes ( Figure 3A ) . Previous analysis , along with our current work , indicates that homozygous tDf2/tDf2 mutants arrest as embryos that display a phenotype closely resembling pha-1 strong loss-of-function mutations , suggesting that pha-1 may be the earliest-acting zygotic gene within the region deleted by the deficiency [31] . If so , then the apparent lack of suppression observed in tDf2/tDf2 embryos , where both sup-35 and pha-1 are deleted , would suggest that loss of sup-35 cannot suppress the pha-1 null phenotype . Alternatively , another early-acting gene within the deficiency , one that is not suppressed by loss of sup-35 , could be responsible for the pha-1-like phenotype observed in tDf2/tDf2 homozygotes . To distinguish between these two possibilities , we introduced an extrachromosomal array containing wild-type copies of pha-1 into a balanced strain that carries the tDf2 deficiency ( tDf2/qC1 dpy-19 glp-1 ) . In the absence of any array , this strain segregates ∼25% tDf2/tDf2 progeny that arrest as dead embryos with morphological defects similar to those observed for pha-1 ( ts ) mutants at 25°C ( Figure 3B ) . Strikingly , in the presence of pha-1 rescuing arrays , we observed a substantial decrease in the frequency of embryonic lethality ( Figure 3B ) . This effect was observed using multiple independently generated arrays , with the extent of embryonic rescue corresponding closely to the transmission frequencies of the individual arrays ( Figure 3B and data not shown ) . Furthermore , we observed a proportional increase in the percentage of array-positive larval-lethal animals ( Figure 3B ) , indicating that some other gene within the deficiency is required for progression through larval development . Taken together , these results demonstrate that pha-1 is the earliest-acting zygotic gene within tDf2 and , most importantly , that loss of sup-35 cannot suppress the pha-1 null genotype . These findings are also consistent with the observation that sup-35 pha-1 ( e2123 ) /tDf2 animals , which carry only a single copy of the pha-1 hypomorphic allele , display much weaker suppression than that of sup-35 pha-1 ( e2123 ) animals , which retain two copies of this allele ( data not shown; [31] . As an additional test , we made use of two recently generated deletion alleles of pha-1 ( tm3671 and tm3569; gift of National Bioresource Project ) . tm3671 is a 203-bp deletion that removes part of the second exon of pha-1 , creating a premature stop codon after 30 amino acids and is a presumed null allele . tm3569 contains an in-frame 568-bp deletion extending from exon 2 through exon 4 , which removes 149 amino acids of PHA-1 ( isoform Y48A6C . 5a ) . Both pha-1 ( tm3671 ) /+ and pha-1 ( tm3569 ) /+ heterozygous hermaphrodites produce ∼25% embryonic-lethal F1 progeny that phenocopy pha-1 ( ts ) embryos ( at 25°C ) . Consistent with our deficiency analysis , growth of pha-1 ( tm3671 ) /+ and pha-1 ( tm3569 ) /+ heterozygotes on sup-35 ( RNAi ) failed to decrease the percentage of embryonic-arrested progeny , further indicating that reduction of sup-35 activity cannot suppress complete loss of function of pha-1 ( data not shown ) . In contrast , sup-35 ( RNAi ) efficiently suppressed the lethality of pha-1 ( ts ) mutants ( at 25°C ) , as well as all tested synthetic phenotypes ( Table 1 ) . Given that loss of sup-35 cannot suppress the pha-1 null genotype , we hypothesized that SUP-35 may function as a negative upstream regulator of pha-1 . Furthermore , because SUP-35 contains C2H2-type Zn fingers that are critical for its activity ( Figure 1A ) , we reasoned that SUP-35 may mediate repression of pha-1 at the level of transcription ( Figure 4A ) . Consistent with this , qRT-PCR experiments revealed embryonic pha-1 mRNA levels to be 2- to 4-fold more abundant in sup-35 ( tm1810 ) mutants as compared with wild type using two independent internal normalization controls ( Figure 4B ) . An even greater increase in pha-1 mRNA levels was observed in embryos from sup-35 ( e2223 ) pha-1 ( ts ) double mutants relative to pha-1 ( ts ) single mutants ( Figure 4C ) . This latter result is significant in that pha-1 mRNA levels were assessed in a genetic background in which PHA-1 activity is compromised . The observed difference in the degree to which pha-1 is upregulated in these strains could reflect a heightened sensitivity to SUP-35 levels in the pha-1 mutant background or could be due to differences between the two sup-35 alleles used in these studies . As a second test , we made use of a previously described strain that expresses a functional full-length PHA-1::GFP fusion protein [29] . Because this fusion protein is regulated by sequences derived from the native pha-1 promoter , its expression should be sensitive to alterations in the activities of endogenous transcriptional regulators . Consistent with data obtained from qRT-PCR , PHA-1::GFP was upregulated at least 2-fold in sup-35 ( tm1810 ) mutants relative to wild-type embryos ( Figure 4D and 4F–4I; Figure S1 ) . These findings also indicate that changes in pha-1 mRNA levels lead to corresponding changes in the abundance of PHA-1 protein . The above results indicate that SUP-35 may negatively regulate pha-1 at the level of transcription or mRNA stability . To distinguish between these possibilities , we assayed expression levels of a Ppha-1::GFP reporter [29] in wild-type and sup-35 mutants . Because this construct contains only the 5′ upstream regulatory region of pha-1 , effects on mRNA stability through the pha-1 3′UTR should not be observed . Using this reporter , we observed that Ppha-1::GFP is upregulated ∼3-fold in sup-35 mutants versus wild-type embryos ( Figure 4E; Figure S1 ) . Taken together , these data provide strong evidence that SUP-35 normally functions to inhibit pha-1 at the level of transcription . If SUP-35 negatively regulates pha-1 , then sup-35 overexpression should cause a reduction in PHA-1 levels and therefore would be expected to phenocopy pha-1 loss-of-function mutations . Consistent with this , extensive attempts to revert the suppression of sup-35; pha-1 mutants through the expression of wild-type sup-35 via an extrachromosomal array failed to generate stable transgenic lines . This includes experiments in which sup-35 was engineered to be present at low copy numbers . In addition , sup-35 transgenic expression was also highly toxic to wild-type animals , as was expression of the SUP-35::GFP fusion protein . Given that SUP-35 may require the pha-1 suppressors SUP-36 and SUP-37 to mediate its activities , we hypothesized that SUP-35 overexpression may not be toxic in genetic backgrounds that remove either sup-36 or sup-37 activities . Consistent with this prediction , we encountered no difficulties in obtaining stable transgenic lines carrying wild-type sup-35 ( or SUP-35::GFP ) at high copy number in either the sup-36 or sup-37 mutant background ( also see Materials and Methods ) . This finding indicates that SUP-36 and SUP-37 function genetically downstream of SUP-35 . However , SUP-36 and SUP-37 could conceivably function upstream of SUP-35 if they are required for SUP-35 activation . To determine directly the phenotypic effects of SUP-35 overexpression in a wild-type background , we performed a series of genetic crosses , an example of which is shown in Figure 5A . sup-35 overexpression was toxic to hermaphrodites that expressed both sup-36 and sup-37 zygotically , even if sup-36 or sup-37 were absent maternally . Surprisingly , sup-35 overexpression was not toxic in males that expressed both sup-36 and sup-37 zygotically , provided that either sup-36 or sup-37 maternal contributions were absent . In contrast , sup-35 overexpression was toxic to both males and hermaphrodites when both sup-36 and sup-37 were present maternally and zygotically . Most strikingly , non-viable sup-35-overexpressing embryos and larvae obtained through these crosses had a phenotype that was identical to strong pha-1 loss-of-function mutations ( Figure 5B and 5C ) . Taken together , these results are consistent with our finding that SUP-35 functions as a negative regulator of pha-1 and further indicate that SUP-35 acts together with SUP-36 and SUP-37 to control pha-1 expression levels . Our above analyses strongly indicate that sup-35 suppression of partial loss-of-function mutations in pha-1 occurs through the upregulation of pha-1 mRNA , which in turn leads to increased PHA-1 protein levels ( Figure 4 ) . An extension of this model is that suppression of the synthetic pharyngeal genotypes by sup-35 ( Table 1 ) may occur through an identical mechanism . If that is the case , then an increase in PHA-1 levels , even in the presence of wild-type sup-35 , should be sufficient to suppress the synthetic phenotype of lin-35; ubc-18 double mutants . To address this , we overexpressed wild-type PHA-1 from high-copy extrachromosomal arrays in lin-35; ubc-18 double mutants and assayed for rescue . Strong suppression of synthetic lethality was observed in three out of three independent transgenic lines , leading to the generation of viable double mutant strains that carried only the PHA-1-overexpression transgenic array . This finding is consistent with the hypothesis that sup-35-mediated suppression of pha-1 ( ts ) and the synthetic genotypes occurs through the same mechanism . A second prediction of the above model is that inhibition of pha-1 activity should revert the suppression observed in lin-35; sup-35 ( tm1810 ) ubc-18 triple mutants ( Table 1 ) . We therefore subjected triple mutants to pha-1 ( RNAi ) feeding and assayed for loss of suppression . Whereas 100% ( n = 255 ) of lin-35; sup-35 ubc-18 animals reached adulthood when grown on vector-RNAi control plates , only 12 . 9% ( n = 200 ) of triple mutants grown on pha-1 ( RNAi ) escaped embryonic or early-larval arrest . This finding further supports the model that sup-35-mediated suppression of both strong loss-of-function pha-1 mutants and the synthetic genotypes occurs through the common mechanism of increasing PHA-1 levels . In considering potential regulatory networks that could account for both the molecular and genetic data described above , we were able to construct a relatively straightforward model . In this scenario , LIN-35 , functioning as a transcriptional repressor ( Figure 6A ) , and UBC-18–ARI-1 , acting as a complex to promote target protein degradation ( Figure 7A ) , are negative regulators of SUP-35 . Thus in lin-35; ubc-18 double mutants , increased levels of SUP-35 would lead to the inhibition of PHA-1 and associated defects in pharyngeal development . We first tested this model by examining the role of LIN-35 in the expression of endogenous sup-35 . Consistent with the model , embryonic levels of sup-35 mRNA are increased ∼4-fold in lin-35 mutants as compared with wild type ( Figure 6B ) . Correspondingly , SUP-35::GFP was upregulated 2- to 3-fold in embryos following lin-35 ( RNAi ) treatment ( Figure 6C; Figure S2 ) , indicating that changes in sup-35 mRNA levels are further reflected by changes in the abundance of SUP-35 protein . Most importantly , we observed an ∼5- to 10-fold reduction in the levels of endogenous pha-1 mRNA in embryos derived from lin-35 mutants versus those from wild type ( Figure 6B ) . This latter result also provides an explanation for why mutations in lin-35 are strongly synthetic with hypomorphic mutations in pha-1 ( also see Discussion ) . We next examined the roles of UBC-18 and ARI-1 in the regulation of SUP-35 and PHA-1 . In contrast to findings from lin-35 mutants , embryonic sup-35 mRNA levels in ubc-18 mutants were identical to those observed in wild type ( Figure 7B ) . Nonetheless , embryonic SUP-35::GFP protein levels were substantially increased following RNAi inhibition of ubc-18 or ari-1 ( Figure 7C; Figure S2 ) . These results indicate that UBC-18–ARI-1 negatively regulates SUP-35 post-transcriptionally , possibly at the level of SUP-35 stability . Consistent with this , we find that SUP-35::GFP is a target for ubiquitination in cell extracts from whole worms ( Figure 7D ) . Furthermore , we observed that the increase in SUP-35 levels in ubc-18 mutants correlates with a decrease in the expression levels of pha-1 mRNA ( Figure 7B ) . These findings , in combination with other molecular and genetic data , strongly support the model that LIN-35 and UBC-18–ARI-1 promote pha-1 transcription by inhibiting SUP-35 expression and stability . In previous studies , we have implicated the C . elegans E2F ortholog , EFL-1 , as a regulatory partner of LIN-35 in the control of pharyngeal development [29] , and have also defined the C . elegans E2F consensus binding motif [13] . Consistent with a role for E2F in the regulation of sup-35 , we identified three candidate E2F bindings sites within the first 700 bp of the sup-35 promoter region . One of these sites , located approximately 230 bp upstream of the predicted transcriptional start site ( GATTCGCGCCT ) , conformed to all published criteria , suggesting that E2F may potentially regulate sup-35 directly . Studies in mammals have implicated HCF-1 ( host cell factor 1 ) , as an important physical and functional co-partner of E2F in the activation of E2F target genes [37] , [38] . For example , loss of HCF-1 activity in hamster cells leads to a reduction in the expression of E2F-regulated genes required for G1 entry resulting in arrest in G0 [39] . Interestingly , this G0 arrest can be bypassed through the inhibition of pRb family members , indicating that mammalian HCF-1 and pRb carry out opposing functions on E2F targets [40] . The presence of a structurally and functionally conserved ortholog of HCF-1 in C . elegans [41]–[43] , led us to hypothesize that a similar regulatory relationship may exist in C . elegans ( Figure 8A ) . To test this , we assayed levels of sup-35 mRNA in lin-35 mutants subjected to hcf-1 ( RNAi ) by qRT-PCR . Notably , we observed an ∼2-fold reduction in the levels of sup-35 mRNA in lin-35; hcf-1 ( RNAi ) embryos as compared with lin-35 mutants treated with a control RNAi ( Figure 8B ) . To see if the observed reduction in sup-35 mRNA levels by hcf-1 ( RNAi ) has a functional consequence in lin-35; ubc-18 and lin-35; pha-1 double mutants , we carried out hcf-1 ( RNAi ) in these backgrounds and assayed for suppression of larval arrest , leading to the generation of fertile adults . Notably , reduction of hcf-1 activity led to pronounced suppression of arrest in both lin-35; ubc-18 and lin-35; pha-1 mutant backgrounds ( Figure 8C ) . We note that the partial phenotypic suppression of the synthetic mutants by hcf-1 ( RNAi ) is consistent with the incomplete correction of sup-35 overexpression in lin-35; hcf-1 ( RNAi ) embryos ( Figure 8B ) . In addition , hcf-1 ( RNAi ) resulted in suppression of pha-1 ( ts ) mutants at the intermediate temperature of 20°C , leading to a marked decrease in L1-larval pharyngeal defects and a corresponding increase in the frequency of fertile adults ( Figure 8C ) .
We report here the molecular identification and analysis of SUP-35 . We provide evidence that loss of sup-35 activity specifically suppresses the embryonic- and larval-lethal phenotypes of pha-1 hypomorphic alleles . Additionally , loss of sup-35 activity efficiently suppressed the synthetic lethal phenotypes of lin-35; pha-1 and lin-35; ubc-18 double mutants , as well as a number of related genotypes . sup-35 is predicted to encode a C2H2-type Zn-finger protein , consistent with a role in transcriptional regulation ( Figure 1 ) , although other functional activities associated with Zn fingers domains are possible . Based on sequence similarity , SUP-35 is also a new member of the RMD family of proteins , several of which have been shown to associate with microtubules [32] . During early embryonic development , a functional SUP-35::GFP protein was expressed predominantly in the cytoplasm of most or all cells . Notably , at the onset of morphogenesis , SUP-35::GFP expression became enriched in pharyngeal nuclei ( Figure 2 ) . How this dynamic pattern of expression , as well as a potential association with microtubules , may contribute to the functions and regulation of SUP-35 is currently unclear . Further studies of SUP-35 , as well as the additional suppressors SUP-36 and SUP-37 , should shed light on these facets of SUP-35 regulation . In previous work , we have shown that LIN-35 , a transcriptional repressor , and UBC-18–ARI-1 , an E2-E3 ubiquitin ligase complex , redundantly regulate pharyngeal morphogenesis [21] , [34] . In addition , mutations in lin-35 , ubc-18 , and ari-1 strongly enhance the pharyngeal morphogenetic defects of partial loss-of-function mutations in pha-1 [29] , [34] . In our current study , we provide both molecular and genetic evidence that LIN-35 and UBC-18–ARI-1 function as negative regulators of SUP-35 , which in turn functions as a transcriptional repressor of pha-1 . Thus , in our model , both LIN-35 and UBC-18–ARI-1 are positive , albeit indirect , regulators of PHA-1 through the inhibition of SUP-35 ( Figure 9 ) . Evidence to support this model includes the findings that pha-1 overexpression efficiently rescued the synthetic lethality of lin-35; ubc-18 double mutants and that the suppression observed in lin-35; ubc-18 sup-35 triple mutants was reversed by pha-1 ( RNAi ) . Furthermore , sup-35 overexpression in a wild-type background phenocopied pha-1 loss of function ( Figure 5 ) . Consistent with the genetic data , qRT-PCR and GFP reporters indicate that sup-35 mRNA and protein levels were upregulated in embryos where lin-35 activity had been compromised , whereas ubc-18 and ari-1 specifically affected SUP-35 protein levels ( Figure 6 ) . Additionally , endogenous pha-1 mRNA levels were decreased in lin-35 and ubc-18 mutants , whereas pha-1 mRNA and protein levels were increased in sup-35 mutants ( Figure 6 , Figure 7 ) . This model accounts for both the synthetic lethality of lin-35; ubc-18 double mutants as well as the genetic interactions observed between pha-1 and lin-35 , ubc-18 , and ari-1 , as pha-1 hypomorphic mutations would be expected to be hypersensitive to conditions that further reduce pha-1 mRNA levels . An additional prediction of this model is that strong loss-of-function pha-1 mutants should minimally phenocopy the defects observed in lin-35; ubc-18 and lin-35; pha-1 mutants . Specifically , lin-35; ubc-18 and lin-35; pha-1 mutants show early-stage defects in the re-orientation of anterior epithelial cells within the pharyngeal primordium [21] , [29] . Surprisingly , however , we had previously failed to observe re-orientation defects in pha-1 ( ts ) embryos grown at 25°C [29] , even though these mutants show severe pharyngeal morphogenesis defects at later stages [29] , [30] . We have subsequently repeated these experiments and , consistent with our earlier study , find little or no evidence for early-stage morphogenesis defects in pha-1 ( ts ) embryos grown at the non-permissive temperature on either NGM or vector-RNAi control plates ( data not shown ) . In contrast , pha-1 ( ts ) mutants grown at 16°C on pha-1 ( RNAi ) plates did display early-stage pharyngeal morphogenesis defects , demonstrating that a specific reduction in pha-1 activity can phenocopy the early-stage defects observed in the synthetic mutants ( data not shown ) . Moreover , the frequency and severity of pha-1 ( ts ) ; pha-1 ( RNAi ) morphogenesis defects were similar to those observed for pha-1 ( ts ) ; lin-35 ( RNAi ) and pha-1 ( ts ) ; ubc-18 ( RNAi ) embryos grown at 16°C ( data not shown ) . These observations indicate that early-stage defects in pha-1 ( ts ) mutants are suppressed by growth at 25°C , suggesting an effect of temperature on the underlying process of cell re-orientation . Most importantly , these findings are internally consistent with our model , in which PHA-1 levels are positively regulated by LIN-35 and UBC-18 through the inhibition of SUP-35 ( Figure 9 ) . Our observation that mutations in sup-36 and sup-37 abolish SUP-35-mediated toxicity indicate that sup-36 and sup-37 act genetically downstream of SUP-35 . Thus , SUP-36 and SUP-37 may potentially function downstream of SUP-35 in a linear pathway to control pha-1 expression . Alternatively , SUP-36 and SUP-37 may act in a complex with SUP-35 , or in a parallel pathway that is required for SUP-35 activation ( Figure 9 ) . We also find that inhibition of hcf-1 by RNAi leads to a partial , though significant , suppression of larval arrest in lin-35; ubc-18 and lin-35; pha-1 mutants as well as the substantive suppression of both the L1 arrest and Pun ( Pharynx unattached ) phenotypes of pha-1 ( ts ) mutants at 20°C . This genetic suppression correlates well with the observed decrease in sup-35 mRNA levels in lin-35; hcf-1 ( RNAi ) embryos . These results are consistent with our current model as well as previously published findings on mammalian HCF-1 [29] , [44] , and append our model with the addition of a phylogenetically-conserved component of the E2F network ( Figure 9 ) . Our finding also indicates that additional novel suppressors may be identified through the use of sensitized strains . Elucidating the mechanistic bases of synthetic genetic interactions will continue to be a major challenge for the field of developmental genetics . These types of interactions will also likely be critical to our understanding of complex disease traits in humans . For example , a recent commentary in the New England Journal of Medicine states that “many , rather than few , variant risk alleles are responsible for the majority of the inherited risk of each common disease” [45] . Our current analysis provides a straightforward model to account for the genetic redundancies observed in an additional case study . Although understanding different sets of genetic interactions will undoubtedly require unique solutions , we contend that certain patterns of redundancy are likely to emerge . In this case , we have shown that a redundancy between a transcriptional regulator , LIN-35 , and a mediator of protein stability , UBC-18–ARI-1 can be explained through the negative regulation of a common target , SUP-35 . Similarly , we have previously shown that LIN-35 and FZR-1 , a substrate-specificity component of the APC ( anaphase-promoting complex ) E3 ligase , mutually inhibit the expression levels of G1 cyclins [14] . Thus , a potential theme to emerge from our studies is the redundant control of common targets through distinct mechanisms of negative regulation . Additional studies into synthetic phenotypes in C . elegans and other systems should further elucidate general themes that may govern genetic redundancy .
C . elegans were maintained using standard procedures [46] . Strains used in our analysis include GE24 [pha-1 ( e2123 ) ] , GE348 [dpy-18 sup-35 ( e2223 ) pha-1 ( e2123 ) ] , WY83 [lin-35; ubc-18; kuEx119 ( lin-35+; sur-5::GFP] , WY119 [lin-35; pha-1 ( fd1 ) ; kuEx119] , sup-35 ( tm1810 ) , WY477 [dpy-18 pha-1 ( e2123 ) ; ari-1 ( tm2549 ) ] , WY482 [sup-35 ( tm1810 ) ; SM469 ( PHA-1::GFP; pRF4 rol-6gf ) ] , WY527–528 , [lin-35;ubc-18; kuEx119; fdEx72–73 ( pBX;rol-6 ( su1006gf ) ) ] , WY529–530 [lin-35; ubc-18; fdEx72–73] GE2158 [tDf2/qC1 dpy-19 ( e1259 ) glp-1 ( q339 ) ] , WY539–542 [unc-13 lin-35; dpy-17 ubc-18 sup-35 ( tm1810 ) ] , GE348 [dpy-18 sup-35 ( e2223 ) pha-1 ( e2123ts ) ] , GE551 [vab-7 ( e1562 ) sup-35 ( t1013 ) pha-1 ( e2123ts ) ] , GE552 [vab-7 ( e1562 ) sup-35 ( t1014 ) pha-1 ( e2123ts ) ] , GE913 [vab-7 ( e1562 ) sup-35 ( t1016 ) pha-1 ( e2123ts ) ] , GE914 [vab-7 ( e1562 ) sup-35 ( t1015 ) pha-1 ( e2123ts ) ] , GE915 [vab-7 ( e1562 ) sup-35 ( t1017 ) pha-1 ( e2123ts ) ] , and WY453–466 [sup-35 ( fd33–46 ) pha-1 ( e2123ts ) ] . SM35 [PHA-1::GFP] , SM36 [Ppha-1::GFP] . To analyze SUP-35 overexpression and toxicity , the following strains were generated using either a sup-35 genomic fragment or a cloned sup-35:GFP construct: WY512–513 [pha-1 ( e2123ts ) ; sup-36 ( e2217 ) ; fdEx57–58 ( sup-35::GFP; rol-6 ) ] , WY514–517 [pha-1 ( e2123ts ) ; sup-36 ( e2217 ) ; fdEx59–62 ( sup-35 genomic fragment; sur-5::GFP ) ] , WY518 [pha-1 ( e2123ts ) ; sup-37 ( e2215 ) ; fdEx63 ( sup-35::GFP; rol-6 ) ] , WY519–520 [pha-1 ( e2123ts ) ; sup-37 ( e2215 ) ; fdEx64–65 ( sup-35 genomic fragment; sur-5::GFP ) ] , WY523–524 [dpy-11 sup-3; fdEx68–69 ( sup-35 genomic fragment;sur-5::GFP ) ]; WY525–526 , [dpy-11 sup-3; fdEx70–71 ( sup-35::GFP; rol-6 ) ] . Strains used for rescue analysis of pha-1 ( e2123ts ) and the chromosomal deficiency tdf2 included WY506–511 [pha-1 ( e2123ts ) ; fdEx51–56 ( pBX/e2123; sur-5::GFP ) ] and WY531–534 [tDf2/qC1 dpy-19 ( e1259 ) glp-1 ( q339 ) ; fdEx74–77 ( pBX; sur-5::GFP ) ] . lin-35 ( n745; ubc-18 ( ku354 sup-35 ( tm1810 ) triple mutants were generated by crossing sup-35 ( tm1810 ) /+ males to dpy-17 ubc-18 unc-32 hermaphrodites . Cross-progeny were allowed to self , and the resulting Dpy non-Unc recombinants were assayed for the sup-35 ( tm1810 ) deletion by PCR . Confirmed dpy-17 ubc-18 sup-35 ( tm1810 ) triple-mutant hermaphrodites were then crossed to unc-13 lin-35/+ males . Following selfing of the cross-progeny , Dpy Unc animals were confirmed for lin-35 ( n745 ) , ubc-18 ( ku354 ) , and sup-35 ( tm1810 ) by PCR and DNA sequencing . To test for rescue of lin-35; ubc-18 double mutants by pha-1 overexpression , plasmid pBX , which contains a rescuing segment of the pha-1 genomic locus [47] , was co-injected with pRF4 , which contains the dominant rol-6 ( su1106 ) marker [48] , into strain WY83 . Stable double transgenics were recognized by the presence of rolling sur-5::GFP ( + ) animals . Rescue was then determined by the presence of rolling viable non-GFP adults that could be further propagated in the absence of kuEx119 . A SUP-35::GFP fusion ( pDF101 ) was constructed as follows . An ∼2 . 5-kb sup-35 genomic fragment , which includes the upstream sup-35 promoter/enhancer region , was amplified using the primer pair 5′-GCTCTAGATGATAGTCGTGTCGGTGGTCGTC-3′ and 5′-CGCGGATCCAATTGAGCACAAGTCAAGGGCGTCG-3′ . This fragment was digested with BamHI and XbaI and cloned in-frame into a similarly restricted pPD95 . 77 vector ( gift of A . Fire ) . All recombinant clones were verified by restriction digestion and sequencing . For the rescue of pha-1 ( e2123ts ) mutants by pha-1 ( e2123ts ) overexpression , a fragment of the pha-1 genomic locus was amplified from pha-1 ( e2123ts ) mutants using the primer pair 5′-CAGGACAATGATCTCGCCTT-3′ and 5′-TATCTTTTCACATGGAATACATGTAG3′ and digested with SalI and BsaBI . This fragment was then used to replace the analogous region of pBX . Recombinant plasmids carrying the e2123ts mutation were identified by digestion with Bst1107I , which recognizes the SNP created by the e2123ts point mutation , and further confirmed by sequencing . RNAi feeding was carried out using standard protocols , and plates were cultured at 25°C to score for suppression [49] . The RNAi constructs JA:Y48A6C . 3 , JA:Y48A6C . 5 , and JA:R01H12 . 6 were used to target sup-35 , pha-1 , and ubc-18 gene products , respectively . RNAi constructs used to target lin-35 and ari-1 were previously described [14] , [34] . hcf-1 ( RNAi ) feeding was carried out using construct JA:C46A5 . 9 , corresponding to exons 2–4 . RNAi injection of hcf-1 was carried out by gonadal injection of dsRNA ( ∼1 . 0 mg/ml ) corresponding to exons 5 and 6 . Fluorescence microscopy was performed using a Nikon Eclipse microscope . Quantification of the GFP fluorescence in embryos was carried out using Open Lab Software Version 5 . 0 . 2 . All images were captured using identical exposure times , and all embryos used in our analysis were of similar developmental stages ( ∼200–300 cells ) . An average of the mean fluorescence was calculated to compare GFP expression levels . P values were determined using a Student's t-test . Because multicopy transgene expression of SUP-35 and SUP-35::GFP was toxic in wild-type backgrounds , arrays were initially generated in sup-36 and sup-37 mutants . To determine the effect of SUP-35 and SUP-35::GFP overexpression in wild-type animals , males of genotype +/+; mIs11 ( myo-2::GFP ) were crossed to pha-1 ( e2123ts ) ; sup-36; fdEx59 hermaphrodites . fdEx59 expresses wild-type sup-35 and the co-injection marker sur-5::GFP . Such crosses resulted in the generation of fdEx59+ F1 males only , which were identified by the presence of both sur-5::GFP and myo-2::GFP . F1 males were then mated to either N2 hermaphrodites or homozygous sup-36 hermaphrodites . When the F1 males were crossed to sup-36 hermaphrodites , non-viable Pun and viable cross-progeny animals were obtained , whereas all the cross-progeny from the N2 hermaphrodite matings were non-viable and exhibited the Pun phenotype . These results were reproduced using three independently generated extrachromosomal arrays in both sup-36 and sup-37 mutant backgrounds . Similar results were also obtained for the SUP-35::GFP construct co-injected with pRF4 . As an alternative approach , males of the genotype dyp-13 unc-24/+ were crossed to pha-1 ( e2123ts ) ; sup-36; fdEx59 hermaphrodites . F1 hermaphrodites were placed on individual plates and allowed to self; cross-progeny were determined by the presence of Dpy Unc animals . In the event that SUP-35 overexpression was non-toxic , half of the cross-progeny F1s [pha-1 ( e2123ts ) /+;sup-36/dpy-13 unc-24; fdEx59] should have segregated one-sixteenth of the F2 animals with a genotype of +/+; dpy-13 unc-24; fdEx59 . Although our crosses resulted in a high frequency of F1 cross-progeny males , they failed to produce F1 hermaphrodites that segregated Dpy Unc F2 animals . To extend these results , F1 cross-progeny males were subsequently crossed to N2 hermaphrodites . This cross resulted in fdEx59+ animals that arrested uniformly as arrested embryos or larvae that exhibited the Pun phenotype . Again , these results were reproducible with other independently generated arrays and when analogous crosses were performed in the sup-37 mutant background Strains were grown at 16°C and total RNA from bleached embryos was isolated using the Trizol reagent ( Invitrogen ) followed by phenol-chloroform extraction . All samples were DNase ( Invitrogen ) treated and cleaned using the RNeasy Midi Kit ( Qiagen ) . cDNA was synthesized using random primers and Superscript reverse transcriptase II ( Invitrogen ) at 42°C for 1 hour . First-strand cDNA was purified using the Qiagen Microelute Kit and eluted in 10 l final volume . Primer pairs used for the various genes include pha-1 [5′-TCGACTGGAGCTTCGTGTAAGTCA-3′ and 5′-ACGGTGCAAGGGCATTAAGGAAAC-3′]; ama-1 [5′-TGATGTGATGACTGCGAAGGGACA-3′ and 5′-TTCGAATGAACAACGCATCAGGGC-3′]; act-1 [5′-TTACTCTTTCACCACCACCGCTGA-3′ and 5′-TCGTTTCCGACGGTGATGACTTGT-3′]; and sup-35 [5′-GATCATGCGAGCGGTTATTCGTC-3′ and 5′-GATCGATGGACTTCTCTCCAGAA-3′] . All primer pairs amplified regions that spanned sizeable introns such that cDNA amplification was strongly favored . Furthermore , we did not detect genomic contamination in our cDNA samples based on several tests including gel-purified amplimer band sizes . Primer pairs used for the act-1 internal normalization are predicted to amplify act-1–3 . Primer pairs used for the ama-1 were specific to this gene . qRT-PCR was performed using a BioRad icycler in a total reaction volume of 50 l using the BioRad SYBR green supermixwith the following reaction conditions: initial denaturation at 95°C for 3 min , followed by 40 cycles of denaturation at 95°C for 30 seconds and a combined annealing and extension step at 60°C for 30 seconds . After the final amplification cycle , a melt curve analysis was performed to examine the specificity of the reaction . The fold-change of the mRNA levels was calculated by the delta-delta Ct method For each qRT-PCR experiment , amplification was done in triplicate for both the test and the normalization genes , and the results were checked for reproducibility using at least one biological duplicate . In addition , all data were reproduced using at least two biological replicates . P values were determined using a Student's t-test . Mixed-stage worms from 10 large NGM-OP50 plates were pooled and washed with M9 and distilled water and resuspended in 500 µl of homogenization buffer ( 20 mM Tris-HCl pH 7 . 5 , 100 mM NaCl , 5 mM MgCl2 , 1 mM EGTA , 1 mM DTT , 1% TritonX-100 , protease inhibitors ) . Worms were then sonicated , incubated on ice , and lysates were cleared of large particles by centrifugation . To immunoprecipitate SUP-35::GFP , precleared worm lysate was incubated with 5 µg of polyclonal anti-GFP antibody ( Santa Cruz ) at 4°C for 2 hrs and the resulting immune complex was pulled down using 30 µl of proteinA-sepharose beads ( Invitrogen ) by over-night incubation at 4°C . Beads were washed 3× with cold homogenization buffer and subjected to SDS-PAGE and western blot analysis . Westerns to detect ubiquitinated products were carried out using either 2 µg of monoclonal anti-ubiquitin antibody ( Santa Cruz ) or 2 µg of monoclonal anti-GFP primary antibody ( Invitrogen ) . Visualization was carried out using HRP-conjugated goat anti-mouse secondary antibodies ( Santa Cruz ) at 1∶5000 and peroxidase activity was detected by the enhanced chemiluminesence assay ( Pierce ) . LLnL-treated Jurkat cell lysate ( Santa Cruz ) was used as a positive control for ubiquitination . | One of the more puzzling aspects of genetics is that the inactivation of many genes fails to produce strong deleterious effects on the organisms that carry those genes . In some cases , however , the combined inactivation of two or more such genes can lead to the expression of robust abnormal phenotypes . These types of synthetic genetic interactions are thought to reflect the presence of functional overlap or redundancy between the involved genes . The root mechanisms that underlie synthetic interactions are thought to be complex and are in most cases poorly understood . Our work here focuses on one case study where we have uncovered the molecular basis underlying a complex set of genetic redundancies in C . elegans . More specifically , we have discovered a novel regulatory network that connects eight genes controlling embryonic foregut development in the nematode C . elegans . By solving mechanisms of this nature , our analysis provides a means for understanding more generally the principles that govern genetic redundancies . Our work also provides insight into the bases of complex disease traits , where the combined interactions of multiple genetic factors leads to outcomes that determine health or disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"developmental",
"biology",
"genetics",
"and",
"genomics/gene",
"expression",
"genetics",
"and",
"genomics/gene",
"function",
"developmental",
"biology/organogenesis",
"developmental",
"biology/developmental",
"molecular",
"mechanisms"
]
| 2009 | A Mechanistic Basis for the Coordinated Regulation of Pharyngeal Morphogenesis in Caenorhabditis elegans by LIN-35/Rb and UBC-18–ARI-1 |
Proteins with low-complexity domains continue to emerge as key players in both normal and pathological cellular processes . Although low-complexity domains are often grouped into a single class , individual low-complexity domains can differ substantially with respect to amino acid composition . These differences may strongly influence the physical properties , cellular regulation , and molecular functions of low-complexity domains . Therefore , we developed a bioinformatic approach to explore relationships between amino acid composition , protein metabolism , and protein function . We find that local compositional enrichment within protein sequences is associated with differences in translation efficiency , abundance , half-life , protein-protein interaction promiscuity , subcellular localization , and molecular functions of proteins on a proteome-wide scale . However , local enrichment of related amino acids is sometimes associated with opposite effects on protein regulation and function , highlighting the importance of distinguishing between different types of low-complexity domains . Furthermore , many of these effects are discernible at amino acid compositions below those required for classification as low-complexity or statistically-biased by traditional methods and in the absence of homopolymeric amino acid repeats , indicating that thresholds employed by classical methods may not reflect biologically relevant criteria . Application of our analyses to composition-driven processes , such as the formation of membraneless organelles , reveals distinct composition profiles even for closely related organelles . Collectively , these results provide a unique perspective and detailed insights into relationships between amino acid composition , protein metabolism , and protein functions .
Low-complexity domains ( LCDs ) in proteins are regions enriched in only a subset of possible amino acids . LCDs can be composed of homopolymeric repeats of a single amino acid , short tandem repeats consisting of only a few different amino acids , or aperiodic stretches with little amino acid diversity [1] . Proteins containing LCDs are relatively common among organisms from all domains of life , and are particularly common among eukaryotes [2–4] . For example , approximately 70% of genes in the Saccharomyces cerevisiae genome possess at least one classically-defined LCD [3] . Furthermore , the total number of LCDs far exceeds the total number of yeast genes ( ~2-fold more LCDs than genes ) , indicating that many genes contain multiple distinct LCDs . Various methods have been developed to assess biopolymer sequence complexity [1 , 5–9] . One of the most commonly employed methods to define LCDs is the SEG algorithm [1] , which scans protein ( or nucleic acid ) sequences using a short sliding window , and calculates the local Shannon entropy for each window ( see [10] for a detailed description ) . Subsequences with a Shannon entropy value below a pre-determined “trigger” threshold are classified as LCDs . LCD boundaries are later extended and refined by merging overlapping LCDs and calculating combinatorial sequence probabilities . Another metric commonly used to assess relative sequence complexity is compositional bias , which involves determining the statistical probability of a sequence given whole-proteome frequencies of the individual amino acids [11 , 12] . These approaches ( or closely-related approaches ) have been used extensively to examine LCDs on a proteome-wide scale [1 , 3 , 12–17] . LCD-containing proteins have been implicated in a variety of normal and pathological cellular processes . For example , Q/N-rich yeast proteins often play a role in transcription regulation , endocytosis , and cell cycle regulation , among other functions [11 , 18] . Many proteins containing Q/N-rich LCDs , or LCDs of related types ( Q/N/G/S/Y-rich LCDs ) have been linked to prion or prion-related processes [11 , 18–21] . Additionally , many prion-like LCDs , which are often composed of short tandem repeats of low-complexity [22] , have been linked to stress granules and processing bodies ( P-bodies ) in eukaryotes ( see [23] for recent review ) . The amino acid composition of these LCDs confers unusual biophysical properties to these domains [24] , which likely relates to their unique behavior in vitro and in vivo [25–30] . However , these unusual characteristics appear to be inseparably linked to pathological processes as well . For example , genetic expansion of regions encoding homopolymeric glutamine repeats ( the simplest type of LCD ) in various proteins can lead to a multitude of neurodegenerative disorders , including Huntington’s Disease and spinocerebellar ataxias ( for review , see [31] ) . Furthermore , mutations in the LCDs of stress granule proteins can alter stress granule dynamics and lead to degenerative diseases [26 , 28 , 30 , 32 , 33] . The importance of LCDs extends well beyond Q/N-rich LCDs , as LCDs of other compositions have also been linked to normal and pathological cellular processes [12 , 14 , 17 , 34 , 35] . Although LCDs can clearly impact protein regulation and function , a number of challenges have thus far limited a proteome-scale understanding of these relationships . One major challenge lies in defining LCDs . Current approaches use statistically-defined thresholds for sequence complexity or compositional bias [1 , 11] , or arbitrarily-chosen repeat lengths for proteins with homopolymeric repeats [34–41] . Although these definitions of LCDs , compositionally biased sequences ( herein referred to as “statistically-biased domains” to avoid later confusion ) , or homopolymeric repeats have facilitated important discoveries , the biological relevance of these thresholds has not been rigorously examined . Furthermore , these proteins are often grouped into a single class even though their compositions , and therefore physical properties , can differ dramatically ( a limitation that was appreciated in a recent review [42] ) . To address these limitations , we have developed an alternative approach to infer relationships between amino acid composition and protein metabolism and function . By focusing on amino acid composition , which is the fundamental feature underlying both sequence complexity and statistical amino acid bias , we examined links between local compositional enrichment and various aspects of protein regulation and function without appealing to pre-defined sequence complexity or statistical bias thresholds . We find that local compositional enrichment correlates with differences in nearly all core aspects of a protein’s tenure in the cell , including translation efficiency , abundance , half-life , protein-protein interaction promiscuity , subcellular localization , and function . However , enrichment for different amino acids is associated with different effects , even for residues often grouped based on physicochemical similarities , highlighting the importance of distinguishing LCDs of different types . These relationships are discernible at compositions below those required for classification as low-complexity or statistically-biased , suggesting that the thresholds in traditional methods may not be biologically optimized . Finally , analysis of experimentally-defined protein components of stress granules and P-bodies reveals both shared and distinct compositional features associated with these organelles .
Fundamentally , both sequence complexity and statistical amino acid bias are indirect measures of local amino acid composition . Since composition is a more direct indication of overall protein domain properties , we sought to examine whether composition alone could be used to infer residue-specific relationships between local amino acid composition and protein regulation and function . We first developed an algorithm to partition the yeast proteome on the basis of maximum local composition for each amino acid using a series of scanning window sizes ( Fig 1; see Methods ) . For all amino acids , the majority of proteins are partitioned into composition bins of ≤ 25% ( Fig 2 and S1 Table ) . However , the number of proteins achieving higher local compositions , indicated by a right-hand shoulder or tail in the distribution , were strongly residue-dependent . For example , proteins containing local enrichment of highly hydrophobic residues ( I , L , M , and V ) , aromatic residues ( F , W , and Y ) , or cysteine are almost exclusively limited to composition bins of ≤ 45% for the smallest window size , whereas alanine and proline distributions extend to slightly higher composition ranges ( up to 60–65% ) . Proteins containing local enrichment of polar ( G , N , Q , S , and T ) or charged ( D , E , and K ) residues in composition bins of ≥ 40% are relatively common even among larger window sizes ( albeit to differing degrees ) , whereas histidine and arginine rich regions are relatively rare . These data indicate that relatively high local enrichment is tolerated for some amino acids , while compositional enrichment for other amino acids appears to be restricted in yeast . While the origins and evolution of LCDs have been extensively explored [3 , 4 , 14 , 38 , 43 , 44] , the regulation and metabolism of LCD-containing proteins remain poorly-understood . Proteins with intrinsically disordered segments , which often qualify as LCDs [45 , 46] , have been associated with lower protein half-lives [47] . However , not all intrinsically disordered regions lead to short protein half-lives , and not all LCDs are intrinsically disordered [15] . Additionally , proteins with homopolymeric repeats , when considered as a single class , are associated with lower translation efficiency , lower protein abundance , and lower protein half-life compared to proteins lacking homopolymeric repeats [37] . However , the regulation and structural properties of proteins with LCDs or homopolymeric repeats is likely strongly dependent on the predominant amino acids within the domain of interest [42] . To explore relationships between local compositional enrichment and protein metabolism , we first examined possible links between local compositional enrichment and protein abundance . Recent advances in proteomic methods have facilitated remarkable proteome coverage for both protein abundance [48] and protein half-life [49] measurements in yeast . At each window size/percent composition bin , the distribution of protein abundance values for all proteins partitioned into that bin was compared to the protein abundance distribution for all other yeast proteins ( Mann-Whitney U test ) . Transitions from significantly lower median abundance to significantly higher median abundance or vice versa are observed upon enrichment for many amino acids individually ( Fig 3 ) . However , the direction of the trends upon progressive compositional enrichment are dependent on amino acid type . For the majority of amino acids ( C , D , F , H , I , L , M , N , P , Q , R , S , T , W , or Y ) compositional enrichment is associated with lower median protein abundance . However , compositional enrichment of A , G , or V is associated with higher median protein abundance . Two very similar transitions are observed for both E-rich and K-rich sequences: as compositional enrichment increases , the relative median protein abundance transitions from high to low , then back to high . Collectively , these trends are consistent with , yet much stronger than , previously observed correlations between protein abundance and whole-protein composition [50 , 51] . This suggests that the trends observed previously may actually reflect the effects of local compositional enrichment , which would increase apparent whole-protein composition for the enriched amino acid yet be dampened by confounding effects from the remainder of the protein sequence . Similar trends are observed when compositional enrichment is compared to protein half-lives ( Fig 4 ) . Compositional enrichment for the majority of amino acids ( C , H , K , M , N , P , S , or T ) is associated with lower protein half-life , whereas enrichment for A , G , I , or V is associated with higher protein half-life . Enrichment for F leads to an initial transition from lower to higher half-lives , while further enrichment leads to a transition back to lower half-lives . It is worth noting that similar trends were observed in an independent protein half-life dataset when the proteins were analyzed based on whole-protein amino acid composition [52] , suggesting that maximum local composition is sufficient to detect associations between amino acid composition and half-life . Although for many amino acids the trends are readily apparent , the strength of the association between compositional enrichment and protein half-life appears to be slightly weaker than the association between compositional enrichment and protein abundance . This is likely due , at least in part , to limited proteome coverage ( relative to the protein abundance dataset ) . However , a recent study also suggested that protein half-life is strongly affected by factors other than sequence characteristics [53] , which would likely further dampen relationships between compositional enrichment and protein half-life . Finally , protein half-life is generally less-conserved than protein abundance [54] , perhaps suggesting that specific relationships between conserved sequence features and protein half-life may not be particularly strong . Therefore , it is rather surprising that we observe the indicated trends in spite of these limitations , and could suggest that half-life is more strongly influenced by local composition than particular primary sequence motifs . Direct measurement of protein synthesis rates is more experimentally challenging . Consequently , proteome-wide coverage for experimentally-derived translation efficiency remains substantially lower than coverage for protein abundance and half-life . The normalized translation efficiency ( nTE ) , a reported metric of translation elongation efficiency [55] , is based on codon usage frequencies and tRNA gene copy numbers , allowing for calculation of translation efficiency for the entire proteome . Therefore , we first examined relationships between local compositional enrichment and calculated translation elongation efficiency . nTEs were calculated for whole-protein sequences using the corresponding coding region on mRNA transcripts ( see Methods ) . Translation efficiency is strongly dependent on the locally-enriched amino acid ( Fig 5 ) . For the majority of amino acids ( C , D , E , F , H , I , K , L , M , N , P , Q , R , or Y ) , local enrichment is associated with significantly lower median nTEs suggesting that , as a single class , proteins with local compositional enrichment tend to be translated relatively inefficiently . Proteins with domains enriched in S , T , or W are generally associated with significantly lower median nTEs , although proteins with very high S , T , or W enrichment are associated with significantly higher median nTEs . However , proteins with domains enriched in A , G , or V residues are consistently associated with significantly higher median nTEs , suggesting that these proteins may be translated relatively efficiently . Remarkably , nearly identical trends are observed between local compositional enrichment and the experimentally-derived protein synthesis rates reported for a limited proteome ( S1 Fig ) despite a substantial reduction in sample size ( n = 1115; [56] ) , suggesting that nTE can serve as a good surrogate for overall protein synthesis efficiency . Collectively , these results indicate that local amino acid enrichment is associated with differences in protein production rates in a composition-dependent manner . For most amino acids , we noticed a remarkable correspondence in the trends for translation efficiency , protein abundance , and protein half-life , despite the fact that these values are derived from entirely different methods and experiments . For example , local enrichment for many amino acid types is associated with low nTE values , low protein abundance , and low protein half-life ( Table 1 ) . While translation efficiency and protein degradation rate are largely functionally independent in cells , protein abundance depends , at least in part , on both translation efficiency and protein half-life [49] . This may suggest that protein abundance for these proteins is limited in cells by a combination of poor translation efficiency and rapid degradation rate . In contrast , local enrichment for some amino acids is associated with high protein abundance also tended to have higher nTE values and higher half-lives , perhaps suggesting that high protein abundance for these proteins is achieved by a combination of efficient translation and poor degradation . As a model eukaryotic organism , S . cerevisiae provides a number of important advantages in proteome-scale studies relating protein sequence to protein metabolism and function . In addition to the unmatched proteome coverage in protein abundance and protein half-life datasets , and the availability of yeast-specific tools such as nTE , sequence-function analyses in yeast are further simplified by the absence of tissue-specific effects and limited alternative splicing ( only ~4% of yeast genes contain introns and , of those genes , only a small fraction is capable of producing alternative protein products [57 , 58] ) . With these caveats in mind , we sought to examine whether similar relationships between local amino acid composition and protein abundance could be detected in a model multicellular eukaryotic organism . We decided to focus on whole-organism protein abundance measurements in C . elegans [59] for four main reasons: 1 ) due to technical experimental challenges , protein abundance measurements in C . elegans are substantially more robust than protein half-life measurements; 2 ) on a proteome-wide scale , protein abundance is more strongly conserved across yeast species than protein half-life [49] , suggesting that final protein levels tend to be constrained across organisms , while regulation of the metabolic pathways that contribute to protein abundance may vary; 3 ) protein abundance is , at least partially , a function of translation efficiency and protein half-life; and 4 ) the parameters underlying the translation efficiency method ( namely the “s-vector” , or the efficiency of wobble base pairing between tRNA isoacceptors ) were optimized for yeast [60] . Therefore , the nTE method may not be amenable to application in other organisms . In order to examine relationships between maximum local composition and protein abundance , we first determined the proteome distribution of C . elegans proteins as a function of maximum local composition for each amino acid . The C . elegans-specific proteome distributions ( S2 Fig and S2 Table ) were overall quite similar to the yeast proteome distributions ( Fig 2 ) . However , the maximum local composition for S and N appear to be slightly more constrained in C . elegans ( indicated by contraction of the shoulder to lower maximum compositions ) , while G , P , and T achieve slightly higher maximum local compositions , indicating relaxed constraints on local enrichment of these residues . These results are consistent with previous observations noting both shared and organism-specific homopolymeric repeat signatures or bulk proteome compositions across proteomes from different organisms [4 , 38 , 40 , 41 , 44–46 , 61] . As observed in yeast , progressive compositional enrichment results in a transition from higher to lower median abundance for the majority of amino acids with a clear trend ( C , F , I , M , N , P , S , W , and Y; Fig 6 ) . Furthermore , all three amino acids ( A , G , and V ) that exhibit a transition from lower to higher median abundance upon progressive enrichment in yeast exhibit the same trend in C . elegans as well . Indeed only one amino acid with a clear transition in protein abundance upon local enrichment differs between C . elegans and S . cerevisiae: in yeast , local K enrichment is associated with mixed effects on protein abundance ( depending on the degree of K enrichment ) , whereas in C . elegans local K enrichment is weakly ( yet consistently ) associated with higher protein abundance . Therefore , nearly identical residue-specific relationships are observed between local amino acid enrichment and protein abundance in a more complex eukaryote . An important advantage of approaching LCDs from a composition-centric perspective is the ability to examine relationships between amino acid composition and protein outcomes without appealing to pre-defined thresholds of statistical amino acid bias [11] or sequence complexity [1 , 10] , which may not reflect biologically-relevant thresholds . Indeed , the transitions observed in the median translation efficiencies , protein abundances , and protein half-lives often occur at surprisingly mild levels of compositional enrichment , suggesting that these trends may be observed even in the absence of classically-defined statistically-biased or low-complexity domains . Statistical amino acid bias conceptually parallels our investigation of compositional enrichment , and has been used to investigate the functions of proteins with statistically-biased domains [11 , 12] . To examine whether compositional enrichment may be linked to biologically-relevant effects on protein metabolism independently of statistically-biased domains , a conservative bias threshold was employed to define statistically-biased domains using previously developed methodology [12] ( also , see Methods ) . Proteins with statistically-biased domains were then filtered from the yeast proteome ( n = 866 statistically-biased proteins for the yeast translated proteome of sequences ≥ 30 residues in length ) . However , even in the absence of statistically-biased domains , compositional enrichment resulted in robust trends in translational efficiency , protein abundance , and protein half-life that re-capitulated those originally observed ( S3–S5 Figs ) . This suggests that compositional enrichment affects protein metabolism at thresholds preceding those required for classification as statistically-biased by alternative methods . The SEG algorithm , by default , employs substantially more relaxed criteria when classifying protein domains as low-complexity [1] . Indeed , of the 5 , 901 proteins of length ≥30 amino acids in the translated ORF proteome , 4 , 147 proteins contain at least one LCD , which is consistent with previous estimates [3] . Nevertheless , despite a large reduction in proteome size , many of the trends in protein metabolism are discernible even when all proteins with a SEG-positive sequence are filtered from the proteome ( S6–S8 Figs ) . This suggests that compositional enrichment exerts biologically relevant effects even among non-LCD-containing proteins . Proteins containing homopolymeric amino acid repeats ( often defined as five or more identical amino acids in succession ) , were recently reported to have lower translation efficiency , lower protein abundance , and lower protein half-life when compared to proteins without homopolymeric repeats [37] . Homopolymeric repeats are effectively short sequences of maximum possible single-amino acid density . Therefore , proteins with homopolymeric repeats are expected to be disproportionately common among compositionally enriched domains , raising the possibility that the trends observed in the present study have been mis-attributed to compositional enrichment alone . To examine this possibility directly , the relationship between compositional enrichment and nTE , abundance , and half-life was re-evaluated for a filtered proteome that excludes all proteins containing at least one homopolymeric repeat ( n = 755 proteins excluded ) . While exclusion of these proteins preferentially reduces the sample sizes at higher compositional enrichment percentages , the absence of homopolymeric repeat proteins has little effect on the trends in nTE , abundance , and half-life as a function of compositional enrichment ( S9–S11 Figs ) . This does not definitively rule out the possibility that homopolymeric repeats may , in some way , specifically affect translation efficiency , abundance , and half-life . However , since homopolymeric repeats per se are not absolutely required , the effects of homopolymeric repeats may instead be explained simply by local compositional enrichment . Collectively , these results suggest that compositional enrichment affects translation efficiency , protein abundance , and protein half-life at thresholds preceding those required for classification as low-complexity or statistically-biased by traditional methods . It is worth noting that in the course of eliminating proteins with classically-defined low-complexity , statistically-biased , or homopolymeric domains , proteins with multiple distinct domains strongly enriched in different amino acid types , or with single domains strongly enriched in more than one amino acid , are eliminated from the proteome before re-evaluation . Therefore , the trends in protein metabolism observed upon enrichment of a given amino acid are not due to confounding effects of domains strongly enriched in other amino acids occurring within the same protein sequences . Local enrichment of a single amino acid can dramatically influence the physicochemical properties of a given protein domain [24] . In a cellular context , these physicochemical properties likely influence interactions between proteins and surrounding molecules , including other proteins . To examine whether local compositional enrichment affects protein-protein interactions , we explored relationships between enrichment for each of the amino acids and protein-protein interaction promiscuity ( defined as the number of unique interacting partners per protein ) . Proteins found in a range of high-percent composition-bins for most amino acids ( A , D , E , G , K , N , P , Q , R , and V ) are associated with significantly more interacting partners relative to all other proteins ( Fig 7 ) , suggesting that these domains are relatively promiscuous . Additionally , proteins with mild enrichment for select hydrophobic residues ( I , L , and M ) are generally associated with more interacting partners , although fewer comparisons reach statistical significance ( blue or red dots ) . These results are consistent with previous reports that , as a single class , proteins with LCDs or homopolymeric repeats tend to have more protein-protein interaction partners [16 , 37] . However , proteins in a range of high-percent composition-bins for each of the aromatic residues ( F , W , and Y ) are associated with significantly fewer interacting partners relative to other proteins , suggesting that aromatic residues tend to lack the interaction promiscuity observed at higher percent compositions for other amino acids . Furthermore , proteins with moderate to high local C content and proteins with extremely high maximum local S or T content are also associated with significantly fewer interacting partners relative to other proteins , suggesting that these domains are relatively non-promiscuous as well . This is particularly interesting , given that these trends were not observed upon enrichment for other polar residues . Again , this highlights the potential pitfall of grouping amino acids with related physicochemical properties into a single category . Collectively , these results indicate that protein-protein interaction promiscuity varies for proteins with high compositional enrichment in a residue-specific manner . Previous studies have attempted to associate proteins containing LCDs , statistically-biased domains , and homopolymeric repeats with particular cellular functions [12 , 16–18 , 34 , 37] . However , one important consideration when inferring relationships between proteins with LCDs and cellular functions , for example , is the prevalence of proteins with multiple LCDs [3] , and of LCDs strongly enriched in more than one amino acid type [11 , 14 , 18 , 36] . Therefore , attempts to associate cellular functions to specific LCD types , without controlling for other LCDs within the same protein sequences , risk mis-attributing functions to unrelated protein features [12 , 14 , 34 , 36] . While multiple LCDs within the same protein ( or multiple amino acid types enriched within the same LCD ) may cooperate to generate novel structures or functions , this complicates interpretation of the role of each individual amino acid type within LCDs . Furthermore , because some types of LCDs are more common than others , general attempts to associate cellular functions with LCDs , statistically-biased domains , or homopolymeric repeats likely reflect the functions associated with only the most common types when considered as a single , unified class [16 , 37] . Therefore , definitive assignment of cellular functions to each individual class of LCD necessitates exclusion of proteins with other types of LCDs . In order to minimize possible confounding effects introduced by proteins with multiple regions enriched in different amino acid types , a modified version of the initial calculation performed by the SEG algorithm ( namely , the Shannon entropy; see Methods ) was employed to define proteins with only a single type of compositionally-enriched domain ( CED ) . In an effort to incorporate our results ( which indicate that compositional enrichment may exert biologically-relevant effects at compositions preceding the SEG algorithm threshold ) into our definition of single-CED proteins , percent composition bins for which at least 75% of the residing proteins contained a SEG-positive sequence ( as defined above ) were pooled to generate a single list of CED-containing proteins for each amino acid . Proteins that contain multiple types of CEDs were then removed from the dataset , resulting in a non-redundant set of proteins with only one type of CED . Importantly , this method captures the exclusion of proteins containing more than one type of CED , as well as proteins with CEDs strongly enriched in more than one amino acid type . Gene Ontology ( GO ) term analysis was performed separately for each window size within each single-CED category . For each type of CED , there is strong overlap in the enriched GO terms across the range of window sizes , suggesting that the associations between functions and residue-specific CEDs are not strongly length-dependent at this scale . Therefore , for simplicity of interpretation , significantly enriched GO terms for each window size were pooled to generate a single non-redundant list of enriched GO terms for each CED type . Removal of proteins with multiple types of CEDs reveals a remarkable degree of specialization for CEDs of different types ( Fig 8 , and S3 Table ) , which is often not observed for CEDs when considered as a single category or when multi-CED proteins are not excluded . For example , L-rich proteins are predominantly associated with functions at the ER and vacuole membranes , whereas I-rich proteins are more strongly associated with carbohydrate transport at the plasma membrane . A-rich proteins are associated with a variety of processes or cellular components , including translation , protein kinase activity , the cell wall , and carbohydrate/alcohol catabolism . N-rich proteins are strongly ( and perhaps exclusively ) associated with functions related to transcription , whereas Q-rich proteins appear to be more weakly associated with transcription and , instead , are associated with a larger variety of functions including endocytosis , mating projection of the membrane , and response to glucose . Finally , although yeast cell wall proteins are often radically S/T-rich , after controlling for co-enrichment of S and T in the same proteins , S-rich proteins are more strongly associated with membrane-related processes ( cell wall , cellular bud tip , cellular bud neck , mating tip projection , etc . ) , protein kinase activity , and transcription , whereas T-rich proteins tend to be associated with nucleic acid binding and helicase activity , with fewer associations with membrane-related processes . Therefore , after controlling for the presence of multiple CEDs within the same proteins , specialized functions emerge even among commonly grouped amino acids . Furthermore , CEDs enriched in some amino acids share functions despite the removal of multi-CED proteins , suggesting some degree of co-specialization . For example , D- , E- , and K-rich CEDs were each associated with functions in the nucleus/nucleolus , including ribosomal RNA processing , nucleic acid binding , transcription , and histone/chromatin binding . Intriguingly , intrinsically disordered domains with opposite net charges ( along with other charged macromolecules such as nucleic acids and polyADP-ribose ) can drive phase separation or complex coacervation in the nucleus [62–64] . It is possible that these domains , along with nucleic acids and other polyionic molecules , may participate in nuclear processes via dynamic electrostatic association with these or other membraneless assemblies . By contrast , H-rich CEDs are associated with processes related to zinc ion transport and regulation . There were no GO terms significantly associated with R-rich CEDs . However , compositional enrichment for R appears to be constrained , as evidenced by the sharp decline in the number of proteins with R-rich domains toward higher maximum local percent compositions ( see Fig 2 ) , which may be further impacted by the removal of proteins with other types of CEDs . In summary , when examined as separate classes , different types of CEDs can have overlapping or specialized roles in the cell . The molecular specialization observed for CEDs indicates that proteins with enrichment of particular residues may localize to particular subcellular compartments in order to execute their specialized functions . Furthermore , protein quality control factors can differ between subcellular compartments ( for review , see [65] ) , which may contribute to composition-dependent differences in protein metabolism . Therefore , we applied a bottom-up approach to infer the composition profiles associated with the major subcellular compartments ( see Methods ) . Largely aqueous subcellular compartments are almost exclusively associated with proteins containing domains enriched in charged residues , polar residues , and proline ( Fig 9; see also S12 Fig ) . However , differences in compositional enrichment profiles are apparent even among related aqueous compartments . For example , significant associations with charged , Q , or N residues reach more extreme percent compositions in the nucleus , whereas as significant associations with P enrichment reach higher percent compositions in the cytoplasm . By contrast , the highly membraneous internal organelles ( e . g . the endoplasmic reticulum and Golgi apparatus ) are predominantly associated with enrichment of hydrophobic or aromatic residues ( Figs 9 and S13 ) . The yeast vacuole is also associated with composition profiles resembling those of membraneous compartments , with additional weaker associations with S and C enrichment . Few weak associations are observed for mitochondria . The yeast cell wall is strongly associated with S enrichment ( likely related to its ability to be glycosylated ) , with additional moderate associations with T and A enrichment , and a weak association with mild V enrichment ( Figs 9 and S14 ) . As expected , the plasma membrane is associated with enrichment for a variety of hydrophobic and aromatic residues . However , the plasma membrane is also significantly associated with enrichment of a select subset of polar residues ( namely C , G , S , and T ) , further corroborating the specialized roles observed for these CEDs at the outer membrane . Indeed , G-rich CEDs are significantly associated with amino acid transport ( see S3 Table ) , and S- or T-rich CEDs of the plasma membrane could have overlapping functions or interactions with S- and T-rich CEDs of the cell wall . Together , these observations indicate that subcellular compartments may tolerate or prefer proteins with specific types of CEDs . Recent observations indicate that a variety of Q/N-rich and G-rich domains can form highly dynamic protein-rich droplets in aqueous environments [25–30] , a process referred to as liquid-liquid phase separation . These types of LCDs are prevalent among components of membraneless organelles such as stress granules and P-bodies [23] . Furthermore , stress granules and P-bodies share many properties with protein-rich liquid droplets formed in vitro , suggesting that the fundamental biophysical properties of these domains are related to the formation of membraneless organelles in vivo . However , while amino acid composition is acknowledged as a critical determinant of this behavior , the precise compositional requirements associated with membraneless organelles remain largely undefined . Therefore , we also applied our bottom-up approach to infer the compositional enrichment profiles associated with protein components of stress granules and P-bodies ( as defined in [66] ) . Stress granules and P-bodies have overlapping protein constituents and can exchange protein components [67 , 68] , suggesting that they are closely related yet distinct organelles . Accordingly , we observe both shared and unique features in the composition profiles associated with stress granule and P-body proteins ( Fig 10 ) . As expected , both stress granules and P-bodies are strongly associated with proteins containing Q-rich or N-rich domains . For example , minimum Q or N compositions significantly associated with stress granules range from ~15–100% at small window sizes ( ≤30 amino acids ) and ~10–30% at large window sizes ( ≥80 amino acids ) , although these values vary slightly depending on window size and residue . Similarly , minimum Q or N compositions significantly associated with P-bodies range from ~15–100% at small window sizes and ~10–40% at larger window sizes . In addition to the commonly appreciated link between stress granule/P-body components and Q/N-rich domains , we identify and define a variety of currently underappreciated compositional features common to stress granule and P-body components . Components of both stress granules and P-bodies are strongly associated with P-rich domains , weakly associated with K-rich domains , and very weakly ( yet significantly ) associated with Y-rich domains . Furthermore , while both stress granules and P-bodies are associated with proteins containing G-rich domains , stress granule components are associated with a much broader range of G enrichment , suggesting that G enrichment may be a more characteristic feature of stress granules than P-bodies . This is particularly striking in light of recent observations indicating that high glycine content helps maintain the liquid-like characteristics of phase-separated droplets and prevents droplet hardening in vitro [69] . Additionally , some compositional features are unique to either stress granules or P-bodies . For example , stress granule constituents are significantly associated with A-rich , M-rich , E-rich , and R-rich domains , whereas P-body constituents exhibit little or no preference for these compositional features ( a key role for arginine in the phase separation of stress granule-associated proteins was also recently reported [69] ) . By contrast , P-body components are weakly associated with H-rich domains , whereas stress granule components are not enriched among proteins containing H-rich domains . To our knowledge , this represents the first attempt to systematically define the range of amino acid compositions associated with membraneless organelles such as stress granules and P-bodies . These observations suggest that components of related , membraneless organelles have overlapping yet distinct compositional preferences . It is possible that shared compositional features facilitate the physical interactions between stress granules and P-bodies and allow for the exchange of components , while differences in compositional features facilitate their ability to function as independent organelles .
Protein domains categorized as low-complexity , statistically-biased , or homopolymeric encompass broad , heterogeneous classes of sequences with diverse physical properties and cellular functions . These domains can play important roles in normal and pathological processes . However , challenges in categorizing proteins on the basis of sequence complexity or statistical bias have thus far precluded a complete , proteome-wide view of the effects of these domains on protein regulation and function . Here , we adopt an alternative , unbiased approach to examine proteome-wide relationships between local amino acid enrichment and the birth , abundance , functions , subcellular localization , and death of proteins . For nearly all amino acids , progressive local enrichment corresponds to clear transition thresholds with regard to translation efficiency , protein abundance , and protein half-life . Transition thresholds ubiquitously occurred at compositions preceding those required for classification as low-complexity or statistically-biased by traditional methods , indicating that our observed transition thresholds more closely reflect biologically-relevant composition criteria . Protein sequences can range from perfectly diverse ( i . e . a completely homogeneous mixture of amino acids with maximal spacing between identical amino acids ) to lacking any diversity ( i . e . homopolymeric sequences ) . While homopolymeric regions represent an extreme on this spectrum and can influence protein metabolism [37] , classically defined homopolymeric regions are not absolutely required for these effects ( see S9–S11 Figs ) . This suggests that compositional enrichment may affect protein metabolism even upon some degree of primary sequence dispersion ( i . e . greater linear spacing between identical amino acids ) . Defining the limits of this dispersion may shed additional light on the relationship between amino acid composition and protein metabolism . An advantage of assessing compositional enrichment ( as opposed to sequence complexity ) is the ability to distinguish the effects of compositional enrichment for each amino acid type . The nature of the trends in translation efficiency , protein abundance , and protein half-life depend on the amino acid enriched in the protein sequences , indicating that local enrichment of different amino acids can have opposite effects . This highlights a key limitation when considering low-complexity , statistically-biased , or homopolymeric domains as a single class–grouping domains composed of radically different amino acids effectively skews any trends observed toward those of the most common type and , in some cases , can completely mask the effects of less common low-complexity , statistically-biased , or homopolymeric domains . Furthermore , even grouping these domains on the basis of common physicochemical properties can introduce the same complication . This is exemplified by the non-aromatic hydrophobic amino acids; while I-rich , L-rich , and M-rich domains are associated with poor translation efficiency , low abundance , and rapid degradation rate , A-rich and V-rich domains are associated with high translation efficiency , high abundance , and slow degradation rate . Additionally , the cellular functions associated with domains enriched in hydrophobic residues tend to differ; L-rich domains are predominantly associated with the ER or vacuole membrane , whereas I-rich domains are predominantly associated with carbohydrate transport at the plasma membrane . Similarly , N-rich domains are strongly associated with transcription-related processes , whereas Q-rich domains are more strongly associated with endocytosis and other processes in the cytoplasm . While there is some overlap between these two groups , this suggests that domains enriched in remarkably similar amino acids may yet be favored for specialized roles in the cell . Finally , a bottom-up application of our composition-centric algorithm to membraneless organelles provides the first step in defining the distinct compositional profiles associated with each type of organelle . We find that even closely related and physically interacting organelles are associated with discernible differences in compositional enrichment , which may relate to differences in their properties , regulation , and function in vivo . It is important to note that , while the observed trends in compositional enrichment are significantly associated with stress granule proteins or P-body proteins as respective groups , these features may not be absolutely required for individual proteins to be incorporated into stress granules and/or P-bodies . It is possible , for example , that two proteins possessing non-overlapping subsets of the associated compositional features may still be recruited to stress granules , and that some stress granule proteins may be recruited for reasons entirely distinct from compositional enrichment ( e . g . via RNA-binding domains ) . One might even imagine that differences in compositional features , while still allowing recruitment to stress granules and/or P-bodies , could favor differences in the dynamics of individual protein components ( e . g . the kinetics of entry/exit , dwell time , the strength of the interactions , or the depth of penetration within the stress granule/P-body ) . Therefore , while the associated composition ranges observed here are collectively enriched among proteins associated with these membraneless organelles , each individual protein need not possess all of the compositional features simultaneously in order to function as a stress granule or P-body protein . While a great deal of attention is rightfully devoted to understanding the effects of primary amino acid sequence on protein fates ( including folding , regulation , and functions ) , amino acid composition is increasingly believed to drive a variety of cellular and molecular processes . Here , we have developed an approach to examine relationships between local compositional enrichment and protein fates for each of the canonical amino acids , in the absence of a priori assumptions or pre-defined thresholds . Our results provide a coherent , proteome-wide view of the relationships between compositional enrichment and the fundamental aspects of protein life cycle , subcellular localization , and function in model eukaryotic organisms .
Protein sequences were parsed using FASTA sequence parsing module from the Biopython package [70] . For each amino acid in the set of 20 canonical amino acids , each protein in the translated ORF proteome ( latest release from the Saccharomyces Genome Database website , last modified 13-Jan-2015 ) or the ORF coding sequences ( organismID:UP000001940_6239 , release date 23-May-2018 downloaded from the UniProt website ) was scanned using a sliding window of defined size ( ranging from 10 to 100 amino acids , in increments of 10 ) . The percent composition of the amino acid of interest ( AAoI ) is calculated for each window , and the protein is sorted into bins based on the maximum percent composition achieved for the AAoI ( ranging from 0 to 100 percent composition in 5 percent increments ) . Analyses were performed for all possible AAoI , window size , and percent composition combinations . Translation efficiency for each gene was estimated using the normalized translation efficiency ( nTE ) scale [55] , which is based on tRNA gene copy number , codon-anticodon wobble base-pairing efficiency , and transcriptome-wide codon usage . However , the original nTE algorithm plots all nTE values for each codon to generate a separate translation efficiency profile for each gene . In order to condense translation efficiency information to a single value for each gene ( in a manner analogous to the tRNA adaptation index; [60] ) , the geometric mean of nTE values across the transcript was calculated as nTEgene= ( ∏k=1lsnTEiks ) 1ls ( 1 ) where nTEiks represents the translation efficiency value of the ith codon defined by the kth triplet in nucleotide sequence s , and ls represents the length of the nucleotide sequence excluding stop codons . Therefore , nTE values reported in the current study represent whole-gene nTE values . nTE analyses were performed using an in-house Python script . The Shannon entropy of each sequence was calculated as SE=−∑i=1N=20niL ( log2niL ) ( 2 ) where N represents the size of the residue alphabet ( N = 20 , for the canonical amino acids ) , ni represents the number of occurrences of the ith residue within the given sequence window of length L . For comparison with established measures of sequence complexity , we defined low-complexity domains by using the default window size ( 12 amino acids ) and Shannon entropy threshold ( SE ≤ 2 . 2bits ) used in the first pass of the SEG algorithm to initially identify LCDs [1 , 10] . In the SEG algorithm , the complexity state vector used to calculate the Shannon entropy is blind to the amino acid composition ( i . e . the ni values in Eq 2 are not attributed their respective amino acids ) . Therefore , when indicated , in order to distinguish LCDs on the basis of the predominant amino acid , sequences for which the SE ≤ 2 . 2bits and nAAoI ≥ nmax within the complexity state ( indicating that the AAoI is a major contributor to the sequence’s classification as an LCD ) were assigned to the corresponding amino acid category ( e . g . A-rich LCDs , C-rich LCDs , etc . ) . Single-LCD/CED proteins are proteins classified as LCDs or CEDs that do not appear on multiple amino acid-specific LCD/CED lists . Statistical amino acid bias was calculated as described in [12] . Briefly , the lowest probability subsequence for each protein was determined by exhaustively scanning proteins with window sizes ranging from 25 to 2500 amino acids . For each window , the subsequence bias probability ( Pbias ) was defined as Pbias=[w ! n ! ( w−n ) ! ]× ( fx ) n× ( 1−fx ) w−n ( 3 ) where w denotes the window size , n denotes the number of occurrences of the amino acid of interest within the subsequence , and fx denotes the fraction of the amino acid of interest in the yeast translated proteome . The lowest probability subsequence for each protein is the subsequence with the lowest Pbias . A suitable threshold to define statistically-biased proteins within the yeast protein was determined as previously described [12] , except that more relaxed criteria were used in order to include additional proteins with less extreme biases . Briefly , the Pbias corresponding to the lowest probability subsequence ( Pmin ) for each protein was plotted on a log-log plot against whole-protein sequence length . A line was fitted , then the y-intercept was decreased until only 15% of the proteome had Pmin values below the line ( previous analyses used a more stringent cutoff of 10% to define statistically-biased proteins [12] ) . Additionally , a length-independent threshold was designated as the Pmin value at which 15% of the proteome had smaller absolute Pmin values . This threshold was used when it was less than the Pmin threshold given by the length-dependent method to avoid unreasonably relaxed bias criteria for small protein sequences . Amino acid bias was calculated using values from the translated orf proteome only , and implemented via an in-house Python script with pre-computed look-up tables for computational efficiency . Proteins containing homopolymeric sequences were defined simply as any protein with a subsequence of five or more contiguous residues of the same amino acid , as previously described [37] . Yeast protein abundance values ( in average number of molecules per cell per protein ) were obtained from [48] ( n = 5 , 391 ) . Protein abundance values for C . elegans were obtained from [59] . Yeast protein half-life data were obtained from [49] . For simplicity of interpretation , only proteins with unambiguous , non-zero half-life or abundance values were included in the datasets . Proteins listed on separate lines with identical half-life or abundance values were retained , whereas protein half-life or abundance measurements assigned to more than one protein on the same line were excluded ( these were often highly homologous genes , suggesting that the measurement could not be unambiguously assigned to one of the proteins ) . Furthermore , all proteins corresponding to “low-confidence” measurements in the half-life dataset were excluded ( see [49] for criteria ) . n = 3 , 525 for the filtered yeast half-life dataset , and n = 5 , 952 for the filtered C . elegans protein abundance dataset . For all AAoI/window size/percent composition bins , the distribution of nTE , abundance , or half-life values for proteins included in the given bin was compared to the distribution of the respective values of all proteins excluded from the given bin . Statistical significance was estimated using a two-sided Mann-Whitney U test ( also referred to as the Wilcoxon rank-sum test; refer to Supplemental Experimental Procedures from [47] for a detailed description and rationale ) . Where indicated , p-values were adjusted within each window using the Bonferroni correction method for multiple hypothesis testing . All statistical tests were performed using modules available in the SciPy package with default settings , unless otherwise specified . All plots were generated using Matplotlib or Seaborn modules . GO term enrichment tests were performed using the GOATOOLS package ( version 0 . 7 . 9 ) [71] for each set of proteins contained in a given amino acid/window size/percent composition bin . For each test , the set of background proteins was defined as all proteins from the translated ORF proteome of sequence length greater than or equal to the given window size . All reported p-values were adjusted using the Bonferroni correction during GO term association . To evaluate the compositional enrichment profiles associated with GO terms related to subcellular compartments , we applied a minimum-threshold-scanning approach to all partitioned proteomes . For each AAoI , window size , and percent composition bin , all proteins with maximum local compositions greater than or equal to the current percent composition under consideration are pooled and evaluated for possible enriched GO terms . This effectively evaluates possible GO term enrichment iteratively with increasing maximum local composition criteria . GO term results were subsequently evaluated for significant enrichment of a single GO term describing each subcellular compartment ( or two related GO terms , “outer membrane” and “plasma membrane” , in the case of the plasma membrane ) . p-values were further adjusted within each window size using the Bonferroni correction method . Similar analyses were performed for the sets of experimentally-defined stress granule ( n = 83 ) and P-body ( n = 52 ) proteins [66] . Specifically , a minimum-threshold-scanning approach was applied to all partitioned proteomes . For each AAoI , window size , and percent composition bin , all proteins with maximum local compositions greater than or equal to the current percent composition under consideration are pooled . Significant enrichment of experimentally-defined stress granule or P-body proteins within each pool of proteins was evaluated using Fisher’s exact test ( p < 0 . 05 ) . | Low-complexity domains in protein sequences are regions that are composed of only a few amino acids in the protein “alphabet” . These domains often have unique chemical properties and play important biological roles in both normal and disease-related processes . While a number of approaches have been developed to define low-complexity domains , these methods each possess conceptual limitations . Therefore , we developed a complementary approach that focuses on local amino acid composition ( i . e . the amino acid composition within small regions of proteins ) . We find that high local composition of individual amino acids is associated with pervasive effects on protein metabolism , subcellular localization , and molecular function on a proteome-wide scale . Importantly , the nature of the effects depend on the type of amino acid enriched within the examined domains , and are observable in the absence of classically-defined low-complexity ( and related ) domains . Furthermore , we define the compositions of proteins involved in the formation of membraneless , protein-rich organelles such as stress granules and P-bodies . Our results provide a coherent view and unprecedented resolution of the effects of local amino acid enrichment on protein biology . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"sequencing",
"techniques",
"protein",
"abundance",
"membrane",
"proteins",
"fungi",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"proteins",
"proteomics",
"molecular",
"biology",
"cell",
"membranes",
"yeast",
"biochemistry",
"eukaryota",
"cell",
"biology",
"proteomes",
"protein",
"domains",
"biology",
"and",
"life",
"sciences",
"protein",
"sequencing",
"organisms"
]
| 2018 | Proteome-scale relationships between local amino acid composition and protein fates and functions |
The characterization of the blood virome is important for the safety of blood-derived transfusion products , and for the identification of emerging pathogens . We explored non-human sequence data from whole-genome sequencing of blood from 8 , 240 individuals , none of whom were ascertained for any infectious disease . Viral sequences were extracted from the pool of sequence reads that did not map to the human reference genome . Analyses sifted through close to 1 Petabyte of sequence data and performed 0 . 5 trillion similarity searches . With a lower bound for identification of 2 viral genomes/100 , 000 cells , we mapped sequences to 94 different viruses , including sequences from 19 human DNA viruses , proviruses and RNA viruses ( herpesviruses , anelloviruses , papillomaviruses , three polyomaviruses , adenovirus , HIV , HTLV , hepatitis B , hepatitis C , parvovirus B19 , and influenza virus ) in 42% of the study participants . Of possible relevance to transfusion medicine , we identified Merkel cell polyomavirus in 49 individuals , papillomavirus in blood of 13 individuals , parvovirus B19 in 6 individuals , and the presence of herpesvirus 8 in 3 individuals . The presence of DNA sequences from two RNA viruses was unexpected: Hepatitis C virus is revealing of an integration event , while the influenza virus sequence resulted from immunization with a DNA vaccine . Age , sex and ancestry contributed significantly to the prevalence of infection . The remaining 75 viruses mostly reflect extensive contamination of commercial reagents and from the environment . These technical problems represent a major challenge for the identification of novel human pathogens . Increasing availability of human whole-genome sequences will contribute substantial amounts of data on the composition of the normal and pathogenic human blood virome . Distinguishing contaminants from real human viruses is challenging .
Research on the human microbiome has been primarily directed to the prokaryotic composition of the human microflora . Because most of the analyses use 16S rRNA gene-based amplification , the viral content has been rarely captured in large-scale microbiome studies . In contrast , analysis of the whole human genome by next-generation sequencing is an exercise in metagenomics: after mapping sequencing reads to the human reference genome , there is a significant proportion ( generally 5% of all sequence data ) that is left uncharacterized [1] . Bacterial but also archaea , non-human eukaryotic and viral sequences are thus a by-product of the sequencing of the human genome . Previous studies of the human virome have addressed the viral component of the gut flora [2–4] and skin [5–7] , with particular attention to the very abundant bacteriophages [7 , 8] . A thorough review has been published recently [9] . Many viruses are present in peripheral blood—in particular , members of the Herpesviridae and Anelloviridae families are identified in the absence of disease . Metagenomic studies on blood have identified great genetic diversity of anelloviruses [10–12] . Metagenomic studies also lead to the identification of novel RNA viruses—for example the identification of two rhabdoviruses [13] . Other viral sequences in the blood of healthy individuals are related to members of the Picornaviridae , Poxviridae , Flaviviridae , and Phycodnaviridae families ( reviewed in [9] ) . Finally , a number of viruses , prominently retroviruses , are integrated in the human genome as provirus , while others may integrate occasionally or accidentally [14] . The study of the human virome is particularly relevant in the context of current discussions of next-generation sequencing for surveillance of viruses in blood and for transfusion safety [11 , 15 , 16] . Only viruses that are both pathogenic and transfusion-transmissible are routinely tested for and excluded from blood-derived products . Rejecting all virus-infected donations irrespective of pathogenicity would not be sustainable as most donors are anellovirus positive . The time required to develop and implement specific virus nucleic acid tests to emerging viral pathogens in the blood supply has greatly improved as seen with the response to recent Zika virus outbreak [17] . Exclusionary steps for viruses can also vary depending on the recipients in whom sequelae may vary in severity such as the use of parvovirus B19-reduced plasma pool to derive products for pregnant B19 seronegative women and immunocompromised patients . Seasonal variation in virus prevalence can also affect when testing is implemented such during mosquitos season for West Nile virus RNA . As the rate of human genome and associated DNA viruses sequenced from blood continues to grow data a baseline will be available to compare rates of infections with various DNA viruses , as described in this study , to that in future populations . There are many open questions on what could be considered a “normal” human blood virome . Recently , the National Heart , Lung , and Blood Institute of the National Institutes of Health convened a working group on the microbiome that identified studies of the human virome a key priority [18] . The present study aims at establishing the DNA virome in over 8 , 000 individuals participating in a large-scale sequencing effort of the whole human genome [1] . A careful definition is key to diagnosing infections , to understanding the role of the virome in chronic disease , and for settling claims for the identification of new viral species in humans .
We sequenced the genomes of 8 , 240 individuals . On average , each sequencing reaction generated 1 billion reads . The total input approached 1 PB . The majority ( 95% ) of reads were successfully mapped ( S1 Fig ) to the human reference genome GRCh38 ( hg38 ) . Among the remaining reads , similarity search assigned 9% to non-reference human sequences , 1% to other primate sequences , 0 . 2% to bacteria , and 0 . 01% to viruses . The bulk of unmapped reads mainly represents reads with multiple mappings to the human reference , but also microbial genomes absent in the database , and low quality reads . We launched 0 . 5 trillion similarity searches against the NCBI viral genomes ( Fig 1 ) . This step mapped sequences to 94 viruses ( S1 Table ) . Samples carried a median of about 400 , 000 viral reads . However , the majority corresponded to phiX174 , used as spike-in control in the sequencing process , or to human endogenous retroviruses ( HERV ) that are discarded during alignment ( Fig 2 ) . Samples that carried phiX174 were also enriched in reads from multiple phages , which we interpret as contamination of the commercial preparation of phiX174 . Epstein-Barr virus ( EBV , HHV4 ) reads were abundant in sequences of the human reference genome NA12878 ( www . nist . gov/programs-projects/genome-bottle ) and in a subset ( n = 148 ) of participant samples where the input DNA material was , in retrospect , from cell lines that use EBV in the process of cell immortalization . Furthermore , we observed cross-contamination from the EBV content in the human genome immortalized cell line NA12878 to other samples on the same flow cell ( S2 Fig ) . The human reference genome NA12878 is used as standard reagent in sequencing laboratories . In a second step , viral candidate reads were searched against a comprehensive database of viruses , vectors , bacteria , archaea , human , and other eukaryotes to reduce false-positive matches from the initial search . We identified 11% reads that would result from plasmid sequences engineered with sequences such as viral promoters . Therefore , we removed from downstream analysis reads of phiX174 and associated contaminant phages , HERVs , reads from samples containing EBV used in cell immortalization , and EBV reads from samples that were potentially contaminated and plasmids and vectors . Flow cells with high-titer samples of human papilloma virus ( HPV ) and parvovirus B12 contained other positive samples that were potential false positives ( S3 Fig ) . Single indexing , where the barcodes are embedded in one of the sequencing library adapters , comes with a risk of misidentification of sequences sharing flow cells [19] . The quality control steps are depicted in Fig 1 . We compared the sensitivity of detection of viruses using nucleotide-based search with individual reads versus using protein-based search after de novo assembly of reads into contigs and translation ( Fig 1 and S4 Fig ) . The mapping of single reads identified 19 human viruses . In contrast , contigs could only be assembled for 8 viruses because it required the presence of 1 to 4 orders of magnitude more viral reads in the sample ( S4 Fig ) . Overall , viruses were detected by both read mapping or contigs in 137 samples , and only by read mapping in 3 , 342 samples . Because of the low sensitivity of the approach using contigs , the study proceeded using individual reads . While it would have been ideal to perform a complete search of translated read-to-translated NCBI nt database using tools such as TBLASTX , this approach would be prohibiting in terms of computational demands . of translated read-to-translated NCBI nt database using tools such as TBLASTX , this approach would be prohibiting in terms of computational demands . Among the 94 different viruses identified in the study materials , we identified viral reads for 19 human viruses ( Fig 3 and Table 1 ) . Among the herpesvirus ( HHV ) , HHV7 was found in 20% , and EBV was identified in 14% of the individuals . Analysis of sequence diversity identified the presence of both EBV subtypes 1 and 2 . The estimated proportion was 80% for subtype 1 and 20% for subtype 2 , consistent with previous knowledge [20] . HHV6A and HHV6B were identified in 1 . 5% and 5% of individuals , respectively . We identified fewer individuals carrying sequences of other human herpesviruses: Herpes simplex 1 ( HSV1 ) , Cytomegalovirus ( CMV , HHV5 ) , and HHV8 . We identified a significant presence of anelloviruses ( Torque teno virus [TTV] and TTV-like mini virus [TLMV] ) in 9% of the individuals . Other viruses were identified in less that 1% of the study population ( Fig 3 and Table 1 ) . We took interest in the presence of sequence reads for papillomavirus ( 7 different types: 2 , 10 , 16 , 92 , 137 , 163 , and 179 ) in 17 individuals . Upon validation , we identified a cluster of individuals with the oncogenic type 16 in the same flow cell . We identified the wrong inclusion of a tumor sample in the analysis . This sample corresponded to a head and neck tumor containing large presence of papillomavirus 16 that led to contamination of samples sharing the same flow cell . Parvovirus B19 was identified in 10 individuals; however , four positive samples shared the flow cell with the sample with the highest load of viral copies ( > 300 million viral copies/100 , 000 cells ) and where thus classified as contaminants . We aimed at reconstructing viruses across many samples ( Fig 4 ) . The purpose of this step is to provide proof that the viral presence is confirmed by demonstrating broad and average coverage of each viral genome , and not the result of skewed accumulation of local reads—for example at CMV promoters in plasmids . It also offers a detailed view on viral polymorphism and subtypes . This was done for viruses with enough reads or present in numerous individuals , where we could reconstruct the viral genomes with significant coverage ( Fig 4 ) . For viruses where only a few reads could be identified , we checked them manually for unambiguous mapping . HHV6 can integrate in the human genome in telomeric regions and can be inherited through the germline [21] . We identified integrated HHV6A/B in 0 . 5% of the individuals . Fig 5 depicts the expected binomial distribution where samples with integrated copies have 100 , 000 viral copies per 100 , 000 human cells ( one integration event in every cell ) . The precision of this number attests to the highly quantitative nature of the sequencing protocol . Actual proof of integration was also obtained for most of those samples though the identification of chimeric reads or virus-host paired reads ( Fig 5 ) . In contrast , samples without integration have 3 to 4 orders of magnitude lower abundance . Other than the integration events of HHV6 –and the presumed events ( insufficient reads to identify the integration site ) for human immunodeficiency virus ( HIV ) and human T lymphotropic virus ( HTLV1/2 ) –we did not have direct proof for other integrated viruses . However , we identified two individuals carrying DNA sequence reads of RNA viruses , influenza and hepatitis C virus ( HCV ) . In the first individual , we observed 4 reads of influenza virus . The reads were mapped to different regions of the viral matrix genes ( M1 and M2 ) ( S5 Fig ) , as well as in the terminal read , a short plasmid tail representing the cloning site of common vector backbones . A possible explanation is that this individual received a DNA-based vaccine . In the second individual , we identified 18 HCV reads . The resulting sequence is similar to HCV clone from Pakistan , which coincides with the demographic information on the presumed carrier ( S5 Fig ) . An additional sample contained many paired-end chimeras between CMV and human chromosome 11 and 15 . Closer inspection revealed a lack of coverage of the CMV genome , with a large number of reads uniquely mapping to CMV regulatory elements used in expression vectors [22] . A similar situation was found in a sample that contained many reads of SV40 of plasmid origin . We identified a few viral sequences of Mollivirus in 8 individuals with a median of 2 reads per sample , Paramecium bursaria Chlorella virus in 3 individuals with a median of 2 sequence reads per sample , Apis mellifera filamentous virus in 2 individuals with a median of 2 sequence reads per sample , Melbournevirus in 2 individuals with a median of 3 sequence reads per sample , and Acanthamoeba polyphaga moumouvirus in 1 individual with 2 sequence reads . We observed the presence of occasional reads with correct match to animal retroviruses ( Fig 1 ) : Feline immunodeficiency virus and RD114 feline retrovirus , Ecotropic , Polytropic and Moloney murine leukemia virus , and Porcine endogenous retrovirus . The source of these viruses is likely to be through contamination of cell lines or the environment [23 , 24] . We identified in a single individual the presence of 8 reads ( abundance = 2 , 432 particles ) of a virus corresponding to the sewage-associated gemycircularvirus . This virus was also identified in transfusion plasma pools and clinical samples [16] , thus raising awareness for the possibility of gemycircularviruses infect humans or alternatively , reflecting contamination occurring during phlebotomy or plasma pool processing . We identified a few viral sequences of archaeal viruses ( Archaeal BJ1 virus and Halovirus ) in 4 individuals with a median of 9 reads per sample . There is debate in the literature whether these viruses should be referred to as phages [25] , and there is no sufficient information on whether archaea , and thus their viruses , may represent actual flora of humans [26] . Complete demographic information was available for 4 , 505 individuals . We observed a greater prevalence of circulating viruses in men than in women ( Fig 6 and S6 Fig ) . We also observed difference in viral prevalence in relation to age and ancestry ( S6 Fig ) . Deltaretroviruses were predominantly identified in individuals of African ancestry from different geographical locations . Twenty out of 22 human T-lymphotropic virus ( HTLV ) infections ( 90% ) were HTLV-2 . CMV , HHV6A and B and HHV7 were more prevalent in the younger groups , with higher loads of HHV7 identified in them ( Fig 6 ) . Statistical significant differences for demographic characteristics and viral prevalence or viral load are summarized in S2 Table . Overall , viral presence associated with age ( p-value = 5 . 6e-25 ) after adjustment for ancestry ( p-value = 1 . 3e-20 ) and sex ( p-value = 1 . 4e-9 ) ; ( S6 Fig ) .
The current work defines the human DNA blood virome in more than 8 , 000 individuals that we consider as representing a general population . The study leverages sequencing of the human genome that generates approximately 5% of reads ( the sequence of a fragment of the genome ) that do not map to the human reference genome . This large pool of reads primarily includes unmapped and repetitive human reads , bacterial reads , but also lesser numbers of sequences from archaea , eukaryotes , and viruses [1] . We identified 94 different viruses , including human DNA viruses , however , the pools of non-human reads are known to contain contaminant DNA from reagents [27 , 28] . The routine process of sequencing human DNA does not capture RNA viruses except through the identification of proviruses and other possible viral integration events . Among sequences that mapped to 94 viruses , we identified 19 human viruses in 42% of the study participants . In addition to a wide representation of human herpesviruses and anelloviruses , the study identified 7 different papillomavirus types , including the oncogenic type 16 , HIV , HBV , 3 different polyomavirus types and parvovirus B19 . These viruses generally correspond to those known to be highly seroprevalent in the human population [29] . Viral sequences in the study represent a concentration of two to millions of genome copies per 100 , 000 cells . We identified sequences of most members of the herpesvirus with the notable exception of Varicella-Zoster virus . This virus is easily identified in blood from immunosuppressed hosts and in immunocompetent subjects with active herpes zoster disease [30 , 31] . It is however reported absent in blood in the immunocompetent host [32] . We also observed papillomavirus reads in 0 . 2% of the study participants . Papillomavirus DNA was previously identified via PCR amplification in 8 . 3% ( 15/180 ) of healthy male blood donors [33] . The Merkel cell polyomavirus ( MCPyV ) , found in 0 . 55% of the study participants , is highly seroprevalent in the population [34] . MCPyV was reported in 22% of blood samples from healthy donors using PCR [35] . We also identified Trichodysplasia spinulosa polyomavirus ( TSPyV ) [36] , which is also seroprevalent in humans [37] . TSPyV viremia has been described , via PCR amplification , in immunosuppressed individuals but not in healthy controls [38] . The presence of viruses in blood products can be relevant for transfusion medicine . Currently , laboratory testing of donated blood prior to transfusion includes screening of HIV-1 and HIV-2 , HTLV-1 and 2 , HCV , HBV , West Nile virus , and Zika virus . The clinical impact , if any , of transmission of the highly prevalent GBV-C ( aka pegivirus A ) and of anelloviruses , is to be deciphered [39 , 40] . Parvovirus B19 [41] and other parvoviruses [42] are of concern to transfusion safety because these viruses are not routinely screened for and they lack a lipid envelope , rendering pathogen inactivation procedures less effective . The observation of other human DNA viruses in the study population—for example HPV , MCPyV , HHV8 and adenovirus—adds to the list of viruses that could be potentially transmitted via blood products [43] . The coverage ( 30X ) required for sequencing of the human genome [1] limits the ability to map integration events . This would rely on abundance of sequencing paired reads that encompass viral and human sequences . However , integration into the human genome was observed for HHV6A and B , known to occur in about 0 . 5% to 1% of humans [44 , 45] . Integration by RNA viruses ( other than retroviruses ) has been described occasionally [14] , and we were intrigued to identify one individual carrying few sequence reads of influenza virus that we attributed to the possible use of a DNA-based influenza vaccine ( because of the presence of a small plasmid fragment in the sequence ) . The second surprising event was the identification of multiple sequence reads of HCV matching to viral clones from Pakistan , in an individual from the same geographical origin . There has been discussion on the role of reverse transcriptase activity determining the accidental integration of viral RNA in the genome [46] , and specific to HCV , the occasional claim of integration [47] . Younger study participants were more likely to have human viruses identified in blood—which is consistent with the impact of seroconversion window at younger age . Differences in viral prevalence and type of virus varied also by ancestry: geography and local epidemiology may be the driving epidemiological factor . We observed an unexpected bias towards greater prevalence of circulating viruses in men than in women that remained significant after adjusting for the other demographic factors . There have been many descriptions on differences in prevalence , susceptibility to infection and disease severity across sex . The current thinking is that females tend to mount higher innate , cell-mediated , and humoral immune responses than males [48] . Next-generation sequencing is used for the discovery of new human pathogens—particularly in the setting of acute infection . Although we identified 94 different viruses , we found that large numbers of viral sequences represented contamination . Specifically , we observed a very significant presence of phage DNA associated with use of phage phiX174 used to allow real-time quality metrics during sequencing . Although there is a possibility that some phage DNA could translocate from the gut [49] , the presence of other phages and viruses each time that phiX174 was used is revealing of intrinsic contamination of the commercial phiX174 materials . Phage DNA can also derive from bacteria contaminating the reagents [27 , 50] . Beyond phages , there are reports of false-positive results and claims of viral pathogen discovery traced back to specific steps in the process of sequencing; for example , the identification of parvovirus-like sequences in nucleic acid extraction columns [51 , 52] or Moloney MuLV genome in cancer cell lines [53] . Therefore , the presence of a novel DNA virus in blood would require the use of numerous control experiments to exclude contamination . More generally , we identified animal retroviral sequences that likely reflect the contamination of cellular reagents or from environmental sources—a critical consideration given the past history of claims such as with Xenotropic murine leukemia virus-related retrovirus ( XMRV ) that was reported to be associated with prostate cancer and chronic fatigue syndrome . A massive effort was required to reverse those claims [54] . Finally , many reads were falsely attributed to viruses due to contamination with plasmid sequences that use viral regulatory cassettes . We evaluated the presence of the recently discovered giant viruses [55] . Our finding of a small number of reads in only 0 . 2% of the study population suggests that giant virus DNA is not a frequent finding in blood or that its detection also reflects reagent or laboratory contamination [56] . In addition , the presence of samples with high viral-titers leads to misidentification of samples , due to sharing of barcodes in single-index sequencing libraries [19] . This problem has also been described as “sample bleeding” that refers to the incorrect assignment of reads to multiplexed samples that are being sequenced in the same sequencing lane [57] . Dual-indexing will be needed for more accurate studies of the human virome . Many of the observed viruses might be truly present in human blood—however , it is difficult to distinguish them from prevalent contaminant viral sequences . Study design , epidemiological setting and downstream validation by independent techniques are needed to propose novel viruses . Overall , the analysis aims at defining the normal DNA virome background in blood in a presumably healthy population against which novel discoveries can be proposed . This study has the following limits . It analyzes a convenience population that does not contribute specific data on infectious diseases . However , this can be seen as an advantage in terms of better representing a general population . The nature of the sequencing protocol implies limited amplification of the viral genetic material , and a significant competition from the larger human genome . Therefore , this approach may not identify lower concentration viruses that could be revealed by using viral particles enrichment [58 , 59] or viral genome capture [60 , 61] . The latter methods rest on the ability to capture closely related sequences by hybridization to short conserved probes . Other recent approaches include methods that enable human viral epitope-wide exploration of immune responses in large numbers of individuals . This latter approach is effective for determining past viral exposure [62] . The study was not conceived for the discovery of highly divergent , novel human viruses , as this requires the use of less stringent similarity criteria for detecting divergent ( relative to those already known ) viral sequences . Lastly , the study did not address the RNA virome in human blood . Thus , the highly prevalent blood-borne RNA pegivirus A ( GBV-C ) in the Flaviviridae family was not detected here . The interest of the study derives from the size of the investigation that serves to define the human DNA blood virome . The second , and equally important part of the study is the description of the contamination profile during genome sequencing that may confound the discovery of novel human viruses . Increasing numbers of humans undergoing whole genome and transcriptome sequencing will support the precise description of the human blood DNA and RNA virome .
Participants were representative of the spectrum of age ( between 2 months and 102 years with a median of 56 ) , and of major human populations and ancestries . Specifically , the study included EUR , European = 5 , 384; AFR , African = 1 , 049; MDE , Middle Eastern = 213; EAS , East Asian = 159; AMR , CSA , Central South Asian = 94; Admixed American = 16; and Multi-Racial and Others = 1 , 325 . The study population was not ascertained for a specific infectious disease status . Other aspects of the study and the performance of genome sequence are detailed in Telenti et al . [1] . New ( Western Institutional Review Board , www . wirb . com ) and existing IRB-approved consent forms for participation in research and collection of biological specimens and other data used in this publication were reviewed and confirmed to be appropriate for use . All adult subjects provided informed consent , and a parent or guardian of any child participant provided written informed consent on their behalf . Library preparation was carried out using the TruSeq Nano DNA HT kit ( Illumina Inc . ) . Libraries were combined into 6-sample pools and clustered . Flow cells were sequenced on the Illumina HiSeqX sequencer utilizing a 150 base paired-end single index read format . Despite of the use of TruSeq technology , several ssDNA viruses were identified . It is possible that this is a reflect of extensive secondary structure of the naked viral DNA [63] and of replicative intermediate forms that are dsDNA [64] . For each BAM file , we extracted read pairs with either one or both of the reads not mapping to hg38 using sambama [65] with filtering for “unmapped” or “mate_is_unmapped” . Read pairs with average base quality below 30 were removed . Read pairs with low complexity identified using String Graph Assembler [66] with the following parameters dust-threshold = 2 . 5 and quality-filter = 50 then they were removed . Samples with more than 10% unmapped reads were excluded from further analysis . Unmapped reads were in a first step searched for putative viral matches by blastn [67] against the NCBI RefSeq [68] viral reference genomes ( > 8 , 000 viruses and phages ) [69] using an e-value ≤ 1e-10 . In a second step , candidate reads with viral hits were searched against a more comprehensive database comprised of NCBI RefSeq genomes of viruses , representative bacteria ( 1 , 636 species and strains ) , archaea ( 389 species and strains ) , and fungi ( two species ) , and UCSC genomes of human , chimp , mouse , chicken , and fruit fly , and NCBI nt vectors ( 274 , 565 sequences ) and plasmids ( 778 sequences ) using blastn with e-value ≤ 1e-20 . Viral hits were filtered for bit-score ≥ 190 . Reads with hits other than viruses with bit scores greater than or equal to the viral hits were discarded . Finally , randomly selected reads with viral hits of the human viruses were manually and visually verified by searching ( blastn ) against NCBI nt ( online ) and by aligning the reads to the corresponding viral genomes . The normalized abundance of a virus in a sample was estimated in genome copies per human cell ( viral genomes per human diploid genome ) with the following equation: virus abundance=2×number of reads mapped to viral genomevirus genome sizenumber of reads mapped to human genomehuman genome size For ease of interpretation , values are referred to a “viral copies per 100 , 000 human cells” . The fraction of viral reads has been shown to generally correspond to its viral load as determined by real time PCR [3 , 58 , 70] . The unmapped reads were also assembled in contigs using SOAPdenovo [71 , 72] with k-mer size 91 for each sample . Contigs that were mapped to the human reference with > 90% identity on > 30% length were removed . The remaining contigs were then mapped to the hg38 regions that were masked as repeat in UCSC goldenPath using blastn [67] without low complexity filtering to remove contigs that contain > 20% repeat sequences . Contigs passing the above filtering steps were clustered into non-redundant set using CD-Hit [73 , 74] with 90% global identity threshold . Non-redundant clusters were searched for matches to viral proteins using DIAMOND [75] against NBCI non-redundant proteins ( nr ) . To detect potential cases of integration between the viral genome and the human genome , identified viral reads were aligned to a database comprised of the viral genomes and the human reference genome hg38 to detect potential cases of integration , which were predicted via the identification of chimeric reads and chimeric mates using BWA [76] with the maximal exact matches algorithm “bwa mem” . An integration event was predicted when either one mate of a paired-end read aligned to a virus genome and the other mate aligned to the human genome or a single mate chimerically split into two alignments where one part mapped to a virus genome and the other part mapped to the human genome . We conducted a logistic regression analysis under a generalized linear model ( GLM ) with binomial distribution for the presence of human viruses in response to the individuals’ sex , ancestry , and age along with the cohort information as the covariate using the ‘glm’ method in R , followed by the `step`method for identifying the optimal model . The significance of the interactions was determined by chi-squared tests for the deviance table of the GLM . Statistical significances of the differences in prevalence and abundance across the demographic characteristics for each virus were estimated using chi-square test and Kruskal-Wallis test , respectively , followed by multiple test correction for the generated p-values . Virome reads are available for downloading at www . HLI-OpenData . com/Virome/ . In addition , see the Data Access Statement ( www . humanlongevity . com/wp-content/uploads/HLIDataAccessAgreement020416 . docx . ) for information on extended access . | Novel sequencing technologies offer insight into the virome in human samples . Here , we identify the viral DNA sequences in blood of over 8 , 000 individuals undergoing whole genome sequencing . This approach serves to identify 94 viruses; however , many are shown to reflect widespread DNA contamination of commercial reagents or of environmental origin . While this represents a significant limitation to reliably identify novel viruses infecting humans , we could confidently detect sequences and quantify abundance of 19 human viruses in 42% of individuals . Ancestry , sex , and age were important determinants of viral prevalence . This large study calls attention on the challenge of interpreting next generation sequencing data for the identification of novel viruses . However , it serves to categorize the abundance of human DNA viruses using an unbiased technique . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"hepacivirus",
"pathogens",
"microbiology",
"human",
"genomics",
"viruses",
"rna",
"viruses",
"dna",
"viruses",
"genome",
"analysis",
"microbial",
"genomics",
"research",
"and",
"analysis",
"methods",
"viral",
"genomics",
"genomic",
"libraries",
"genomics",
"medical",
"microbiology",
"microbial",
"pathogens",
"biological",
"databases",
"hepatitis",
"c",
"virus",
"hepatitis",
"viruses",
"blood",
"anatomy",
"flaviviruses",
"virology",
"physiology",
"database",
"and",
"informatics",
"methods",
"genetics",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"genomic",
"databases",
"organisms"
]
| 2017 | The blood DNA virome in 8,000 humans |
The opportunistic fungal pathogen Cryptococcus neoformans is a major cause of mortality in immunocompromised individuals , resulting in more than 600 , 000 deaths per year . Many human fungal pathogens secrete peptidases that influence virulence , but in most cases the substrate specificity and regulation of these enzymes remains poorly understood . The paucity of such information is a roadblock to our understanding of the biological functions of peptidases and whether or not these enzymes are viable therapeutic targets . We report here an unbiased analysis of secreted peptidase activity and specificity in C . neoformans using a mass spectrometry-based substrate profiling strategy and subsequent functional investigations . Our initial studies revealed that global peptidase activity and specificity are dramatically altered by environmental conditions . To uncover the substrate preferences of individual enzymes and interrogate their biological functions , we constructed and profiled a ten-member gene deletion collection of candidate secreted peptidases . Through this deletion approach , we characterized the substrate specificity of three peptidases within the context of the C . neoformans secretome , including an enzyme known to be important for fungal entry into the brain . We selected a previously uncharacterized peptidase , which we term Major aspartyl peptidase 1 ( May1 ) , for detailed study due to its substantial contribution to extracellular proteolytic activity . Based on the preference of May1 for proteolysis between hydrophobic amino acids , we screened a focused library of aspartyl peptidase inhibitors and identified four high-affinity antagonists . Finally , we tested may1Δ strains in a mouse model of C . neoformans infection and found that strains lacking this enzyme are significantly attenuated for virulence . Our study reveals the secreted peptidase activity and specificity of an important human fungal pathogen , identifies responsible enzymes through genetic tests of their function , and demonstrates how this information can guide the development of high affinity small molecule inhibitors .
Cryptococcus neoformans is an opportunistic fungal pathogen responsible for 40% of all AIDS-related deaths [1 , 2] . Of the one million new infections occurring worldwide annually , greater than 60% result in death due to the limited efficacy and availability of therapeutics [3] . Only three classes of drugs are currently approved for treatment of fungal infections , thus there is a significant need for development of new antifungal compounds [3–5] . Peptidases are secreted by many types of pathogens including bacteria , fungi and parasites and often serve critical roles related to survival and virulence [6–11] . Direct targeting of peptidases expressed by pathogenic organisms has proven to be a successful therapeutic strategy , notably in the development of Hepatitis C Virus ( HCV ) and Human Immunodeficiency Virus ( HIV ) protease inhibitors [12 , 13] . Additionally , the identification and characterization of peptidases secreted by pathogens have contributed to the formulation of new diagnostic approaches based on detection of these proteolytic activities [14–16] . Pathogenic fungi express extracellular peptidases for wide-ranging functions including host tissue invasion , nutrient acquisition and regulation of mating [17–19] . A single organism may simultaneously secrete multiple peptidases with divergent substrate specificities and requirements for activity that are tailored to their biological functions . In addition , peptidase secretion and activation are often stimulated by extracellular conditions , as distinct proteolytic functions can be important for different environments . Candida albicans and Aspergillus fumigatus , two prominent fungal pathogens , each secrete several peptidases with defined roles in virulence , while dermatophytes and the causative agent of white-nose syndrome Pseudogymnoascus destructans use extracellular peptidases to degrade host tissues [20–26] . Multiple peptidases have been identified in the secreted proteome of C . neoformans , including a metallopeptidase that is required for dissemination to the central nervous system ( CNS ) in a mouse infection model [27–34] . Interestingly , the level of peptidase secretion has been shown to vary between isolates in Cryptococcus species and in many cases higher secretion has been correlated with increased virulence [35–38] . Although these findings suggest that extracellular peptidases are involved in C . neoformans pathogenicity , the delineation of their functions and their validation as therapeutic targets is limited by poor understanding of their activity , specificity and regulation . In this work , we used a comprehensive activity-based approach to characterize secreted peptidases in C . neoformans culture supernatants . This strategy , termed Multiplex Substrate Profiling by Mass Spectrometry ( MSP-MS ) , relies on mass spectrometry to identify cleavage events within a defined 228-member library comprising physiochemically diverse tetradecapeptides [39] . The scope and design of the library allows detection of cleavage events from multiple peptidases simultaneously , and the resulting data are informative for understanding activity on both a global and individual enzyme level . Activity-based profiling stands in contrast to traditional proteomics methods that catalog which peptidases are present but do not provide information on how each enzyme contributes to the overall proteolytic activity [11 , 27] . Likewise , candidate-based approaches focusing on single proteolytic activities isolated from cultures may not accurately represent how these enzymes function within the secreted peptidase milieu [31 , 32] . To investigate the secreted peptidases of C . neoformans and test the influence of environment on global proteolytic activity , we cultured fungal cells under two different conditions and then isolated the cell-free supernatants for substrate specificity profiling . These experiments revealed that overall peptidase specificity differs greatly in response to extracellular conditions . To uncover the contribution of individual enzymes to the total proteolytic activity , ten candidate peptidases were individually deleted and conditioned media generated from each mutant strain was compared to the parental strain . Through this approach , we identified and defined the putative substrate preferences of three peptidases , including a previously uncharacterized secreted aspartyl peptidase . We found that this enzyme is the dominant contributor to extracellular endopeptidase activity at acidic pH and determined that this activity is required for tolerance to low pH environments . Analysis of its substrate specificity enabled us to screen an appropriately focused library of aspartyl peptidase inhibitors , which led to the identification of potent in vitro antagonists . Finally , we found that deletion strains of this enzyme are attenuated for virulence in a mouse inhalation model of C . neoformans infection . Our in-depth characterization of extracellular peptidases in C . neoformans establishes a framework for uncovering the biological functions of these enzymes . As demonstrated by our identification of a peptidase required for virulence , examining the roles of these enzymes is critical to understanding the pathogenicity of C . neoformans . Furthermore , the methods described here are applicable to the discovery and characterization of secreted peptidases from other pathogenic organisms .
C . neoformans was cultured in either microbial minimal media ( yeast nitrogen base [YNB] pH 5 . 0 ) or mammalian tissue-culture media ( DMEM pH 7 . 4 ) , and supernatants from each condition were assayed using a panel of internally quenched ( IQ ) fluorogenic peptides ( Fig 1A , S1 Table for sequences ) . These substrates were previously developed to detect a broad range of microbial peptidases from diverse peptidase families [40–42] . The speed and flexibility of this assay allowed us to optimize the conditions for peptidase activity and to determine which class-specific inhibitors affect it . Although peptidase activity was evident under both culture conditions , differential substrate cleavage reflected differences in specificity . Notably , IQ-2 and IQ-6 were cleaved more efficiently by peptidases in YNB media , while proportionally higher activity was observed against IQ-3 and IQ-4 in DMEM media ( Fig 1A ) . These differences suggested that alternate peptidases were active in each culture condition , which was further confirmed by assaying the substrates in the presence of class-specific peptidase inhibitors . This analysis revealed that aspartyl peptidase activity was present in YNB conditioned media while metallopeptidase activity could be detected in DMEM media ( S1 Fig ) . Adjustment of YNB supernatants from pH 5 . 0 to 7 . 4 ( the pH of DMEM media ) yielded only very low levels of any peptidase activity , while lowering the pH of DMEM supernatants from 7 . 4 to pH 5 . 0 produced a peptidase activity pattern similar to YNB media ( S1 Fig ) . This result suggests that growth in DMEM media using mammalian cell culture conditions induces peptidases optimized for neutral pH in addition to the acidic pH activities detected after growth in YNB media . To investigate global peptidase substrate specificity , MSP-MS was conducted on YNB and DMEM supernatants at the optimal pH for activity observed for each condition , pH 5 . 0 and 7 . 4 , respectively . In the MSP-MS assay , peptide sequencing by LC-MS/MS is used to identify all peptide cleavage products within the 228-member library , revealing peptidase substrate specificity preferences ( Fig 1B ) . The reproducibility of these substrate specificity preferences , and of the cleavage events from which they derive , was confirmed by assessing three technical replicate samples ( S2 Fig and S2 Table ) . Importantly , since there are no modifications to either the N- or C-termini in the peptide library , both exo- and endo-peptidases can be identified in an unbiased manner . Using MSP-MS we observed that peptidases in YNB media cleaved at 423 total sites , whereas peptidases in DMEM media cleaved at 283 total sites ( Fig 1C ) . Only 107 of these sites were cleaved by enzymes in both samples . This difference in cleavage site preference indicated that peptidase activity and specificity differs between the two culture conditions . Positional analysis of all bonds cleaved within the tetradecapeptides of the MSP-MS library illustrates the proportion of endo- and exo-peptidase activity in each sample ( Fig 1D ) . In YNB supernatants , the most frequently cleaved bond was the carboxyl terminal bond between amino acids thirteen and fourteen , representing 32% of total proteolysis events . In fact , 137 of the 228 peptides had their carboxyl-terminal amino acid cleaved . Moreover , we observed that single amino acids were often sequentially hydrolyzed from the carboxyl termini of substrates until an unfavored residue was reached , consistent with the presence of abundant carboxypeptidase activity . This cleavage preference was not observed for proteases in the DMEM media . These studies indicate that carboxypeptidase activity is more abundant in conditioned media from C . neoformans cultures grown in YNB . To further illustrate the differences in proteolytic activity between the two conditions , representative examples of peptides cleaved in both samples are shown ( Fig 1E ) . To identify which C . neoformans peptidases may be contributing to the global substrate specificity profile , we conducted a proteomic analysis of secreted proteins . We observed 199 and 131 proteins in YNB and DMEM conditioned media respectively , with 52 proteins common to both conditions ( S3 Table ) . Recently , Geddes and colleagues identified 61 proteins in the secretions of C . neoformans grown for 16 to 120 hours in minimal media [43] , while Campell and coworkers identified 22 secreted proteins after 168 hours growth in RPMI media [29] . In total , 24 of the proteins detected in our study contained predicted secretion signals ( SignalP 4 . 0 ) [44] , 127 were predicted to be non-classically secreted ( SecretomeP 2 . 0 ) [45] , and 17 have been associated with extracellular microvesicles [46] . The remaining proteins identified had no known mechanism of secretion ( S3 Fig , S3 Table ) . Seven of the proteins with predicted signal sequences were peptidases and included members of the aspartyl , metallo and serine peptidase families . Both endopeptidases and carboxypeptidases were identified , consistent with our predictions from analysis of C . neoformans extracellular proteolytic activity ( Fig 1 ) . Five of these enzymes have been detected in studies of the C . neoformans secretome by other groups; however Prc1 and CNAG_05872 have not been observed previously . To determine which enzymes are responsible for the proteolytic activity present in C . neoformans conditioned media , we performed targeted gene deletions on ten candidate secreted peptidases ( Table 1 , S4 Table ) . Of the seven aforementioned peptidases with predicted signal sequences that were identified by our secretome proteomics , one was predicted to be GPI-anchored ( CNAG_04380 ) [27 , 47] , and thus excluded from further analysis , as our study was focused on non-cell wall anchored enzymes . One other peptidase could not be mapped unambiguously to a single gene , as three paralogs of this enzyme exist in the C . neoformans var grubii genome [48] . Therefore , all three genes were individually targeted for deletion ( CNAG_00919 , CNAG_01040 and CNAG_02966 ) . Because these genes are unnamed and lack orthologs in Saccharomyces cerevisiae , we propose naming them Carboxypeptidase D 1 , 2 and 3 ( CXD1-3 ) , respectively . This resulted in eight genes deleted based on our proteomics results ( Table 1 , S4 Table ) . We additionally targeted two secreted peptidases that were not identified here but have been reported in previous proteomics studies [27] . Two independent isolates of each of the ten deletion strains were generated and are indicated in the text and figures by the gene name or CNAG number followed by “-1” or “-2” ( S4 Table ) . Based on our characterization of secreted peptidase activity present in wild type C . neoformans , we selected deletion strains for in-depth substrate profiling analysis by MSP-MS under either DMEM or YNB culture conditions . Subsequently , by comparing the secreted peptidase activity in conditioned media from the wild type and mutant strains , we were able to correlate extracellular proteolytic activities to specific candidate enzymes . To analyze the peptidase substrate specificity of DMEM media conditioned by wild-type cells , we generated a frequency plot from the 283 cleavage events detected by MSP-MS ( Fig 2A , S5 Table ) [50] . The amino acid preferences are shown for four positions on either side of the cleaved bond ( P4-P4' ) , as the majority of substrate specificity is determined by residues closest to the scissile bond . This analysis revealed that peptidases in DMEM supernatants prefer positively charged residues on either side of the cleaved bond , as well as hydrophobic residues in the P1' position . Negatively charged amino acids are disfavored at the majority of positions , and proline and glycine are both highly disfavored in most positions from P2-P2' ( Fig 2A ) . To identify the enzymes responsible for this activity , we examined proteolytic activity in peptidase deletion strains . Because DMEM conditioned media contained metallopeptidase activity ( S1 Fig ) and a single metallopeptidase ( Mpr1 ) was identified by proteomics ( Table 1 , S3 Table ) , we began by investigating the contribution of this enzyme to the global specificity profile . Mpr1 had previously been characterized as a secreted factor that is important for C . neoformans invasion of the CNS [28] . Matched comparison of the substrate specificity profiles obtained from DMEM media conditioned by wild type or mpr1Δ cells revealed that Mpr1 deficiency caused a loss of the P1' preference for hydrophobic amino acids seen in wild type ( Fig 2A and 2B ) . However , the selection for positively charged residues on either side of the cleaved bond remained unaltered and the same amino acids were disfavored in most positions . To further analyze the impact of MPR1 deletion , a Venn diagram was used to compare the overlap of cleavage events between wild type and mpr1Δ ( Fig 2C ) . A majority of cleaved peptides were detected in both samples; however 107 cleavage events were detected in wild type but not media conditioned by mpr1Δ . These cleavages , presumed to be absent due to the loss of this enzyme , were used to generate a frequency plot representing the putative specificity of Mpr1 ( Fig 2C ) . A prominent feature of this substrate specificity profile is enrichment for phenylalanine , leucine and norleucine ( a replacement for methionine in the MSP-MS library ) at the P1' position , a result that is consistent with the specificity of other peptidases predicted to be related to this enzyme ( members of the M36 peptidase family ) [51] . It is also notable that the P1' site exhibits the greatest degree of selectivity of any position from P4-P4' . To further illustrate the changes in substrate specificity observed in the mpr1Δ deletion strain , a representative example of a peptide cleaved by enzymes in both wild type and mpr1Δ supernatants is shown ( Fig 2D ) . An additional activity in DMEM media conditioned by wild type displays a trypsin-like preference for proteolysis between two positively charged residues , indicating the presence of serine peptidase activity [51] . This specificity is particularly evident in the mpr1Δ culture media ( Fig 2B ) . Two serine endopeptidases were present in the deletion collection and DMEM conditioned media was analyzed from both strains ( prb1Δ and CNAG_00150Δ ) . Deletion of either gene did not substantially impact the extracellular peptidase activity profile , suggesting functional redundancy or the existence of additional , unidentified peptidases ( S4 Fig ) . One predicted serine peptidase with a secretion signal , KEX2 , was identified in a genome search . However , our attempts to delete this gene were unsuccessful , indicating it may be essential for C . neoformans survival . In some cases , media conditioned by knockout cells produced additional peptide cleavage sites as compared to wild type conditioned media , despite similar overall peptidase specificity profiles ( e . g . , S4C Fig ) . This observation is consistent with the fact that iterative cleavage of an MSP-MS substrate peptide can hinder identification of a given cleavage event due to loss of the cleavage’s reaction product . In this way , the loss of a minor peptidase activity can result in the appearance of new cleavage sites [25] . A substrate specificity profile constructed from the 423 cleavages observed in YNB media conditioned by wild-type cells indicated a preference for hydrolysis between hydrophobic residues , while positively charged residues , proline and glycine are disfavored ( Fig 3A ) . From positional analysis of these cleavage sites , we identified carboxypeptidase activity as the dominant proteolytic activity in this media ( Fig 1D ) . Since carboxypeptidases cleave the carboxyl-terminal bond , no enrichment of amino acids is evident in the P2' to P4' positions in these substrates ( Fig 3A ) . To determine whether any of the three carboxypeptidase D paralogs we identified in our proteomics analysis were responsible for the observed carboxypeptidase activity , the gene for each enzyme was deleted and conditioned media from the resulting mutant strains ( cxd1Δ , cxd2Δ and cxd3Δ ) was profiled by MSP-MS . After comparison of the specificity profiles , it was clear that the carboxypeptidase specificity was most dependent on Cxd1 . In media conditioned by this deletion strain only 17 primary carboxyl-termini were cleaved , as compared to 137 in wild type ( Fig 3B , 3D and 3E ) . By comparison , 134 and 123 primary carboxyl-termini were cleaved in media conditioned by cxd2Δ and cxd3Δ , respectively , suggesting that these enzymes do not contribute substantially to the extracellular carboxypeptidase activity ( S5 Fig , S5 Table ) . As anticipated , endopeptidase cleavages were not affected in any of the three carboxypeptidase deletion strains ( Fig 3B and 3D , S5 Fig ) . Fluorogenic assays demonstrated aspartyl endopeptidase activity in wild type YNB supernatants ( S1 Fig ) . To assign activity to the candidate aspartyl peptidases , conditioned YNB media was profiled from the two aspartyl peptidase deletion strains listed in Table 1 ( CNAG_05872Δ and pep4Δ ) . Proteolytic activity remained unchanged relative to wild type in the pep4Δ strain ( S5 Fig ) . In contrast , CNAG_05872Δ conditioned media exhibited a near-total loss of endopeptidase cleavage events as well as substantially decreased carboxypeptidase activity as evidenced by proteolysis of only 55 primary carboxyl termini ( Fig 3C–3E ) . This result suggests that CNAG_05872 is the dominant endopeptidase under these culture conditions . This finding is consistent with fluorogenic assays , where deletion of CNAG_05872 led to a loss of endopeptidase activity in conditioned YNB media , while all other strains exhibited activity levels similar to wild type ( Fig 3F , S6 Fig ) . As the putative dominant endopeptidase , we propose renaming CNAG_05872 to Major Aspartyl peptidase 1 ( MAY1 ) . Due to its substantial contribution to peptidase activity in YNB supernatants , we performed an in-depth biochemical characterization of May1 . This enzyme consists of a 16 residue secretory signal ( SignalP 4 . 0 ) [44] followed by an 82 residue prodomain ( Fig 4A ) . The prodomain is positively charged ( pI = 9 . 97 ) , which likely facilitates interaction with the negatively charged catalytic domain ( pI = 4 . 03 ) at neutral or slightly acidic pH [52] . The pepsin-like aspartyl peptidase domain includes residues 100–434 and is expected to auto-activate in acidic environments , causing release of the pro-domain . The N-terminal region ( position 101–223 ) is also an N-terminal xylanase inhibitor domain , TAXi_N [53] . Homology between xylanase inhibitors and fungal aspartyl peptidases has been noted previously and likely indicates an evolutionary relationship [54] . May1 readily cleaves IQ-2 between phenylalanine and leucine ( S1 Table , S6 Fig ) , which allowed us to use fluorogenic assays to monitor enrichment of this enzyme from YNB supernatants and investigate the impact of pH on its activity . Ion exchange chromatography was used to enrich May1 from conditioned YNB media , resulting in a 292 nM peptidase stock solution . May1 was diluted from this stock into buffers ranging from pH 1 . 5 to 7 . 0 and activity against IQ-2 was tested . Optimal activity was observed between pH 3 . 5–4 . 5 , a range that is consistent with other members of the aspartyl peptidase family of enzymes ( Fig 4B ) [55] . The aspartyl peptidase antagonist pepstatin A fully inhibited proteolysis of IQ-2 in this assay , providing further verification that May1 is the predominant endopeptidase activity under these conditions . To investigate the time- and growth-dependent secretion of May1 , we added a CBP-2xFLAG tag to the carboxyl-terminus through homologous recombination . By monitoring activity using IQ-2 , we confirmed that the addition of this tag did not diminish May1 activity in YNB conditioned media . Interestingly , although recombinant May1 activity could be detected in the culture supernatant after overnight growth , it could not be detected by immunoblot even after three days of growth . We hypothesized that the enzyme rapidly hydrolyses the C-terminal tag; therefore pepstatin A was added to the culture to inhibit this processing . This inhibitor also prevented activation of pro-May1 to mature May1 , resulting in the observation of only the pro form of recombinant May1 . Under these conditions the protein was detected by immunoblot after 48 hours of growth ( Fig 4C ) . The apparent molecular weight of pro-May1 was approximately 13 kDa greater than the predicted 56 . 8 kDa for the tagged enzyme , suggesting that this protein could contain post-translational modifications . When C . neoformans was cultured in YNB media buffered to pH 6 . 5 , the May1 activity detectable in supernatants using IQ-2 was approximately 10-fold lower than in unbuffered YNB , and no signal could be seen by immunoblot after 48 hours of growth ( Fig 4D ) . This result suggests that low extracellular pH could stimulate May1 secretion and activation . We observed that may1Δ strains grown in YNB had a lower cell density at saturation than wild type or any of the other nine peptidase deletion strains ( Fig 5A , S7 Fig , S6 Table ) . Higher cell density could not be achieved by the may1Δ strains even after culturing for 96 hours . These data suggested that may1Δ strains were not merely slow to replicate but were incapable of growth at high density . In fact , during the exponential growth phase , may1Δ strains had an average doubling time of 2 . 36 hours , which was on par with the other nine peptidase deletion strains . However , all other deletion strains exhibited wild-type saturation densities ( S6 Table ) . Studies of deletion strains did not clarify whether May1 activity , as opposed to simply the presence of the May1 protein , was required to reach a saturation density equivalent to wild type . Therefore , we assessed growth of wild type C . neoformans in the presence and absence of pepstatin A . Treatment of wild type cultures with this aspartyl peptidase inhibitor resulted in a saturation density defect equivalent to that observed for the MAY1 deletion strains ( Fig 5B ) . Importantly , the inhibitor had no effect on may1Δ cells , suggesting that the growth defect observed in wild-type cells treated with pepstatin A was mediated through inhibition of May1 . Plating assays on various stress conditions were conducted with two independent isolates of each of the ten peptidase deletion strains ( S8 and S9 Figs ) . After 48 hours may1Δ strains exhibited a growth defect at pH 3 . 5 but not at pH 5 . 0 or 6 . 5 ( Fig 5C ) . Longer growth periods at pH 3 . 5 did not allow may1Δ colonies to overcome this sensitivity and after three days of growth , it became apparent that may1Δ colonies also had a slight defect at pH 5 . 0 but not pH 6 . 5 ( S8 Fig ) . None of the other peptidase deletion strains displayed sensitivity to acidic pH , although pep4Δ was sensitive to hydrogen peroxide and SDS , and prb1Δ had a slight sensitivity to hydrogen peroxide ( S8 and S9 Figs ) . Based on this result , we hypothesized that the inability of may1Δ strains to reach wild-type saturation densities in YNB was a result of acidification of the media and would be rescued by buffering the media to pH 6 . 5 . These culture conditions fully rescued the saturation density of may1Δ ( Fig 5D ) . Surprisingly , it also allowed the final saturation densities of both wild type and may1Δ cultures to approximately double , revealing that low pH is a condition limiting growth even for wild-type C . neoformans . We also assessed melanization , an established virulence factor , for each of the ten peptidase deletion strains . Because melanin production occurs extracellularly , we hypothesized that this process could be influenced by secreted peptidase activity . Only the serine endopeptidase deletion strain prb1Δ exhibited a hypomelanization phenotype ( S9 Fig ) . While pepstatin A inhibits May1 with an IC50 of 1 . 4 nM , it is a broad acting antagonist of many members of the aspartyl class of peptidases , thereby limiting its utility . To determine whether additional inhibitors targeting May1 could be obtained , we conducted an in vitro screen using knowledge of May1 substrate specificity derived from MSP-MS analysis . We screened a panel of 21 peptidomimetic molecules with similarities to May1 substrate preferences but with a non-cleavable bond between the P1 and P1' position ( S7 Table ) . Compounds 1 to 11 are linear peptidomimetics , while compounds 12 through 21 are macrocycles ( S7 Table ) [56–59] . We also screened ten HIV protease inhibitors because some of these molecules have been reported to inhibit C . neoformans peptidase activity [60 , 61] . May1 was incubated with 100 μM , 10 μM and 1 μM of each inhibitor and activity was detected using IQ-2 ( Fig 6A , S10 Fig ) . IC50 values were then calculated for the most potent compounds . The best inhibition by an HIV protease inhibitor was observed with Brecanavir , which reduced activity by 80% at 1 μM and had an IC50 of approximately 352 nM ( S10 Fig ) . Among the peptidomimetic molecules , the macrocycles were the most potent , with the best compounds ( 15 to 21 ) containing P2 –P3' tethered side chains , statines in P1 and an α-amino acid in P2' ( Fig 6 , S7 Table ) . Compounds 16 , 21 , and 18 all exhibited nanomolar IC50 values of 1 . 6 nM , 9 . 4 nM and 41 nM , respectively ( S10 Fig ) . Among the linear peptidomimetic inhibitors , those with a phenylstatine or hydroxyethylamine scissile bond isoster ( compounds 4 , 7 , 8 and 11 ) were superior to compounds with a reduced bond ( 1 , 2 , 5 , 6 and 9 ) or a homo-amide ( 2 ) . Compound 4 was the most potent May1 antagonist out of this group of inhibitors , with an IC50 of 3 . 1 nM ( S10 Fig ) . From analysis of the four most effective inhibitors identified in our screen ( compounds 4 , 16 , 18 and 21 ) , it is clear that a phenylalanine side chain , either unsubstituted ( 4 and 16 ) or with a small substituent ( 18 and 21 ) , is preferred in P1 while a bulkier P1 side chain leads to decreased potency , for example compounds 17 , 19 and 20 . These results match the P1 substrate preference for phenylalanine that we had predicted for May1 and fit our expectations that bulkier residues such as tryptophan are not well tolerated in this position ( Figs 3 and 6B ) . Next , we selected the two best in vitro hits to test their potency in culture relative to pepstatin A by measuring inhibition of May1 and restriction of culture growth using fluorogenic assays and OD600 respectively . Wild-type C . neoformans was grown in YNB treated with 5 μM , 1 μM or 0 . 1 μM of compound 4 , 16 or pepstatin A and the culture density and May1 activity were measured at saturation . While compound 16 exhibited an in vitro IC50 comparable to pepstatin A , it was not as effective at inhibiting May1 activity or restricting culture growth ( Fig 6D , S11 Fig ) . Curiously , despite having an in vitro IC50 approximately twice that of compound 16 , compound 4 was better at inhibiting culture growth . None of the three compounds affected the culture density of a may1Δ strain , consistent with the idea that May1 is the compounds’ relevant target in this context ( S11 Fig ) . These results demonstrate that May1 can be targeted by small molecule inhibitors and provide a discovery framework for further inhibitor development . However , additional medicinal chemistry efforts are necessary for in vivo applications . Therefore , subsequent studies investigating the role of May1 in virulence were carried out using deletion strains of this enzyme . Because may1Δ strains exhibit phenotypes in both peptidase activity assays and growth at low pH , we examined the role of this protein in virulence using an established mouse inhalation model of Cryptococcal infection [81] . Wild-type cells were mixed with an equivalent number of may1Δ cells and used to infect mice intranasally ( Fig 7A ) . These experiments were conducted using two independent isolates of may1Δ as well as a negative control known not to affect fungal replication in this assay ( sxi1Δ ) [63] . In each isolate , genes neighboring may1Δ were tested by RT-qPCR to confirm expression ( S12 Fig ) . Ten days after infection , mouse lungs were harvested and plated for colony forming units ( CFUs ) , at which point may1Δ strains contributed only 22 ± 3% of the colonies recovered from the lungs , a substantial decrease from the approximately 50% present upon infection . This result reveals that may1Δ cells have a growth defect within a mammalian host because the ratio between deletion strain and wild type cells was reduced after host infection . Given the defect in may1Δ cell accumulation in the lung , we evaluated the role of May1 during Cryptococcus interaction with macrophages . Opsonized wild type and may1Δ strains were phagocytosed with equal efficiency by bone marrow-derived macrophages ( S13A Fig ) . We next tested the ability of may1Δ cells to accumulate within macrophages after phagocytosis , since the phagolysosome environment may represent a low-pH setting in which May1 is active . Indeed , cells lacking May1 accumulated significantly more slowly within macrophages than did wild type cells ( S13B Fig ) , consistent with a role for May1 within host cells . The defect in accumulation of may1Δ cells during intranasal infection and within macrophages suggested that these strains would be attenuated for virulence . We directly investigated the virulence of may1Δ strains by performing monotypic infections [62] . Ten mice per group were infected intranasally with wild type , may1Δ-1 or may1Δ-2 cells ( Fig 7B ) . Loss of May1 caused significant attenuation of virulence , with mice infected by may1Δ-1 or may1Δ-2 exhibiting a mean survival time of 60 . 1 and 60 . 7 days respectively , whereas those infected by wild type had a mean survival time of 25 days . The results from these in vivo experiments indicate an important role for May1 during mammalian infection .
In this work , we investigated secreted proteolytic activity in C . neoformans var . grubii culture media using an unbiased approach that can detect both endo- and exo-peptidase activity . In combination with proteomic methods and single gene deletion approaches , this strategy allowed us to characterize peptidase activity from a global perspective as well as interrogate the roles of individual enzymes in the C . neoformans secretome . By comparing the overlap in peptidase activity between wild type and these deletion strains , we were able to identify and define the substrate specificities of a carboxy , aspartyl and metallopeptidase which contribute substantially to the total peptidase activity profile . Additionally , we delineated the substrate specificity of an unidentified trypsin-like peptidase activity , an intriguing result given previous reports implicating secreted serine peptidases in C . neoformans pathogenicity [31 , 34] . Deletion of some peptidase genes , such as the predicted carboxypeptidase D genes CXD2 and CXD3 , caused no significant change in secreted proteolytic activity or cellular phenotype . Instead , it appears that a third carboxypeptidase D paralog CXD1 is responsible for the majority of exopeptidase activity under these conditions . The broad specificity of Cxd1 suggests that one possible role for this enzyme could be in nutrient acquisition by providing C . neoformans with free amino acids from extracellular protein sources [17 , 18 , 64] . The serine peptidase deletion strain prb1Δ also had a minimal effect on total secreted peptidase activity; however , a phenotype of reduced melanin production was evident , indicating function under these conditions ( S9 Fig ) . One possibility is that this gene encodes an enzyme with very strict substrate specificity , thus its deletion did not have a substantial impact on total extracellular peptidase activity as measured by the MSP-MS assay ( S4 Fig ) . Through the application of our global profiling approach to different culture conditions , we were able to demonstrate that the landscape of secreted peptidase activity shifts in response to alterations in environment . This result raises the possibility that changes in extracellular proteolytic activity could be relevant for adaptation . For example , we detected the activity of the metallopeptidase Mpr1 only after growth under neutral pH conditions , whereas we find that May1 is optimally active under acidic conditions . Thus , these enzymes may function in different settings within the host or within other environments encountered by C . neoformans . Through proteolytic profiling and mutant characterization assays , we identified the aspartyl peptidase May1 as the dominant endopeptidase at low pH and found that its activity is required for tolerance to acidic environments . The strongest determinant of May1 substrate specificity was shown to be a preference for cleavage between hydrophobic residues , in particular phenylalanine , leucine and norleucine ( Fig 3A ) . Based on these results , we screened a focused panel of aspartyl peptidase inhibitors with similarity to the P1-P1' substrate specificity of May1 . Several of these compounds had IC50 values in the nanomolar range , whereas the HIV protease inhibitors had relatively poor affinity for May1 . Previous reports have shown that some HIV protease inhibitors reduce secreted aspartyl peptidase activity produced by C . neoformans [60 , 61] . The concentrations of inhibitors required to achieve statistically significant inhibition in previous studies were much higher than those used in the experiments reported here although the trends for inhibitor potency match our results [60] . Therefore , it is possible that the inhibition of C . neoformans aspartyl peptidase activity seen in previous publications could be explained by the inhibition of May1 . We found that strains lacking May1 are attenuated in a competition infection assay , a macrophage accumulation assay and a monotypic infection assay . In microbial culture it is likely that May1 cleaves one or more secreted or cell wall-bound fungal proteins to facilitate low pH tolerance . However , it is possible that during an infection May1 cleaves host proteins and either or both of these proteolytic events impacts virulence . An additional important consideration for defining the role of May1 in C . neoformans pathogenicity is the cleavage context within the host . Our pH titration determined that May1 has very low levels of activity above pH 6 . 5; however few environments of lower pH than this exist within the mammalian host . Therefore , it is possible that residual May1 activity at neutral pH is important , or alternatively that May1 could be relevant for survival in acidic host environments such as dendritic cell or macrophage phagolysosomes , which exhibit a pH of ~5 . 0 in the context of Cryptococcal phagocytosis [65 , 66] . A third possibility is that a combination of these factors contributes to the attenuated virulence of may1Δ strains . We have identified orthologs of MAY1 in many other basidiomycetes including the opportunistic pathogens Trichosporan asahii and Cryptococcus gattii , the latter of which is capable of infecting immunocompetent individuals [67] [48] . Many pathogenic ascomycetes also contain MAY1 orthologs , including Histoplasma capsulatum , Coccidioides immitis and Aspergillus species , although the sequence identity is low [48] . None of the MAY1 orthologs in basidiomycetes has been well studied and only one ortholog in an ascomycete has been examined . This enzyme , from A . fumigatus , encodes a protein secreted during infection of the virulence model Galleria mellonella [68] . The hypovirulent phenotype observed in C . neoformans may1Δ strains and the identification of May1 orthologs in other fungal pathogens raises the possibility that this peptidase family displays a conserved virulence function and suggests that the roles of these orthologs are important to investigate . Small molecule drug development requires a thorough understanding of the target enzyme as well as the surrounding peptidase milieu [69–73] . The results described in this report lay the groundwork for investigating the functions of C . neoformans secreted peptidases and the use of inhibitors to modulate their activity . The substrates and inhibitors presented here may also be of value for interrogating related fungal peptidases . Furthermore , our approach to investigating secreted peptidases through integration of activity profiling , proteomics , and genomics strategies is broadly applicable to other genetically tractable pathogens .
Studies in mice were carried out according to the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . All protocols were reviewed and approved by the Institutional Animal Care and Use Committee , University of California , San Francisco , approval number AN091509-02C . During infections , mice were anesthetized by an intraperitoneal injection of ketamine ( 75 mg/kg ) and dexmedetomidine ( 0 . 5 mg/kg ) , which was reversed by an intraperitoneal injection of atipamezole ( 1 . 5 mg/kg ) . Mice were sacrificed in accordance with protocol guidelines by CO2 inhalation and cervical spine dislocation . Conditioned media was prepared from wild type C . neoformans cultured in YNB ( 32 hours ) or DMEM ( 48 hours ) as described below and concentrated using Millipore centrifugation filters , ( 3 kDa molecular weight cutoff ) . Trypsin digestion was conducted as previously described using a 1/40 mass ratio of trypsin/protein [39] . Peptides were recovered and desalted using C18 tips ( Rainin ) . Peptide identification was conducted as previously described using the LTQ-Orbitrap XL mass spectrometer ( Thermo ) [25] . To identify proteins , searches were carried out against the Uniprot database ( downloaded March 21 , 2012 ) , with Cryptococcus species entered as the taxonomy . This database was concatenated with a fully randomized set of proteins for determination of false-identification rate . Peptides were matched with up to 2 missed trypsin cleavages , carbamidomethylated cysteines as a fixed modification and oxidation of methionine , N-terminal methionine loss with or without acetylation , N-terminal acetylation or oxidation and pyroglutamate from glutamine at the N-terminus as variable modifications . Tolerance for mass accuracy was 20 ppm for parent and 0 . 8 Da for fragment errors . For protein identification from the database search , the Protein Prospector settings were: 15 for the minimum protein score and 10 for the minimum peptide score . The maximum expectation value for proteins was set at 0 . 009 and for peptides it was 0 . 05 . At the time of this study , the Uniprot database did not contain annotated C . neoformans var grubii genes , thus protein matches were identified within other C . neoformans serotypes and the var grubii orthologs were identified by searching the H99 genome either manually or through BLASTp searches using the NCBInr database ( http://blast . ncbi . nlm . nih . gov/blast/Blast . cgi ) . SignalP version 4 . 0 was used to predict secretion signals , while SecretomeP version 2 . 0 was used to predict non-classical secretion pathways [44 , 45] . Data are reported in S3 Table . Identification of May1 orthologs was conducted by searching for CNAG_05872 in FungiDB ( www . fungidb . org ) [48] . The functional domains of May1 were annotated using BLASTp . Isoelectric point and molecular weights were determined using ExPASy ( http://www . expasy . org/ ) [52] . Bone-marrow derived macrophages ( BMDMs ) were isolated from C56Bl/6 mice and used for phagocytosis assays as described previously [80] . Briefly , BMDMs were plated in a 96-well plate ( 10 , 000/well ) and simulated with 100 ng/ml Interferon-γ ( Roche ) starting 24 hr prior to assay initiation and continuing throughout . Overnight cultures of C . neoformans ( 14–16 hr ) were grown in YNB media , after which cells were isolated , washed in DMEM and resuspended in BMDM growth media . Next , cells were opsonized with mAb1255 ( 10 μg/ml ) at 37°C for 1 hr . Cryptococcus cells were added to macrophages at an MOI of 10 , and this concentration was confirmed by plating yeast serial dilutions on rich media . After 24 hr at 37°C and 5% CO2 , cells were washed three times with PBS to remove non-adherent yeast . Finally , ~200 BMDMs were quantified per well , with 6 wells per genotype , to determine the fraction of yeast-associated macrophages ( phagocytic index ) . Cryptococcus accumulation within macrophages was assessed as described previously [80] . Briefly , BMDMs were plated in 24 well plates at a concentration of 100 , 000 cells/well . Stimulation was performed as above , after which macrophages were exposed to opsonized C . neoformans at an MOI of 0 . 1 . After 24 hr at 37°C and 5% CO2 , supernatants were removed and macrophages were lysed . Serial dilutions were plated to determine CFU . The ratio of yeast present at 24 hr versus input was determined and analyzed by bootstrapping , generating 95% confidence intervals . C . neoformans strains were grown in liquid YNB cultures overnight ( 14–16 hr ) , and then centrifuged and washed twice in PBS . For competitive co-infection experiments , mixtures of a wild-type strain and a deletion strain of interest were prepared by determining cell concentration using a hemocytometer and then mixing strains in a 1:1 ratio to a final concentration of 1x107 cells per ml PBS . Concentrations of viable cells were confirmed by plating serial dilutions . A/J female mice ( Jackson Laboratory ) aged 5–6 weeks were anesthetized by intraperitoneal injection of ketamine ( 75 mg/kg ) and dexmedetomidine ( 0 . 5 mg/kg ) , then suspended from a silk thread by their front incisors , as described previously [81] . Intranasal infections of 50 μl were delivered by pipette , resulting in a dose of 5x105 cells . After an additional 10 minutes of suspension , the mice were lowered and anesthesia reversed by intraperitoneal injection of atipamezole ( 1 . 5 mg/kg ) . Three mice were infected with each C . neoformans genotype , and were monitored until a defined terminal time point of ten days after infection . At this time , mice were sacrificed by CO2 inhalation and cervical spine dislocation . Next , lungs were harvested and homogenized in PBS using a PRO200 homogenizer ( Grainger ) . The ratios of C . neoformans strains in the input and organ samples were determined by plating in serial dilutions on Sabouraud agar plates containing 40 mg/ml gentamicin and 50 mg/ml carbenicillin , and then testing the nourseothricin resistance status of ~200 colonies . As a negative control , mice were infected with a 1:1 ratio of wild-type cells and a sxi1Δ strain , which is known to have a wild-type phenotype [63] . For monotypic infections , female A/J mice were intranasally infected with 50 μl PBS containing C . neoformans cells of a single genotype at a concentration of 1 . 0x107 cells per ml , as described above . Concentrations of viable cells were confirmed by plating serial dilutions . Ten mice were infected per genotype , and were monitored until severe morbidity ( as indicated by a loss of 15% of initial body weight or other symptoms ) , at which point they were sacrificed . Survival data was analyzed using the Online Application for the Survival Analysis of Lifespan Assays Performed in Aging Research [82] . | Many pathogenic organisms secrete peptidases . The activity of these enzymes often contributes to virulence , making their study crucial for understanding host-pathogen biology and developing therapeutics . In this report , we employed an unbiased , activity-based profiling assay to examine the secreted peptidases of a fungal pathogen , Cryptococcus neoformans , which is responsible for 40% of AIDS-related deaths . We discovered which peptidases are secreted , identified their substrate specificity , and interrogated their biological functions . Through this analysis , we identified a principal enzyme responsible for the extracellular peptidase activity of C . neoformans , May1 , and demonstrated its importance for growth in acidic environments . Characterization of its substrate preferences allowed us to identify compounds that are potent substrate-based inhibitors of May1 activity . Finally , we found that the presence of this enzyme promotes virulence in a mouse model of infection . Our comprehensive study reveals the expression , regulation and function of C . neoformans secreted peptidases , including evidence for the role of a novel aspartyl peptidase in virulence . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"blood",
"cells",
"cryptococcus",
"neoformans",
"medicine",
"and",
"health",
"sciences",
"cryptococcus",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"enzymes",
"pathogens",
"animal",
"models",
"of",
"disease",
"immunology",
"enzymology",
"microbiology",
"fungi",
"peptide",
"libraries",
"fungal",
"pathogens",
"research",
"and",
"analysis",
"methods",
"animal",
"models",
"of",
"infection",
"mycology",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"animal",
"studies",
"microbial",
"pathogens",
"proteomics",
"biochemistry",
"peptides",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"proteases",
"cellular",
"types",
"macrophages",
"organisms"
]
| 2016 | Integrated Activity and Genetic Profiling of Secreted Peptidases in Cryptococcus neoformans Reveals an Aspartyl Peptidase Required for Low pH Survival and Virulence |
Hebbian plasticity describes a basic mechanism for synaptic plasticity whereby synaptic weights evolve depending on the relative timing of paired activity of the pre- and postsynaptic neurons . Spike-timing-dependent plasticity ( STDP ) constitutes a central experimental and theoretical synaptic Hebbian learning rule . Various mechanisms , mostly calcium-based , account for the induction and maintenance of STDP . Classically STDP is assumed to gradually emerge in a monotonic way as the number of pairings increases . However , non-monotonic STDP accounting for fast associative learning led us to challenge this monotonicity hypothesis and explore how the existence of multiple plasticity pathways affects the dynamical establishment of plasticity . To account for distinct forms of STDP emerging from increasing numbers of pairings and the variety of signaling pathways involved , we developed a general class of simple mathematical models of plasticity based on calcium transients and accommodating various calcium-based plasticity mechanisms . These mechanisms can either compete or cooperate for the establishment of long-term potentiation ( LTP ) and depression ( LTD ) , that emerge depending on past calcium activity . Our model reproduces accurately the striatal STDP that involves endocannabinoid and NMDAR signaling pathways . Moreover , we predict how stimulus frequency alters plasticity , and how triplet rules are affected by the number of pairings . We further investigate the general model with an arbitrary number of pathways and show that depending on those pathways and their properties , a variety of plasticities may emerge upon variation of the number and/or the frequency of pairings , even when the outcome after large numbers of pairings is identical . These findings , built upon a biologically realistic example and generalized to other applications , argue that in order to fully describe synaptic plasticity it is not sufficient to record STDP curves at fixed pairing numbers and frequencies . In fact , considering the whole spectrum of activity-dependent parameters could have a great impact on the description of plasticity , and a better understanding of the engram .
Synaptic plasticity , one of the paramount biological mechanism supporting learning and memory in the brain , has been the object of a wide literature spanning from experimental works [1–3] to computational investigations [4–6] . In 1949 , Donald Hebb’s pioneering work postulated that long-term modifications of the synaptic efficacy can be induced in response to patterns of activity of the pre- and postsynaptic neurons [7] . Since then , many experimental studies have confirmed and extended Hebb’s postulate and have highlighted the complexity of the signaling pathways and their neuromodulation leading to synaptic efficacy changes in response to pre- and postsynaptic activity patterns [1 , 2 , 8 , 9] . Numerous mathematical models were also developed to emulate this diversity and infer their computational capabilities [4–6] . Spike-timing-dependent plasticity ( STDP ) is a biological phenomenon of activity-dependent change in synaptic connectivity that is viewed as a synaptic Hebbian learning rule . STDP has been widely studied in the last two decades and experimentally observed at many synapses in various forms , and those were classified depending on the manner in which they implement Hebb’s postulate [8 , 9] . STDP is assessed experimentally through repetitive paired activations of the pre- and postsynaptic sites with a prescribed timing that is denoted in this paper Δt . By convention , we consider Δt < 0 when the postsynaptic stimulation occurs before the paired presynaptic one ( post-pre pairings ) , and Δt > 0 when the presynaptic stimulation occurs before the postsynaptic one ( pre-post pairings ) . Classically , the same paired stimulation is repeated between 80 and 150 times at a constant frequency ( between 0 . 1 and 5 Hz ) [8–10] . In many cases , these pairing patterns induce long-term plasticity exhibiting various polarities ( increase or decrease of the synaptic weight as a function of the sign of Δt ) or magnitudes . In the vast majority , the expression of STDP is restricted to a narrow interval of values for Δt; thus , when pre- and postsynaptic activities are separated by a large Δt ( |Δt| > 50 ms in most of the cases ) , no long-term plasticity is observed [11 , 12] . In this study , we term Hebbian STDP the plasticities whereby sequences of presentations of a presynaptic spike followed by a postsynaptic spike lead to Long-Term Potentiation ( LTP ) when repeated a specific number of times ( denoted NPairings ) at a certain frequency , whereas reverse sequences induce Long-Term Depression ( LTD ) . Hebbian STDP was reported in various structures such as the hippocampus [11 , 13–15] , the cerebral cortex [12 , 16–19] and the striatum [20–23] . Conversely , we will term here ( bidirectional ) anti-Hebbian STDP , the forms of STDP exhibiting a reverse polarity when compared to the aforementioned Hebbian STDP: causal pre-post pairings induce LTD and anti-causal post-pre pairings induce LTP . Bidirectional anti-Hebbian STDP was also observed , for instance in the striatum [24–28] , in the somatosensory cortex [29] or in the cerebellum-like structure of the electrical fish [30] . Unidirectional anti-Hebbian STDP , i . e . LTD for both post-pre and pre-post pairings , is another main form of STDP observed in the neocortex [31 , 32] , the dorsal cochlear nucleus [33] , the cerebellum [34 , 35] and the hippocampus [36] . We underline that different definitions of ( anti- ) Hebbian STDP were used in the literature; the present study follows the terminology of early experimental studies [11 , 12] , or Figure 2 of the review [8] , but differs , e . g . , from the definitions used in [37] . These plasticities were shown to be dependent upon the parameters of the stimulation beyond spike-timing: for instance , varying the frequency at which pairings are presented or the total number of pairings , presenting distinct spike patterns ( triplets , single spike , theta bursts , … ) [17 , 38–41] or changing neuromodulatory tones [21] may lead to distinct forms of STDP . Despite the existence of multiples forms of STDP [8 , 9] , all of them have in common the crucial role played by the calcium transients in the pre- and postsynaptic compartments for the induction and maintenance of plasticity . Postsynaptic calcium influxes through NDMA receptors ( NMDAR ) and voltage-sensitive calcium channels have been demonstrated to be key factors governing plasticity expression and polarity [10] . Regarding Hebbian plasticity , calcium-dependent mechanisms act as coincidence detectors , essential to implement any type of STDP . In addition , distinct signaling pathways appear to be involved , namely ( i ) calcium triggering downstream cascades modulating calcium/calmodulin-dependent kinase II ( CaMKII ) [42] which ultimately regulates the gene expression and/or ( ii ) the endocannabinoid ( eCB ) system , whose synthesis and release is calcium-dependent , acting retrogradely on the presynaptic element [43–45] . Importantly , both of these pathways are able to trigger LTP or LTD depending on the spatio-temporal kinetics of the calcium [19 , 40 , 46] . Calcium dynamics thus constitute a key factor in synaptic plasticity induction and in selecting plasticity forms . Accordingly , numerous mathematical models were based on calcium transients and described various forms of STDP [5] . In particular , Graupner and Brunel [47] proposed simple calcium-based models able to account for a wide range of experimental observations on synaptic plasticity . However , while calcium-based models succeed in reproducing the results of the “classical” STDP ( ∼ 100 pairings ) , they do not take into account the dynamics of the establishment of plasticity and the variety of timescales involved in plasticity induction . Indeed , in computational neuroscience , it is implicitly admitted that the synapse gradually amplifies synaptic changes as the number of stimulus presentation increases to reach the final plasticity profiles . However , plasticity occurs at vastly distinct timescales and protocols based on one hundred trials ( i . e . , pairings ) , as classically performed in STDP experiments , only reveal an extreme steady state outcome . Actually , a dozen of trials can be sufficient to induce plasticity , if not less in the case of associative memory [39 , 40 , 48 , 49] . Importantly , it was recently shown that depending on the number of pairings , STDP on the cortex-to-striatum synapses ( cortico-striatal STDP ) exhibits three distinct forms of plasticity: an NMDAR-mediated LTP and an eCB-mediated LTD for 100 post-pre and pre-post pairings , respectively [20 , 21 , 24 , 25 , 27] , and an eCB-mediated LTP for 5-15 post-pre pairings [39 , 40] . Note that at cortico-striatal synapses , GABA operates as a Hebbian/anti-Hebbian switch [27 , 28] and that without the blockade of GABA transmission , an anti-Hebbian STDP is induced as observed in vivo [26] . These phenomena were reproduced in a biophysical model of the cortico-striatal synapse accounting for receptors activation dynamics ( 36 equations and 150 parameters ) [40] . However , no simple phenomenological model reproduces these phenomena , and in particular models of plasticity based on the calcium hypothesis fail to reproduce such complex dynamical emergence of plasticity . Here , we propose a new model built upon the calcium hypothesis and taking into account the existence of multiple signaling pathways at a given synapse that may be activated at distinct calcium levels . We instantiate one model to fit the data from cortico-striatal STDP [39 , 40] , and show that the model accurately reproduces the experiments on the dependence of STDP on both the number and frequency of pairings . We use this model to predict the response of the system as the stimulus frequency and number of presentations are varied , and extend the model to show how triplet rules will depend on the number of stimulus presentations . Our model goes beyond the case of the cortico-striatal synapse with two signaling pathways , and we further explore the diversity and limited range of dynamical plasticity establishments that can be unfolded from classical Hebbian STDP . In the face of this diversity , we further propose experimentally implementable protocols to differentiate those scenarii . This study thus sheds a new light on the interplay of multiple signaling pathways at single synapses and how this multiplicity endows the synapse with the capacity of encoding multiple STDP profiles depending on the number and frequency of stimulus presentation , and argues that experiments based on stereotypical stimulus presentations are not sufficient to finely account for the complexity of plasticity , even in widely studied synapses .
We developed a general calcium-based model of the synapse allowing to take into account the presence of multiple pathways and past activity in the establishment of plasticity . Our developments build upon the Graupner and Brunel model [47] , and extend it by ( i ) introducing multiple plasticity pathways , and ( ii ) taking into account the fact that receptor activation thresholds may depend on past activity . We provide here the details of the models and the emergent changes in synaptic weight , as well as a theoretical formula thereof . The model we built in the previous section is general and is thus able to reproduce a variety of synapses and plasticity mechanisms relying on calcium dynamics . We study in this section the case of STDP at the cortico-striatal synapse , which was studied experimentally with varying Npairings [39 , 40] . In these contributions , it was shown that STDP at the cortico-striatal synapse relied both on NMDAR and endocannabinoid pathways ( see Fig 1 ( a ) ) , and that synaptic changes after paired pre- and postsynaptic spikes not only depended on the timing between the pre- and postsynaptic spikes , but also varied with the number and the frequency of the pairings presented . Namely , it was shown that for pre-post pairings ( 0 < Δt < +40 ms ) , an eCB-LTD progressively appeared as the number of pairings was increased , while for post-pre pairings ( −30 ms < Δt < 0 ms ) , a biphasic STDP emerged with an eCB-LTP for a low number of pairings ( 5 − 15 pairings ) , an absence of plasticity between 25 and 50 pairings , leaving room for NMDAR-LTP at higher numbers of pairings ( ≥ 75 pairings ) . A schematic representation of the biological pathways involved is provided in Fig 1 ( a ) together with the biophysical mechanisms and proteins cascades ( described in more detail in [40] ) . A minimal model of cortico-striatal plasticity thus requires taking into account two different and independent calcium-dependent pathways ( P = 2 ) , an eCB-dependent mechanism ( α = e ) which induces both LTP or LTD depending on the specific timing Δt of the pairings , and an NMDAR-dependent ( α = n ) associated to LTP only . This yields to the following system of stochastic differential equations ( see schematic diagram in Fig 1 ( b ) ) : { τdρedt=−ρe ( 1−ρe ) ( ρ*−ρe ) +γep ( 1−ρe ) Θ[ c ( t ) −θep ( c˜t ) ]−γedρeΘ[ c ( t ) −θed ( c˜t ) ]+Noisee ( t ) τdρndt=−ρn ( 1−ρn ) ( ρ*−ρn ) +γnp ( 1−ρn ) Θ[ c ( t ) −θnp ( c˜t ) ]+Noisen ( t ) ( 6 ) The complete synapse model is made of N ≫ 1 independent pairs ( ρ e i , ρ n i ) i = 1 ⋯ N satisfying Eq ( 6 ) , and a synaptic change deduced from the proportion of synapses that switch from being potentiated to depressed or reciprocally , through the sigmoidal map H of Eq ( 4 ) . As described above , in response to pre- and postsynaptic spike-timing ( Δt = tpost − tpre ) , the calcium dynamics c undergoes jumps followed by exponential relaxation as described in Eq ( 2 ) , activating eCB-LTP , eCB-LTD and NMDAR-LTP as soon as c exceeds specific LTP or LTD thresholds ( see Fig 1 ( c ) , where the LTP and LTD thresholds are represented by the green and red lines respectively ) with both thresholds taken from the adjusted parameters of Table 1 for the eCB pathway . When only one presynaptic spike ( thus without postsynaptic spike ) is evoked , the calcium concentration amplitude exceeds the LTP threshold for a short amount of time , and remains below the level of LTD induction: repeating this protocol does not lead to significant plasticity . For a pre-post stimulation ( Δt > 0 ) , the summation of the pre- and postsynaptic calcium spikes triggers both LTP and LTD . The same is valid for Δt < 0 , but the relative time spent above the LTP and LTD thresholds would be significantly different depending on Δt , underlining the importance of the timing and order between the spikes in the resulting plasticity: in the example depicted in Fig 1 ( c ) and parameters γ α x from Table 1 , a pre-post stimulation yields LTD whereas a post-pre stimulation yields LTP , consistent with anti-Hebbian STDP at cortico-striatal synapses ex vivo [27 , 28] in the absence of GABAA receptor antagonist and in vivo [26] . STDP at the cortico-striatal synapse , studied in the previous section , provides a realistic example of plasticity with multiple pathways . Our model , relying on only two equations and a small number of biologically interpretable parameters emulating NMDAR- and eCB-dependent pathways , reproduces all the phenomena reported at the cortico-striatal synapse , and allowed to draw predictions on plasticity for more complex stimuli such as triplet rules . The present model is however much more general than the case of the cortico-striatal synapse: it can indeed emulate synapses with more than two signaling pathways with arbitrary independent plasticity rules , and thus allows unraveling the dynamics of plasticity expression in a variety of synapses with distinct plasticity . Interestingly , while being quite versatile , the repertoire of behaviors that can be reproduced given a fixed number of pathways remains limited , and the model thus also provides predictions on the minimal number of pathways involved given a plasticity profile . Indeed , a single pathway shall induce a monotonic establishment of plasticity if there is no inactivation of the pathway , whereas situations with two pathways can lead to four changes of plasticity ( LTP and LTD inactivation for each of the two pathways ) , possibly with periods of non-significant synaptic changes . More generally , plasticity with P pathways may lead to up to 2P changes of monotonicity , possibly interspersed with periods of non-significant plasticity . We investigate in the next sections a few possible scenarii relying on at most two signaling pathways that could lead to Hebbian or anti-Hebbian plasticity and suggest experiments that could distinguish distinct situations .
Synaptic plasticity is a complex phenomenon relying on the activation of a number of receptors and signaling pathways [3 , 10] . A substantial difficulty for experimentalists is to characterize plasticity in the large variety of possible situations occurring in vivo . To reduce this complexity , a protocol designed to reveal plasticity consists in considering changes in synaptic transmission after the reiterated presentation of a fixed spike pattern a large number of times ( on the order of one hundred ) and at a slow rate . From these experiments , it remains complex to decipher the multiple signaling pathways involved in the expression of plasticity , and their complex interplay , particularly for low numbers of stimulus presentations or for various pairing frequencies . To disentangle the distinctive role of multiple pathways , we developed and studied a phenomenological model of the evolution of synaptic weights and tested its responses in distinct situations . The model relies on calcium transients triggered by the spiking activity of neurons on both sides of the synapses , and is built upon previous theoretical works ( see [47] and references therein ) . When plasticity ( LTP and LTD ) relies on multiple signaling pathways [10] , the timescales at which these mechanisms activate and inactivate upon repetitive stimulation can lead to a variety of behaviors as a function of the number and of the frequency of pairings , which cannot be inferred from experiments where those are fixed . Our model proposes a general and minimal framework to integrate multiple signaling pathways and their dependences upon repetitive stimulations . We have instantiated this model with two specific pathways , NMDAR- and eCB-dependent , that was inspired by experiments at cortico-striatal synapse showing variations of the emergent plasticity upon variation of the number of pairings [39 , 40] . Our model reduces to two stochastic equations Eq ( 6 ) and a small number of parameters , and accurately reproduced the data obtained in that experimental contribution . To our knowledge , this model is the most parsimonious model reproducing STDP experimental results , yet many models of the class that we introduced can be proposed , including for instance NMDAR-LTD or pathways activated by distinct molecules . We also used the model to predict the response of the system when the number of stimulations , the pairing frequency or the number of spikes , are varied . This led us to draw predictions on the modifications of STDP profiles when the frequency of stimulus presentations was varied . Eventually , we have made new predictions on the dependence of triplet rules upon the number of stimulus presentations , and showed that complex non-monotonic STDP profiles emerge with up to three distinct phases . Our model goes beyond the particular case of the cortico-striatal synapse for which data was available , and we pursued our investigations by considering distinct mechanisms that could underlie another type of plasticity , symmetric anti-Hebbian LTD ( with LTD for pre-post and post-pre pairings ) . In this case , we investigated three distinct possible scenarii involving up to two distinct pathways , and showed that unexpected phenomena may arise upon variations of the number and frequency of pairings , and in particular the emergence of an LTP at 100 pairings for high frequencies . Overall , these results highlight the fact that electrophysiological experiments at a fixed frequency and a prescribed number of pairings may not be sufficient to extrapolate to other situations with smaller numbers of pairings or presentation frequencies . To our knowledge , the present model is the first to take into account distinct signaling mechanisms involved in plasticity in a simple and compact framework . The simplicity of the present model allows to envision the implementation of this type of synapse at the level of a neural network , opening the way to theoretical studies of information processing capacity of networks endowed with complex activity-dependent plasticity rules . In addition to the development of a framework integrating multiple pathways , one of the main novelties of this model compared to other calcium-based models is that we have explicitly incorporated activity-dependent thresholds allowing to recover the response of plasticity mechanisms on the past activity of cells . In the present model , we simply assumed that this history-dependence is parameterized by a cumulative calcium concentration . Explicitly incorporating this dependence allows taking into account in the model multifarious experimental facts including finiteness of the calcium pool in the postsynaptic compartment , desensitization of synaptic-receptors and homeostasic mechanisms [63] . The present model proposing that this dependence on past activity relies on cumulative calcium constitutes a first step , and could be refined in several directions , for instance incorporating a slow decay of past-activity dependence with time ( considering integrated calcium spikes with an exponentially decaying kernel for instance ) , moving averages in the flavor of sliding thresholds in the classical Bienenstock-Cooper-Munro ( BCM ) rule [56 , 57] . In our case , the average activity of the neuron would be simply modeled by postsynaptic calcium concentration ( a reasonable proxy of neural activity ) , or with more refined models involving distinct molecular species and their timescales . Despite a good qualitative agreement and an improved accuracy on the dynamics of the expression of plasticity , we found that our model shows a slight mismatch in the timescales at which plasticity emerges: first , although experiments at cortico-striatal synapses show a significant plasticity arising as early as 5 pairings and reaching a maximum at 10 pairings , we did not find in the model significant plasticity at 5 pairings and the maximal plasticity occurred after a slightly larger number of 12 pairings . Moreover , a unidirectional LTP in the cortico-striatal plasticity at 100 pairings was observed experimentally when the frequency of pairing presentations reached 4 Hz , while the model reproduces this phenomenon slightly above 30 Hz . We believe that this slower response of the present model relies on the bistable nature of the model , following [47] . This bistability makes the system quite rigid and resistant to rapid changes , and a direct perspective would be to implement a more flexible model dropping the bistable model but conserving the long-term stability of macroscopic synaptic strength ensured by the bistable potential . The present model would be also used in future works focusing on the implementation of the cortico-striatal STDP in large stochastic neural networks , with several classes of interneurons , aimed at understanding the possible role of implementing distinct cortico-striatal plasticity , in particular LTP , arising at various timescales and their possible role in information processing in striatum . All in all , the present model suggests to reconsider a current widely admitted implicit hypothesis in models , and questions the usual view of STDP in models that consider a fixed curve solely dependent on the spike timing ( Δt ) . Indeed , in most neural network models with STDP , it is considered that synaptic coefficients are progressively incremented depending on spike timings and according to toy-models of STDP ( e . g . , double-exponential curves ) . This is implemented in various manners , including additive or multiplicative changes depending on all spike pairings or on the nearest-spike ( see e . g . [64] ) . At the level of networks , a number of stochastic models were developed to study the influence of STDP as a synaptic plasticity rule ( see the review [4] ) . In particular , early works showed the influence of classical Hebbian and asymmetric STDP in the dynamics of neuronal networks [65–67] . The role of STDP-based rules in the emergence of structures in recurrent neural networks was also studied in a series of papers , highlighting for instance a possible key impact on the self-organization of microcircuits [68–70] . More recently , the distribution of synaptic weights and its stability in randomly stimulated networks with different triplet rules has been extensively studied [37] . The activity-dependent rule we proposed here , reproducing variable synaptic changes as a function of the number of stimulations , may lead to significant changes in the resulting connectivity and dynamics of neural networks . Our model offers an avenue to revaluate the possible modifications of the resulting dynamics emphasizing the role of timescales in these systems [71] .
Simulations were performed with a custom code implemented in Python 2 . 7 or 3 . 5 , developed within the Spyder environment of the Anaconda suite ( Anaconda Software Distribution . Computer software . Vers . 2-2 . 4 . 0 . Continuum Analytics , Nov . 2015 . Web . <https://continuum . io> ) . The main modules used numpy , matplotlib , math , scipy . Elementary simulations of the model were run on a Macbook Pro ( Intel Core i5 processor and 16 RAM ) and more demanding simulations were executed on the Inria Paris-Rocquencourt computer cluster . Figures and plots were realized using matplotlib module of Python and Illustrator/Photoshop of the Adobe series . Unless specified otherwise , we used the parameter values listed in Table 1 . These parameters were optimized starting from initial guesses chosen for consistency of the model with the data using the extensive analysis of [47 , Fig 2] . For adjusting our parameters to the cortico-striatal plasticity , we used a global optimization algorithm , the differential-evolution function from scipy . optimize module to obtain qualitative fits . Simulations of the model were realized either from the theoretical expressions computed , or with numerical simulations of the system of stochastic equations Eq ( 6 ) . We used temporal discretization using an Euler scheme on t = −1…101 s with Niter steps ( see Table 1 ) and run the simulation for a set of N = 1000 individual efficacies . To compute the change in macroscopic synaptic strength for the different pairings , we run the simulation for NPairings = 100 pairings and store the results for all the pairings during the STDP protocol . Therefore , for each fixed Δt , the results obtained for different pairings are not independent , which has the interest of uncovering the evolution of one given synapse , and has no impact on the global outcome of the simulations as can be seen when compared with analytical results . Except for Fig 3 where the analytical mean-field solutions are represented , all the figures show the numerical simulations . For Fig 6 , we have reproduced 30 independent simulations in parallel to obtain the statistical means and standard deviations depicted . All heatmaps used a logarithmic color bar to represent changes in synaptic strength . The classification in mono- , bi- or tri-phasic regimes in Fig 7 ( c ) was done through a visual inspection of the STDP curves associated to each of the 20 Δt1 and Δt2 . The model we studied is nonlinear , and as such , it was complex to derive the explicit form of the probability distribution of the solutions . Following the approach proposed in the Appendix of [47] , we derived the probability distribution of the solution of an approximate model valid when the system remains in the linear part of the cubic bistable term . The model involves a linear Ornstein-Uhlenbeck with deterministic time-dependent coefficients α ( t ) and β ( t ) that we computed as follow . The solution of linear stochastic differential equations of type [72] ( with B a standard Brownian motion ) : d ρ ( t ) = ( α ( t ) ρ ( t ) + β ( t ) ) d t + σ ( t ) d B ( t ) ( 7 ) with initial condition ρ0 can be easily expressed in closed form as: ρ ( t ) = ρ 0 exp ( ∫ 0 t α ( s ) d s ) + ∫ 0 t exp ( ∫ u t α ( s ) d s ) β ( u ) d u + ∫ 0 t exp ( ∫ u t α ( s ) d s ) σ ( u ) d B ( u ) . ( 8 ) As indicated in the main text , the synaptic change is obtained using a sigmoid transform of the proportion U of synapses that , after the protocol , have crossed upwards the threshold value ρ* , over the proportion D of those that crossed downwards . Since the above described Ornstein-Uhlenbeck process is Gaussian , these probabilities are fully characterized by the mean and single-time variance functions of ρ , which have the following expressions: E [ ρ ( t ) | ρ 0 ] = ρ 0 exp ( ∫ 0 t α ( s ) d s ) + ∫ 0 t exp ( ∫ u t α ( s ) d s ) β ( u ) d u ( 9 ) and Var [ ρ ( t ) | ρ 0 ] = ∫ 0 t exp ( 2 ∫ u t α ( s ) d s ) σ 2 ( u ) d u ( 10 ) We thus derive the time-varying coefficients α and β arising in the approximated model , for the eCB pathway ( the other can be dealt with in the same way ) . These are computed describing the time spent above the various thresholds of the model . We denote by η i x the average time spent above the threshold θ i x: this quantity only depends on the calcium dynamics can be easily computed analytically for each given a pairing protocol . Similarly , we define te = T ne the time at which the eCB-LTP is inactivated at the cortico-striatal synapse , with T being the duration between two pairings and ne the pairing number at which eCB-LTP is first inactivated . Denoting Γ i x = γ i x η i x , we have the following compact formulae for the coefficients of the Ornstein-Uhlenbeck processes αe , βe and σe: α e ( t ) = { - Γ e d + Γ e p τ = - 1 τ 1 if t < t e - Γ e d τ = - 1 τ 2 if t e < t ( 11 ) β e ( t ) = { Γ e p τ = ρ ˜ 1 τ 1 si t < t e 0 si t e < t ( 12 ) σ e ( t ) = { η e d + η e p τ σ = σ 1 τ 1 si t < t e η e d τ σ = σ 2 τ 2 si t e < t ( 13 ) Because of the simple , piecewise constant form of the coefficients , we have , for deterministic initial conditions: E [ ρ e ( t ) | ρ e ( 0 ) ] = { ρ e ( 0 ) exp ( - t τ 1 ) + ρ ˜ 1 ( 1 - exp ( - t τ 1 ) ) if t < t e ρ e ( 0 ) exp ( - t - t e τ 2 - t e τ 1 ) + ρ ˜ 1 exp ( - t - t e τ 2 ) ( 1 - exp ( - t e τ 1 ) ) if t e < t ( 14 ) and Var [ ρ e ] = { σ 1 2 ( 1 - exp ( - 2 t τ 1 ) ) t < t e σ 1 2 exp ( - 2 t - t e τ 2 ) ( 1 - exp ( - 2 t e τ 1 ) ) + σ 2 2 ( 1 - exp ( - 2 t - t e τ 2 ) ) if t e < t ( 15 ) The probability that an initially depressed synapse becomes potentiated is thus given by: U e ( n T ) = P ( ρ e > ρ * | ρ e ( 0 ) = 0 ) = 1 2 ( 1 + erf ( − ρ * − E [ ρ e | ρ e ( 0 ) = 0 ] ( n T ) Var [ ρ e ] ( n T ) ) ) , ( 16 ) and the probability of an initially potentiated synapse to become depressed by: D e ( n T ) = P ( ρ e < ρ * | ρ e ( 0 ) = 1 ) = 1 2 ( 1 − erf ( − ρ * − E [ ρ e | ρ e ( 0 ) = 1 ] ( n T ) Var [ ρ e ] ( n T ) ) ) . ( 17 ) allowing directly to obtain the change in synaptic weight associated as H ( U e ( n T ) D e ( n T ) ) . A comparison of the Ornstein-Uhlenbeck approximation with the numerical simulations of the nonlinear system is provided in Fig 3 ( a ) and S2 ( a ) Fig , showing a good agreement for the parameter set chosen . The data used to fit and validate our results were previously published in [39 , 40] . We refer to these papers for more specific information on the experimental protocol . | The brain’s capacity to treat information , learn and store memory relies on synaptic connectivity patterns , which are altered through synaptic plasticity mechanisms . Experimentally , such plasticities were evidenced through protocols involving numerous repetitive stimulations of a given synapse , and were shown to be supported by multiple pathways . Using a simple biologically grounded mathematical model , we show how activation timescales and inactivation levels of each pathway interact and alter plasticity in an intricate manner as stimuli are presented . Building upon data from the synapse between cortex and striatum , we show that synaptic changes may revert or re-emerge as stimuli are presented , and predict specific responses to changes in stimulus frequency or to distinct simulation patterns . Our general model shows that a given plasticity profile emerging in response to a repetitive stimulation protocol can unfold into various scenarii upon variations of the number of stimulus presentations or patterns , which tightly depends on the underlying activated pathways . Altogether , these results argue that in order to better understand learning and memory , single plasticity responses obtained through intensive stimulations do not reveal the complexity of the responses for smaller number of presentations , which may have a strong impact in fast learning of stimuli with low numbers of presentations . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"medicine",
"and",
"health",
"sciences",
"action",
"potentials",
"neural",
"networks",
"nervous",
"system",
"membrane",
"potential",
"electrophysiology",
"neuroscience",
"simulation",
"and",
"modeling",
"synaptic",
"plasticity",
"calcium",
"signaling",
"neuronal",
"plasticity",
"research",
"and",
"analysis",
"methods",
"developmental",
"neuroscience",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"signal",
"transduction",
"cellular",
"neuroscience",
"anatomy",
"synapses",
"cell",
"biology",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"cell",
"signaling",
"neurophysiology"
]
| 2018 | Interplay of multiple pathways and activity-dependent rules in STDP |
Kaposi’s sarcoma-associated herpesvirus ( KSHV ) is the causative agent of Kaposi's sarcoma ( KS ) and primary effusion lymphoma ( PEL ) , which are aggressive malignancies associated with immunocompromised patients . For many non-viral malignancies , therapeutically targeting the ubiquitin proteasome system ( UPS ) has been successful . Likewise , laboratory studies have demonstrated that inhibition of the UPS might provide a promising avenue for the treatment of KSHV-associated diseases . The largest class of E3 ubiquitin ligases are the cullin-RING ligases ( CRLs ) that are activated by an additional ubiquitin-like protein , NEDD8 . We show that pharmacological inhibition of NEDDylation ( using the small molecule inhibitor MLN4924 ) is cytotoxic to PEL cells by inhibiting NF-κB . We also show that CRL4B is a novel regulator of latency as its inhibition reactivated lytic gene expression . Furthermore , we uncovered a requirement for NEDDylation during the reactivation of the KSHV lytic cycle . Intriguingly , inhibition prevented viral DNA replication but not lytic cycle-associated gene expression , highlighting a novel mechanism that uncouples these two features of KSHV biology . Mechanistically , we show that MLN4924 treatment precluded the recruitment of the viral pre-replication complex to the origin of lytic DNA replication ( OriLyt ) . These new findings have revealed novel mechanisms that regulate KSHV latency and reactivation . Moreover , they demonstrate that inhibition of NEDDylation represents a novel approach for the treatment of KSHV-associated malignancies .
The ubiquitin-proteasome system ( UPS ) and associated pathways are rapidly becoming accepted as major therapeutic targets for the treatment of malignancy [1] , which potentially include those associated with oncogenic viruses . Additionally , small molecule inhibitors have been successfully used for dissecting the biological roles of these intriguing pathways , which is critical for our understanding of their mechanisms of cytotoxicity . Indeed , inhibition of the UPS is cytotoxic to Kaposi’s sarcoma-associated herpesvirus ( KSHV , also referred to as human herpesvirus 8 [HHV8] ) infected cells [2–5] . Infection with KSHV is commonly associated with fatal malignancies , is the causative agent of primary effusion lymphoma ( PEL ) and Kaposi’s sarcoma ( KS ) and is frequently associated with multicentric Castleman’s disease ( MCD ) [6 , 7] . Like all herpesviruses , KSHV infection is lifelong and has two distinct phases to its lifecycle; latent and lytic . During latency , viral gene expression is highly restricted and , in the tumor setting , involves the expression of the latency associated nuclear antigen ( LANA ) , the viral FLICE inhibitory protein ( vFLIP ) , viral cyclin , kaposin and various virally encoded miRNAs . Together these promote tumorigenesis in all known KSHV-associated malignancies . Nevertheless , at least for KS , the lytic phase of KSHV , which results in the expression of the complete viral genome and the production of infectious virions , is necessary for sarcomagenesis . For this reason , the molecular mechanisms governing the switch from latency to lytic reactivation have received much attention as they may provide excellent targets for therapeutic intervention . Current treatments of KSHV-associated malignancies have limited efficacy . PEL is treated using a combination of cyclophosphamide , doxorubicin , vincristine and prednisone ( similar to CHOP therapy ) and/or highly active retroviral therapy ( HAART ) [8 , 9] . For AIDS-related KS , HAART is also favored , and due to the requirement of KSHV lytic infection for the pathogenesis of KS , anti-herpesviral drugs have also been used [10] . More recently , preclinical models have demonstrated that inhibition of the UPS using bortezamib [2–5] , or bortezamib in combination with a histone deacetylase ( HDAC ) inhibitor ( vorinostat ) may provide a promising new avenue [11] . Given the success of bortezamib ( marketed as Velcade ) for the treatment of multiple myeloma and mantle cell lymphoma , there are now various additional small molecule inhibitors of the UPS at various stages of development [1] , as well as drugs that target ubiquitin-like pathways such as the NEDDylation cascade [12] . The attachment of ubiquitin-like ( Ubl ) proteins to substrates represents one of the most vital posttranslational modifications in the cell and , so far , twelve Ubls have been identified including ubiquitin ( Ub ) , NEDD8 and SUMO . The covalent attachment of Ubls to substrates requires an enzyme cascade involving an E1 activating enzyme , an E2 conjugating enzyme and an E3 ligase that provides substrate specificity , all of which are potential ‘drugable’ targets . Ubiquitylation is the principal mechanism by which proteins are targeted for proteasomal degradation , and the largest class of E3 ligases are the Cullin-RING ligases ( CRLs ) [13] . These modular proteins are centered on one of several Cullin scaffolds ( Cul1 , 2 , 3 , 4A , 4B , 5 and 7 ) and consist of a RING-binding protein ( RBX1 or RBX2 ) that interacts with the Ub-loaded E2 and a substrate adapter protein that brings the target into close proximity with the Ub-E2 . Many substrate adapters have been identified , and the various combinations of Cul-adapter complexes means there are several hundred possible CRLs , with several-fold more substrates [14] . Intriguingly , the activity of CRLs is dependent on themselves being modified by the Ubl NEDD8 , which requires the activity of the NEDD8 activating enzyme , NAE1 [15] . Recently , a specific NAE1 inhibitor has been developed ( MLN4924 ) and is currently in clinical trials for various malignancies . MLN4924-mediated inhibition of NEDDylation blocks CRL activity resulting in the global stabilization of CRL targets [12] . While the ubiquitin and SUMO pathways have been extensively studied in infectious diseases [16 , 17] , much less attention has been given to the role of NEDD8 . Recent work has shown that the large tegument protein of gammaherpesviruses has deNEDDylase activity and this was important for EBV infection [18] . Another study demonstrated that inhibition of NEDDylation blocked the ability of HIV to degrade the cellular restriction factor APOBEC3G via Vif-mediated hijack of Cul5 [19] . There are indications in the literature that NEDDylation is likely to play a central role in viral pathogenesis . For example , HIV is known to modulate the activity of at least three cullins [20 , 21] , the KSHV LANA protein forms a CRL complex with Cul5 and targets p53 [22] , and it was recently shown that human cytomegalovirus ( HCMV ) utilizes the nucleotide excision repair ( NER ) pathway , which is regulated predominantly by Cul4A/B [23] , to repair its genomes during viral replication [24] . Here we asked if NEDDylation is important for the KSHV lifecycle , and to begin to dissect the functional consequences of its inhibition . Treatment of cancer cells with MLN4924 leads to dramatic cytotoxicity , and some of the best characterized mechanisms include the induction of DNA re-replication by blocking the degradation of Cdt-1 [25 , 26] , or by inhibiting NF-κB signaling via stabilization of the inhibitor of NF-κB protein IκBα [27] , both of which eventually lead to apoptosis . Given the essential role of NF-κB for the maintenance of viral latency and the survival of PEL [28] , we reasoned that these cells would be highly sensitive to MLN4924 . We found that , indeed , MLN4924 led to significant PEL cytotoxicity and this was mediated via inhibition of NF-κB signaling . Intriguingly , we also showed that NEDDylation was essential for amplification of the KSHV genome during reactivation of the lytic cycle and that treatment with MLN4924 prevented the recruitment of RTA to the origin of lytic replication ( OriLyt ) . This work has highlighted the essential role of NEDD8 and CRL-mediated ubiquitylation during the life cycle of KSHV and suggests that NEDDylation may provide a novel therapeutic target for the treatment of KSHV-associated malignancies .
Firstly , it was important to confirm that NEDDylation still occurred in KSHV-infected cells . Transfection of rKSHV-219 cells with FLAG-NEDD8 and the indicated Myc-tagged Cullin , followed by immunoprecipitation of NEDD8 and immunoblot analysis confirmed that Cullin proteins were modified in infected cells ( S1 Fig ) . As MLN4924 has demonstrated anti-proliferative activity in various cancer cell lines , we assessed its cytotoxic effects in PEL cells . We treated PEL cells ( BCBL-1 , TREx-BCBL-1-RTA and BC-3 ) with varying concentrations of MLN4924 and determined cell viability after 96 h using a luminescence ATP detection assay . These data demonstrated that MLN4924 was indeed cytotoxic to latently infected PEL cells with approximate EC50 values of 1 . 10 μM and 0 . 15 μM , for TREx-BCBL-1-RTA and BC-3 respectively ( Fig . 1A ) . However , it was interesting to note that a low level of ORF57 expression was observed in TREx-BCBL-1-RTA ( see 10 μM lane ) and low levels of endogenous RTA in BC-3 cells treated with MLN4924 , suggesting that drug treatment alone induced lytic gene expression ( Fig . 1B & C ) . We also noted that MLN4924 led to apoptosis in BC-3 cells as shown by the cleavage of PARP ( Fig . 1C ) . To investigate MLN4924-induced cytotoxicity , cell cycle analysis of TREx-BCBL-1-RTA cells was performed 24 h after treatment with 1 μM MLN4924 . Compared to untreated TREX-BCBL-1-RTA cells ( Fig . 1D ( i ) ) , MLN4924 treatment led to a reduction in S-phase and an accumulation of cells in G2/M ( Fig . 1D ( ii ) ) . Clearly , inhibition of NEDDylation was toxic to PEL cells , and as shown for BC-3 this led to apoptosis . To further investigate this , we tested the activation of caspases in TREx-BCBL-1-RTA cells . Similar to BC-3 , treatment of TREx-BCBL1-RTA led to the cleavage of PARP , indicative of Caspase 3/7 activation , as did the reactivation of the KSHV lytic cycle after the addition of Dox . ( Fig . 1E ) . It has been shown previously that induction of the lytic cycle activates initiator Caspase 8 , but not Caspase 9 in order to activate effector caspases such as Caspase 3 and 7 [29] . In agreement with this , we also showed that lytic reactivation did not activate Caspase 9; nevertheless , MLN4924 did , resulting in the activation of Caspase 3 ( Fig . 1E ) . Importantly however , caspase inhibition ( using the pan-caspase inhibitor , z-VAD-FMK ) did not prevent PEL cell death after treatment with MLN4924 ( Fig . 1F ) . This result showed that PEL cytotoxicity was not solely a result of apoptosis and suggests other factors , such as the inhibition latency-associated gene expression may contribute to cell death . Therefore , we next measured the effect of treatment on viral gene expression . Latently infected TREx-BCBL-1-RTA were treated with 1 μM MLN4924 and harvested 24 h later . As shown in Fig . 2A , the expression of all lytic-cycle associated genes tested were induced after inhibition of NEDDylation . In contrast , the expression of latency-associated transcripts ORF73 ( latency associated nuclear antigen [LANA] ) and ORF71 ( vFLIP ) were reduced in the presence of drug . Interestingly , the expression of K12 followed that of latency-associated genes . While K12 is dramatically induced upon activation of the KSHV lytic cycle [30] , its expression during latency is initiated from the LANA promoter [31] . This suggested that MLN4924 treatment modulated latency-associated viral gene expression , rather than inducing the full KSHV lytic cycle . To further investigate the potential of MLN4924 to reactivate lytic cycle gene expression , rKSHV . 219 cells were cultured for 36 h in the presence of 0 . 1 μM MLN4924 , a concentration that was tolerated while in culture . These HEK293-based cells are latently infected with recombinant KSHV that contains a red fluorescent protein ( RFP ) reporter gene under the control of an RTA-responsive promoter ( the PAN promoter ) ; hence expression of RTA ( and thus reactivation of the lytic cycle ) leads to RFP expression . As shown in Fig . 2B , RFP expression was observed in numerous cells demonstrating that MLN4924 was indeed able to induce the expression of RTA . MLN4924 was also capable of initiating expression of ORF57 ( activated by RTA during reactivation ) ( Fig . 2C ) and enhancing lytic cycle gene expression in cells reactivated with TPA alone ( Fig . 2D ) . Taken together , these results confirmed that inhibition of NEDDylation was clearly essential for PEL viability and that its inhibition was able to induce low level , lytic cycle-associated viral gene expression . We next investigated the potential mechanisms of MLN4924-mediated cytotoxicity . The reported mechanisms of MLN4924-induced cytotoxicity is the induction of DNA re-replication ( by blocking the CRL-mediated degradation of Cdt-1 ) [26] , or by inhibiting NF-κB signaling as CRL-mediated degradation of IκBα is prevented , thus precluding NF-κB’s transcriptional activity [27] . It is well established that PEL is dependent on NF-κB signaling , and that KSHV vFLIP drives the constitutive activation of NF-κB in order to maintain viability . NF-κB activity also inhibits RTA-mediated transactivation of lytic genes , and thus it is important for maintaining latency . As the viral gene expression profiles following MLN4924 treatment ( Fig . 2A ) are consistent with an inhibition of NF-κB , we tested whether it was inhibited after MLN4924 treatment . The NF-κB transcription factor is tightly regulated via its interaction with IκBα which maintains it in the cytoplasm and away from its transcriptional targets . Upon stimulation , the Iκκ complex phosphorylates IκBα at ser32 and 36 , which signals for the recruitment of the Cul1-containing βTrCP E3 ligase leading to the subsequent degradation of IκBα . This releases NF-κB allowing to translocate to the nucleus and drive transcription [32] . Therefore , MLN4924 inhibition of CRL1 function should stabilize phosphorylated IκBα ( pIκBα ) leading to an inhibition in NF-κB function . To investigate the most likely mechanism of MLN4924-induced cytotoxicity we performed a timecourse experiment ( Fig . 3A ) . Here we show that as little as 1 h treatment of TREX-BCBL-1-RTA cells with MLN4924 led to an inhibition of NEDDylation ( as shown by the lack of NEDDylated Cul2 compared to DMSO-treated cells ) . We also observed the stabilization of pIκBα as early as 1 h post-treatment . By contrast , the accumulation of Cdt1 did not occur until 4 h with significant levels not observed until 6–8 h later . Furthermore , cell cycle analysis of TREx-BCBL-1-RTA cells revealed an accumulation of cells in G2/M after 24 h , indicative of a block at the DNA damage check point . Concomitant with this , we observed the phosphorylation of p53 at 24 h post treatment , but not at earlier time points . Likewise , we did not observe significant alterations in the levels of phosphorylated H2Ax ( γH2Ax ) —a marker of DNA strand breaks—although γH2Ax is known to be present in latently infected cells due to its association with LANA expression [33] . Importantly , the MLN4924-associated reduction of ORF73 mRNA expression occurs prior to any indication of DNA damage ( Fig . 3B ) . Along with cell cycle analysis ( Fig . 1D ) showing that treatment of TREx-BCBL-1-RTA cells led to a reduction of cells in S-Phase ( which is inconsistent with Cdt1-associated DNA-rereplication ) , and the rapid stabilization of pIκBα suggests that inhibition of NF-κB signaling was responsible for cytotoxicity . Additionally , pIκBα levels increased in an MLN4924-dose-dependent manner in latently-infected TREx-BCBL-1-RTA ( Fig . 3C ) and BC-3 cells ( Fig . 3D ) treated for 24 h , showing that , as expected , treatment did stabilize pIκBα . In accordance with this , the cytoplasmic levels of the p65/RelA subunit of NF-κB were increased in a dose-dependent manner ( Fig . 3E-F—enrichment of the cytoplasmic compartment was verified by the lack of nuclear Lamin B—compared to the nuclear fraction control which maintained Lamin B but lacked GAPDH ) . Previous reports have shown that upon lytic reactivation , IκBα is posttranscriptionally downregulated [34] and consistent with this , cytoplasmic p65/RelA decreased in Dox-treated ( reactivated ) cells further highlighting the involvement of IκBα for MLN4924-induced inhibition of NF-κB . Additionally , we did not observe pIκBα expression in reactivated cells ( those with RTA expression ) regardless of MLN4924 treatment ( Fig . 3C-D ) . This was also highlighted in our confocal analysis of p65/RelA localization showing it had translocated to the nucleus in reactivated cells even when treated with MLN4924 ( S2 Fig ) . We therefore surmised that inhibition of NF-κB , via stabilization of IκBα was responsible for cytotoxicity and demonstrated that inhibition of NEDDylation may provide a therapeutic option for the treatment of NF-κB-dependent PEL . Treatment of cells with MLN4924 blocks total NEDDylation , which undoubtedly has far-reaching implications on cellular function . To determine if inhibition of CRL activity was responsible for the phenotypes we have observed , and to identify which CRLs are implicated , we expressed dominant-negative versions of the different cullin family members ( DNCul ) in cells latently infected with KSHV ( rKSHV . 219 ) and screened for reactivation of the lytic cycle by expression of ORF57 . DNCuls are truncated forms of cullin proteins that are still able to engage their respective substrates , but are unable to bind to the Ub-loaded E2 enzyme ( and are unable to be NEDDylated ) , thus preventing substrate degradation [35] . Using this approach , we showed that CRL1 was able to reactivate ORF57 expression , albeit at a low level ( Fig . 4A-B ) . That CRL1 inhibition reactivated lytic expression is not surprising as it is known to regulate NF-κB signaling , corroborating our earlier analyses ( Fig . 2 & 3 ) [36 , 37] . However , we were surprised to note that CRL4B inhibition was able to reactivate ORF57 expression ( Fig . 4A-B ) . Cul4A and 4B both regulate chromatin-associated functions , and they share ca . 80% amino acid identity [38] . However , Cul4B has an extended N-terminus that contains a nuclear localization signal ( NLS ) and its expression is confined to the nucleus , whereas Cul4A is recruited to the nucleus following its interaction with DDB1 in response to genotoxic stress . Due to the high degree of similarity , they are both capable of binding the same CRL4 components ( e . g . DNA damage binding proteins 1 and 2 [DDB1 , DDB2] and the various DDB and Cul4-associated factors [DCAFs] ) . Interestingly , when DNCul4A and 4B were expressed together , the activation of ORF57 was reduced . We believe that this was due to the fact that these two proteins compete for the same binding partners meaning that the degree of nuclear inhibition of CRL4B was reduced , further implicating the importance of CRL4B in the regulation of KSHV latency . Moreover , as CRL4B resided in the nucleus , this particular E3 ligase is the most likely one to be involved in KSHV biology . To confirm this , independent shRNA knockdown of Cul4B in TREx-BCBL-1-RTA cells was performed which led to an ca . 50% reduction in Cul4B expression ( Fig . 4C ) resulting in the expression of various lytic cycle-associated genes ( Fig . 4D ) . Interestingly , expression of the latency-associated gene LANA did not alter suggesting that Cul4B specifically regulated the expression of lytic genes . Here , we also investigated the expression of K12 , which in contrast to the reduced expression we observed in the presence of MLN4924 , was increased by Cul4B knockdown ( Fig . 4D ) . Although a thorough investigation of individual CRLs is required ( for example , using RNAi experiments to inhibit the various CRL components ) , this data suggests that MLN4924’s effect on KSHV latency is through its inhibition of CRL activity . In addition , we have identified CRL4B as a novel regulator of KSHV latency . The KSHV lytic cycle is intimately linked to the pathogenesis of KSHV malignancies and the role of NEDDylation or the importance of CRL-mediated ubiquitylation during this process is currently unknown . To investigate if ubiquitylation is indeed a feature of KSHV reactivation , we stained reactivated cells with anti-Ub FK2 ( which only recognizes ubiquitylated proteins and not free ubiquitin ) and to increase the stringency of this assay , we removed soluble nuclear material prior to fixation ( see Materials and Methods ) . As shown in Fig . 5A , reactivation is associated with significant levels of ubiquitylation which appeared to localize to the edges of the replication compartments ( Fig . 5A inset ) . However , in the presence of low concentration MLN4924 ( 1 μM ) , the degree of replication compartment ubiquitylation was reduced . There are two potential reasons that explain this observation; that CRLs are the predominant E3 ligases that ubiquitylate replication compartment factors , or that NEDDylation is required within replication compartments , and this is important for recruiting ubiquitin E3 ligases ( as recently demonstrated during the DNA double-strand break repair mechanism [39] ) . As NEDDylation-dependent ubiquitylation occurred around replication compartments , we asked whether this modification was required for KSHV lytic reactivation . For these experiments , we employed TREx-BCBL-1-RTA cells , where the lytic cycle is induced by the addition of doxycycline ( Dox . ) by virtue of Dox-inducible expression of exogenous RTA-Myc . The advantage of using TREx-BCBL-1-RTA cells is that the lytic cycle is robustly induced ( more so than with TPA/NaB ) and any effect is a result of viral gene expression and not due to the general effects of TPA/NaB . Cells were treated with varying concentrations of MLN4924 , the lytic cycle was induced and 24 h later cells were harvested and viral protein expression was assessed by immunoblotting . As shown in Fig . 5B , MLN4924 was able to inhibit the expression of RTA-Myc and its downstream target ORF57 in a dose-responsive manner , corresponding to the levels of Cul2 NEDDylation . As RTA-Myc represents exogenous protein expression ( i . e . not from the KSHV genome ) in TREx-BCBL-1-RTA cells , these results showed that MLN4924 inhibited all transcription in these cells . Using qPCR analysis of viral genomes , we also confirmed that MLN4924 was able to inhibit KSHV DNA replication ( Fig . 5C ) . However , KSHV genome replication was inhibited at lower concentrations than was required to inhibit viral protein expression ( compare protein expression and genome replication levels in cells treated with 1 μM ) highlighting a potentially novel mechanism , mediated by NEDDylation , that uncouples these two features of KSHV biology . These data show for the first time that NEDDylation ( or NEDDylation-dependent ubiquitylation ) plays a significant role during the KSHV lytic cycle . We have shown that the inhibition of NEDDylation leads to caspase activation ( Fig . 3B ) which may account for the ability of MLN4924 to prevent KSHV genome replication . Furthermore , caspase activation has been shown to regulate viral gene expression via Caspase 7 cleavage of ORF57 [29] . We therefore asked whether caspases were responsible for the inhibition of KSHV reactivation . Treatment of cells with z-VAD-FMK efficiently blocked caspase-mediated cleavage of PARP-1 ( Cl . PARP-1 ) and ORF57 ( Cl . ORF57 ) in reactivated cells; caspase inhibition also led to an increase in viral protein expression as noted by the increased levels of full length ORF57 ( FL ORF57 ) and RTA expression in reactivated cells ( lane 4 , Fig . 6A ) . Importantly , inhibition of caspase activity restored ORF57 and RTA-Myc expression , clearly demonstrating that MLN4924-induced apoptosis was responsible for inhibiting viral protein expression ( lane 6 , Fig . 6A ) . To our surprise however , when we treated cells with z-VAD-FMK and MLN4924 and investigated viral genome copy number by qPCR , viral DNA replication was still inhibited ( Fig . 6B ) . An additional measure of viral genome replication involves the incorporation of EdU into DNA specifically marking replicating viral genomes in replication compartments , where proteins such as RTA are also localized ( Fig . 7A ) . However , even in cells clearly expressing RTA , MLN4924 treatment prevented virus replication , as shown by the lack of EdU-positive replication compartments . Inhibition of caspases restored viral protein expression in MLN4924-treated cells; however , viral DNA replication was still inhibited ( Fig . 6B ) . We therefore repeated our EdU analysis of drug-treated cells in the presence of z-VAD-FMK . Inhibition of caspases permitted the formation of replication centers and amplification of KSHV DNA in non-treated cells . Nevertheless , z-VAD-FMK was still unable to restore viral genome replication in MLN4924-treated cells ( Fig . 7B ) . These data confirm that MLN4924-mediated activation of caspases was not responsible for the inhibition of KSHV genome replication upon MLN4924 treatment . Throughout our studies , we noticed that reactivated cells treated with MLN4924 displayed an unusual RTA expression pattern; as shown in Figs . 5A & 7A , MLN4924 treatment is associated with pan-RTA localization , rather than in discrete , EdU-positive foci . We also observed the same phenotype when we costained for RTA and a second replication compartment-associated factor , RNA Pol II ( S3 Fig ) . We therefore hypothesized that this would have implications for RTA’s ability to mediate KSHV genome replication , and therefore provide a rationale for the MLN4924-induced block in replication . OriLyt is the cis-acting loci where proteins required for KSHV genome replication assemble . The mechanisms of this have been well characterized and RTA is central to this process . For example , it has been shown that the pre-replication complex ( which involves RTA , K8 , the core replication proteins and various cellular proteins ) is formed prior to its loading onto OriLyt , and that RTA , through its direct DNA binding activity with RRE , recruits these factors to OriLyt [40–42] . Therefore , as MLN4924 inhibited viral genome replication and prevented the correct localization of RTA , we asked whether treatment precluded the loading of the replication complex . To do this , TREx-BCBL-1-RTA cells were treated with 1 μM MLN4924 ( a concentration that permits viral protein expression but inhibits replication ) and 18 h later , ChIP analysis was performed using antibodies specific for IgG ( isotype control ) , RNA Pol II and Myc-tag ( for RTA ) , and primers specific for the RTA responsive element ( RRE ) of OriLyt . Firstly , we performed this analysis on latently infected cells ( Fig . 7C ) , where we observed an enrichment of RNA Pol II ( 6 . 45-fold over IgG ) , whereas RTA levels were lower than those observed for IgG control . Interestingly , when cells were treated , RNA Pol II occupancy at OriLyt approximately doubled , and RTA levels increased 1 . 6-fold over IgG . These results confirm that MLN4924 was able to induce transcription in latently infected cells; however , they show that only low levels of RTA are required . During lytic replication , both RNA Pol II ( 63-fold over IgG ) and RTA ( 44-fold over IgG ) were increased at OriLyt ( Fig . 7D ) . However , after MLN4924 treatment , RNA Pol II occupancy reduced ca . 3-fold ( to 21-fold over IgG ) and RTA occupancy decreased ca . 4-fold ( to 10-fold over IgG ) confirming that the inhibition in KSHV genome replication was due to a block in the recruitment of the viral pre-replication complex . Consequently , by using MLN4924 to investigate the role of NEDDylation in KSHV biology , we have uncovered a novel mechanism that regulates lytic reactivation .
KSHV infection is responsible for various malignancies , including KS , PEL and many cases of MCD . As these diseases are highly associated with compromised immune function , they represent some of the most common cancers in areas of the world where HIV infection is also high . Indeed , KS is the most prevalent cancer in many sub-Saharan countries . Therefore , understanding the molecular mechanisms that underlie KSHV biology is of the utmost importance if therapeutic targets are to be identified . Given the success of drugs such as bortezamib ( UPS inhibitor ) for the treatment of various malignancies , interest has grown in the development of inhibitors that target additional aspects of the UPS , or additional Ubls . An example of this is MLN4924 , a small molecule inhibitor that blocks the function of NAE1 , the first enzyme ( E1 ) in the NEDDylation cascade . Inhibition of NEDDylation leads to a global stabilization of CRL targets , and this drug has proved to be potently cytotoxic in many cancer models . The principal mechanisms of MLN4924 cytotoxicity in cancer cells appear to involve blocking NF-κB signaling ( via stabilization of IκBα , and the retention of the NF-κB transcription factor in the cytoplasm ) or leading to unlicensed DNA replication ( via stabilization Cdt-1 ) , ultimately leading to apoptotic cells death . Given that PEL cells absolutely require NF-κB signaling for their survival , we asked whether MLN4924 was able to kill these cells . Furthermore , as the lytic cycle plays a significant role during the pathogenesis of KS , we also asked whether the NEDDylation cascade was necessary for virus reactivation , and if so , what role it plays . The activation of NF-κB is central to KSHV infection ( by modulating viral gene expression ) and for the pathogenesis of KSHV-associated malignancies ( via induction of inflammatory mediators and the expression of antiapoptotic genes ) . Inhibiting NEDDylation was clearly cytotoxic to PEL cells , and mechanistically , it appears that this was due to the inhibition of NF-κB . For example , we showed that MLN4924 led to: i ) accumulation of pIκBα , ii ) a KSHV gene expression profile consistent with inhibition of NF-κB and iii ) the induction of apoptosis ( Fig . 8 ) . This corroborates recent work demonstrating that MLN4924 killed NF-κB-dependent ABC-DLBCL cells in a comparable fashion [27] . In fact , ABC-DLBCL cells were more sensitive than GCB-DLBCLs ( that are not reliant on NF-κB ) [27] suggesting that this drug may be more efficacious for cancers such as PEL and KS , where NF-κB activation is a requirement for malignancy . NF-κB is also required for transcription of the latency control locus which expresses an alternatively spliced transcript that produces ORF71 ( vFLIP ) , ORF72 ( vCyclin ) and ORF73 ( LANA ) . As LANA is essential for maintenance of the KSHV genome during mitosis , NF-κB signaling is clearly required for sustained infection . Interestingly , MLN4924 did not appear to overtly affect LANA expression ( despite reducing its mRNA level ) which is in line with previous reports showing that its half-life exceeds the time course of our experiments [43] and signifying that cytotoxicity was not a result of KSHV genome loss . Moreover , the histological signature of KS involves elongated , spindle-shaped endothelial cells and vFLIP expression alone is sufficient to bring about this morphological change in endothelial cell cultures [44 , 45] . Importantly , this phenotypic alteration has been shown to be NF-κB-dependent; therefore , it will be of particular interest to investigate if MLN4924 can inhibit this characteristic feature of KS . In contrast to its role in latency-associated transcription , NF-κB is a negative regulator of numerous lytic cycle-associated genes , including those analyzed in this study ( RTA , ORF57 and ORF47 ) [46] . The KSHV protein RTA is both necessary and sufficient for the induction of the lytic cycle , resulting in the expression of the entire KSHV genome and virion production . Many RTA-responsive genes contain RBP-Jκ binding sites , and the translocation of NF-κB into the nucleus allows it to bind RBP-Jκ and prevent its association with RTA . Hence , the induction of lytic cycle-associated gene expression in PEL and rKSHV . 219 cells is very likely due to MLN4924-induced inhibition of NF-κB . The Cul1-containing βTrCP is the principal E3 ligase responsible for IκBα degradation [37] . In agreement the above hypothesis , we found that individual expression of dominant-negative versions of Cul1 ( DNCul ) was able to activate lytic cycle gene expression . Importantly , these data also suggest that MLN4924’s effect on viral gene expression was due to its inhibition of the CRL activity as opposed to a direct role of NEDDylation in KSHV gene expression or the over-arching effects a global block in NEDDylation would have on cellular function . However , an in-depth investigation of the various CRL components is required to fully elucidate the question of specificity . In addition to the ability of DNCul1 to reactivate lytic cycle associated gene expression , DNCul4B and shRNA knockdown of Cul4B were both able to induce lytic cycle gene expression . Using K12 expression as an indicator of latency versus lytic cycle-associated gene expression , the complete blockade of NEDDylation led to a reduction in its expression , whereas knockdown of Cul4B led to its increase , pointing to differing mechanisms . These data are in line with the hypothesis that complete blockade ( via MLN4924 ) effects NF-κB expression by inhibiting CRL1 and that inhibition of Cul1 NEDDylation is more sensitive than its inhibition of Cul4B modification . It is not currently known whether differences exist in the sensitivity of individual CRLs following MLN4924 treatment . Of the known CRL functions , Cul4 activities are centered on chromatin regulation , such as chromosome condensation , heterochromatin formation and DNA replication and repair processes . Therefore , although Cul4’s potential role in the maintenance of KSHV latency is novel , it might not be surprising . The two Cul4 proteins are virtually identical apart from an extended N-terminal portion found in Cul4B that encodes a nuclear localization signal . Therefore , Cul4B predominantly resides in the nucleus , whereas Cul4A is recruited to the nucleus in response to genotoxic stress . This provides further credence to the hypothesis that Cul4B is important for the regulation of KSHV gene expression . The DNA damage response is a recurring theme in virus biology as it either aids or is a consequence of infection [24 , 47–49] . Cul4A and 4B play a pivotal role in the nucleotide excision repair ( NER ) process , and it was recently reported that human cytomegalovirus ( HCMV ) is dependent on NER during replication of its genome . Therefore , the role of Cul4 during KSHV infection merits further investigation . A striking observation however , was that MLN4924 treatment blocked viral genome replication despite appreciable levels of viral gene expression . At 1 μM MLN4924 , ORF57 expression was not significantly reduced , but genome replication was reduced by ca . 80% . We considered the possibility that the induction of apoptosis may be responsible for this , and so we repeated the experiments in the presence of a pan-caspase inhibitor ( z-VAD-FMK ) . Even though this led to a recovery in KSHV protein expression even when up to 10 μM MLN4924 was used , genome replication was still not observed . Over the course of our studies , we did notice aberrant RTA localization in BCBL-1 cells treated with MLN4924 , followed by reactivation ( e . g . Fig . 7 ) . Given that RTA’s function is to recruit the viral pre-replication complex , along with various cellular factors required for KSHV genome replication , its localization is intimately linked with sites of viral genome replication occurs . These sites are termed replication compartments ( or replication centers ) and can be observed as discrete foci at the nuclear periphery that co-stain with RTA and newly replicated viral DNA ( this can be observed using BrdU or EdU that specifically marks viral DNA due to KSHV’s ability to block cellular DNA replication during reactivation ) . We therefore hypothesized that NEDDylation , or CRL-mediated ubiquitylation was required for KSHV genome replication ( but not viral gene expression ) . Firstly , we showed that reactivation of KSHV is associated with ubiquitylation within replication compartments , and that MLN4924 treatment inhibited this . We next investigated whether NEDDylation was required for RTA’s recruitment to the origins of lytic replication ( OriLyt ) . Using ChIP analysis , we observed that treatment of latent cells led to an increase in Pol II ( ca . 2-fold ) and RTA ( ca . 1 . 6-fold ) occupancy , which agrees with our data showing that treatment led to activation of viral gene expression . Treatment followed by reactivation with Dox . however showed us that NEDDylation was required for RTA recruitment to OriLyt . Here we showed that 1 μM MLN4924 ( a concentration that still permits viral gene expression ) blocked RTA recruitment to OriLyt . Likewise , Pol II occupancy was reduced to similar levels . This confirmed to us that NEDDylation was required for the proper recruitment of the RTA ( and therefore the pre-replication complex ) to OriLyt , and offers and explanation for the aberrant localization of RTA and the inhibition of genome replication . There are various hypotheses that may explain the importance of NEDDylation for the recruitment of the pre-replication complex to OriLyt ( Fig . 8 ) . It might be possible that proteins within the pre-replication complex are themselves targets for NEDD8 modification . We have addressed this for RTA and showed that RTA is not NEDDylated ( S4 Fig ) ; given that RTA is the protein responsible for recruitment , this result suggested that this hypothesis is unlikely [40] . It might also be possible that NEDDylation or CRL-mediated ubiquitylation of OriLyt-associated chromatin might be important for the recruitment of the pre-replication complex , as has recently been reported for the recruitment of repair enzymes during a DNA damage response [50] . Although it is still possible that these modifications play a role in cells , RTA binds to DNA in a sequence-specific manner , and it can do this in vitro in the absence of chromatin suggesting that this may not be the role of NEDDylation during reactivation [51] . A further hypothesis is that modification of a factor that resides at OriLyt during latency ( in order to maintain latency ) may be required during reactivation . This may include CRL-mediated ubiquitylation ( via K48 ) that targets a latency-associated protein for degradation thus allowing access to OriLyt by the pre-replication complex . It may also involve Ub-modification via K63 linkage that might enhance binding in vivo . Various cellular proteins have been reported to interact with OriLyt during reactivation , but in that study , proteins that bound during latency were not investigated [42] . All of the above hypotheses merit further investigation but are unfortunately beyond the scope of this study . In summary , we have demonstrated that NEDDylation is essential for various aspects of KSHV infection . Inhibition of NEDDylation using MLN4924 proved cytotoxic to PEL cells due to their dependence on NF-κB . Remarkably , we also show that a functioning NEDDylation cascade is essential for KSHV genome replication as it was required for the recruitment of the RTA-mediated pre-replication complex to OriLyt . These new findings have opened up new avenues of investigation regarding the regulation of herpesvirus latency and reactivation . Moreover , they demonstrate that inhibition of NEDDylation represents a novel approach for the treatment of KSHV-associated malignancies , including KS that is dependent on both lytic replication and the latency-associated activation of NF-κB .
BCBL-1 and BC-3 are primary effusion lymphoma ( PEL ) B cell lines latently infected with KSHV . TREx-BCBL-1-RTA cells ( a kind gift of Dr . Jae Jung , University of Southern California ) are a BCBL-1-based cell line that has been engineered to inducibly express exogenous Myc-tagged RTA by the addition of doxycycline , leading to a robust reactivation of the full KSHV lytic cycle [52] . The rKSHV . 219 cell line maintains KSHV as a latent infection and was generated by infecting HEK293T cells with a recombinant KSHV that contains a constitutively active puromycin resistance and GFP gene , and an RFP gene that is fused to an RTA-responsive lytic cycle ( PAN ) promoter; hence , expression of RFP can be used as a reporter of RTA activity [53] . HEK293T cells were also used for co-immunoprecipitation experiments . All cells were maintained at 37°C in a humidified incubator with 5% CO2 . The PEL cell lines were maintained in RPMI-1640 ( Lonza ) supplemented with 10% FBS ( Life Technologies ) . TREx-BCBL-1-RTA media also contained 200 μg/ml hygromycin B ( Life Technologies ) . The rKSHV . 219 and HEK293T cell lines were maintained in DMEM ( Lonza ) supplemented with 10% FBS ( Life Technologies ) . The rKSHV . 219 cultures also contained 1 μg/ml puromycin . KSHV reactivation was induced in BCBL-1 and rKSHV . 219 by the addition of 20 ng/ml 12-O-tetradecanoylphorbol 13-acetate ( TPA ) and 1 . 5 mM sodium butyrate ( NaB ) or 1 μg/ml doxycycline hyclate ( Sigma ) in TREx-BCBL-1-RTA cells . MLN4924 [12] ( Millennium Pharmaceuticals ) stock solutions ( 10 mM ) were prepared in DMSO and diluted in media prior to its addition to cells at the indicated concentrations , and for the indicated times . Routinely , 106 cells in 12-well plates were treated . Inhibition of caspase enzymes was achieved by the addition of 50 μM z-VAD-FMK or FMK-negative control ( MBL International ) 30 min prior to MLN4924 and/or reactivation treatment . Inhibitors remained on the cells for the duration of the experiments . Cell viability was determined using the standard trypan blue exclusion method and using the ATPlite Luminescence ATP Detection Assay ( Perkin Elmer ) . For the ATPlite assay , 10 , 000 cells were seeded into white 96-well plates and allowed to settle for 16 h at 37°C . Varying concentrations of MLN4924 were added to the cells followed by 96 h incubation at 37°C . Cellular proliferation was determined according to the manufacturer’s instructions . Cells ( 106 ) were treated with 1 μM MLN4924 ( or left untreated ) for 24 h , washed in cold PBS and fixed for 24 h in cold 70% ethanol at -20°C . Prior to analysis , cells were washed in PBS and treated with 1 ml PBS , 10 μg/ml propidium iodide ( Sigma ) , 0 . 5 mg RNase A for 3 h at 37°C . Cells were then pelleted , resuspended in 1 ml PBS and analysed using a Becton Dickinson BD-LSRFortessa flow cytometer . Data were fitted using ModFit software . Most expression vectors were obtained from Addgene: pcDNA3-DN-hCul1-FLAG ( plasmid 15818 ) , pcDNA3-DN-hCul2-FLAG ( plasmid 15819 ) , pcDNA3-DN-hCul3-FLAG ( plasmid 15820 ) , pcDNA3-DN-hCul4A-FLAG ( plasmid 15821 ) , pcDNA3-DN-hCul4B-FLAG ( plasmid 15822 ) , pcDNA3-DN-hCul5-FLAG ( plasmid 15823 ) [35] , pcDNA3-HA-Cullin4A ( plasmid 19907 ) , pcDNA3-Myc-NEDD8 ( plasmid 19943 ) . pFLAG-CMV-4-NEDD8 was a kind gift from Dr Eric Stebbins ( Rockefeller University ) [54] . RTA expression vectors ( pRTA and pRTAH145L ) were gifts of Dr Gary Hayward ( Johns Hopkins University ) . Cells were plated into 6-well plates and transfections routinely used 1 μg plasmid DNA and Lipofectamine 2000 ( Life Technologies ) following the manufacturer’s instructions . Knockdown of Cul4B expression was accomplished using four individual SureSilencing shRNA expression vectors ( Qiagen ) . TREx-BCBL-1-RTA cells ( 3 x 106 per transfection ) were transfected with 20 μg scramble shRNA ( control ) or 5 μg of each of four shCul4B vectors by nucleofection ( Lonza; solution V and nucleofector program T-01 ) . Four days post-transfection , RNA was extracted using Trizol ( Life Technologies ) and used to generate cDNA for qRT-PCR analysis ( see below ) . Cells were washed in PBS and proteins extracted in lysis buffer containing 50 mM Tris ( pH 7 . 4 ) , 150 mM NaCl , 1% NP-40 and 1x protease inhibitor cocktail ( Roche ) for 15 min on ice and clarified by centrifugation at 12 , 000 xg for 10 min , 4°C . Cells used in Nuclear/cytoplasmic enrichments assays ( ca . 106 ) were washed in PBS , lysed in 50 μl PBS , 1% Triton X-100 and 1x protease inhibitor cocktail ( Roche ) for 15 min on ice and nuclei were pelleted at 2000 xg for 5 min . Cytoplasmic fractions ( supernatant ) were moved to new tubes and the nuclear pellets were washed three times in lysis buffer . SDS-PAGE and immunoblotting of normalized protein concentrations followed standard techniques using the following antibodies [55]: RTA , rabbit antisera ( 1:400 ) , mouse mAb anti-Myc-tag ( 1:5000; Sigma ) , rabbit pAb anti-FLAG-tag ( 1:1000; Sigma ) , mouse mAb anti-HA-tag ( 1:5000; Life Technologies ) , mouse mAb anti-ORF57 ( 1:1000; Santa Cruz ) , rabbit mAb anti-Cullin 2 ( 1:1000; Life Technology ) , mouse mAb anti-PARP-1 ( 1:1000; Cell Signaling Technology ) , mouse mAb anti-GAPDH ( 1:5000; Sigma ) , mouse mAb anti-Lamin B ( 1:1000; Santa Cruz ) , rabbit mAb anti-Phospho-IκBα ( ser32 ) ( 1:1000; Cell Signaling Technology ) , mouse mAb anti-p65/RelA ( 1:1000; Santa Cruz ) , rabbit mAb anti-Caspase 3 ( 8G10 ) ( 1:1000; Cell Signaling Technology ) , rabbit mAb anti-active Caspase 3 ( Abcam; 1:250 ) , mouse mAb anti-Caspase 9 ( Cell Signaling Technology ) , mouse mAb anti-Cdt-1 ( 1:1000; Santa Cruz ) , mouse mAb anti-γH2Ax ( ser139 ) ( 1:1000 , Santa Cruz ) rabbit mAb anti-Phospho-p53 ( ser15 ) ( 1:1000; Cell Signaling Technology ) . Signals were detected using chemiluminescence and densitometry ( ImageJ ) was used to semi-quantify expression levels . As previously reported [56] , total cellular RNA was extracted from cells using Trizol ( Life Technologies ) according to the manufacturer’s instructions and contaminating DNA was removed using the DNA-free kit ( Ambion ) . Complimentary DNA ( cDNA ) was generated from 1 μg RNA in 20 μl reaction volumes using M-MuLV reverse transcriptase ( RT; NEB ) according to the manufacturer’s recommendations with 5 ng oligo ( dT ) . In parallel , negative control reactions were performed for each RNA by omitting RT in order to confirm that quantification represented cDNA and not contaminating DNA . Quantitative PCR reaction mixes ( 20 μl ) included 1x SensiMix SYBR green master mix ( Bioline ) , 0 . 5 μM each primer and 1 μl cDNA reaction mix . Cycling was performed in a RotorGene Q machine ( Eppendorf ) and included an initial 10 min denaturation step at 94°C , followed by 40 cycles of 30 s at 94°C , 30 s at 60°C and 30 s at 72°C . Melting curve analysis was performed between 65 and 95°C ( with 0 . 2°C increments ) to verify amplicon specificity . Quantification of GAPDH mRNA was used to normalize between samples , and the average cycle threshold ( CT ) was determined from three independent RNA samples from independent cultures . Relative expression compared to non-treated or DMSO-treated control cells was calculated using the ΔΔCT method . Cells were cultured on poly-L-lysine coated coverslips in 12-well plates for 24 h at 37°C , gently washed with PBS , fixed using 4% formaldehyde ( in PBS ) for 10 min , permeabilized with PBS , 1% Triton X-100 for 10 min and washed three times with PBS as previously reported [57 , 58] . Where stated , some experiments involved the removal of soluble nuclear proteins ( i . e . those not tightly bound to chromatin ) using an “extraction first” method: here coverslips were treated ( as previously reported [24] ) with CSK buffer ( 10 mM PIPES , 100 mM NaCl , 300 mM sucrose , 3 mM MgCl2 ) containing 0 . 5% Triton X-100 for 2 min followed by 2 washes with CSK buffer alone . These cells were then fixed and processed as normal . Primary antibodies were diluted in PBS , 2% BSA , added to cells and incubated in humidity chambers for 2 h at 37°C or overnight at 4°C followed by 5 washed with PBS . The appropriate secondary antibodies ( Alexa Fluor 488 or 594; Life Technologies ) were diluted 1:500 in PBS , 2% BSA and incubated with cells for 1 h at 37°C followed by 5 washed with PBS . Coverslips were mounted in VECTORSHEILD with DAPI ( Vectorlabs ) . Where experimentally possible , double staining was performed by incubating the coverslips with both primary antibodies , and following washing , both secondary antibodies at the same time . Images were captured using an LSM700 laser scanning microscope ( Carl Zeiss ) and processed using ZEN imaging software ( Carl Zeiss ) . Antibodies included: RTA rabbit antisera ( 1:100 ) , NF-κB RelA mouse mAb ( 1:250; Cell Signaling ) , anti-ubiquitinylated proteins clone FK2 mouse mAb ( 1:250; Millipore ) , mouse mAb anti RNA Pol II ( Millipore ) . Virus reactivation was determined by two complimentary assays—identification of viral replication compartments by imaging the incorporation of EdU into replicating viral DNA and by quantitative PCR analysis of viral genome amplification . Incorporation of EdU was performed using a Click-iT EdU Imaging Kit ( Life Technologies ) . Briefly , 1 x 106 TREx-BCBL-1-RTA cells were drug-treated as appropriate , virus reactivation was induced by the addition of doxycycline and the cells were plated onto poly-L-lysine-coated coverslips . After 16 h , cells were pulsed for 45 min with 10 μM EdU , washed , fixed and permeablized according to the manufacturer’s recommendations . After detection of EdU ( according to the manufacturer’s protocol ) , the cells were washed and further incubated with RTA antisera ( 1:100 in PBS , 2% BSA ) for 1 h at 37°C , washed again and incubated with Alexa Fluor 488 goat anti-rabbit antibody ( Life Technologies; 1:500 in PBS , 2% BSA ) for 1 h at 37°C . Finally , DNA was stained using Hoechst 33342 ( 1:2000 in PBS ) and mounted in VECTORSHIELD ( Vectorlabs ) . Images were captured using an LSM700 laser scanning microscope ( Carl Zeiss ) using 639 nm , 488 nm and 405 nm lasers for the detection of EdU , RTA and DNA respectively and processed using ZEN imaging software ( Carl Zeiss ) . For quantitative PCR analysis 1 x 106 TREx-BCBL-1-RTA cells were drug-treated as appropriate and virus reactivation was induced by the addition of doxycycline . At 24 h post-reactivation , total DNA was extracted using a DNA Minikit ( Qiagen ) and quantified by UV spectrophotometry . Viral DNA was quantified using 3 . 4 ng DNA , in 20 μl reaction volumes as described above for qRT-PCR , using primers specific for the ORF57 gene . Quantification of GAPDH was used to normalize between samples , and the average cycle threshold ( CT ) was determined from three independent samples from independent cultures . Relative levels of viral DNA was calculated using the ΔΔCT method . HEK293T cells were plated into 6-well dishes and co-transfected with the indicated plasmids with pcDNA-Myc-NEDD8 or pFLAG-CMV-4-NEDD8 [54] for 24 h . Cells were washed in PBS and proteins extracted in 1 ml lysis buffer ( see above ) and incubated with anti-c-Myc agarose ( Sigma ) or goat anti-DDDDK conjugated agarose ( for detection of FLAG-tagged proteins; Abcam ) following the manufacturer’s recommendations . Immunoprecipitated ( NEDDylated ) proteins were eluted in Laemmli buffer and subject to immunoblot analysis ( see above ) . Chromatin immunoprecipitation ( ChIP ) assays were carried out as previously stated [59] using the EZ-ChIP chromatin immunoprecipitation kit ( Millipore ) . Per condition , 107 cells were treated with 1% formaldehyde for 10 min prior to three washes with ice-cold PBS . Cells were then resuspended in 3 ml SDS lysis buffer ( 1% SDS , 10 mM EDTA , 50 mM Tris , [pH 8 . 1] ) containing 1x protease inhibitor cocktail II and sonicated 15 time with 20 s pulses at 4°C . Insoluble material was removed by centrifugation for 10 min at 12 , 000 xg , 4°C . For each immunoprecipitation ( IP ) , 100 μl of prepared chromatin was added to 900 μl of dilution buffer ( 0 . 01% SDS , 1 . 1% Triton X-100 , 1 . 2 mM EDTA , 16 . 7 mM Tris-HCl , [pH 8 . 1] , 167 mM NaCl ) containing 1x protease inhibitor cocktail II . The lysates were then pre-cleared with Protein G agarose before being incubated with appropriate antibodies ( Myc-tag and RNA Pol II ) overnight at 4°C . The following day 60 μl of Protein G agarose was added to each IP and rotated for 1 h at 4°C . The Protein G agarose-antibody/chromatin complexes were then washed sequentially in 1 ml of the provided buffers; low salt immune complex wash buffer ( 0 . 1% SDS , 1% Triton X-100 , 2 mM EDTA , 20 mM Tris-HCl , [pH 8 . 1] , 150 mM NaCl ) , high salt immune complex wash buffer ( 0 . 1% SDS , 1% Triton X-100 , 2 mM EDTA , 20 mM Tris-HCl , [pH 8 . 1] , 500 mM NaCl ) , LiCl immune complex wash buffer ( 0 . 25 M LiCl , 1% IGEPAL CA630 , 1% deoxycholic acid sodium salt , 1 mM EDTA , 10 mM Tris , [pH 8 . 1] ) and TE buffer ( 10 mM Tris-HCl , [pH 8 . 0] , 1 mM EDTA ) . The DNA/protein complexes were eluted in 200 μl elution buffer ( 1% SDS , 0 . 1 M NaHCO3 ) before reversal of crosslinks with 5 M NaCl for 4 h at 65°C . Samples were then treated with RNase A at 37°C for 30 min followed by incubation with Proteinase K for 2 h at 45°C . The resulting DNA was purified over the provided spin filter columns , with elution in 50 μl of elution buffer C . The purified DNA was then subject to qPCR amplification using primers directed to the OriLyt ( Forward primer: 5’- ACG GGC CTG GAA TCT CGC CTC TGG-3’ and Reverse primer: 5’- ATG GGC GTA ACC GTA GGA CAA GCT G-3’ ) . Each qPCR reaction performed in triplicate containing 5 μl purified DNA and was quantified as described above . Oligonucleotide primer sequences are available upon request . | Kaposi’s sarcoma-associated herpesvirus ( KSHV ) causes Kaposi’s sarcoma ( KS ) and primary effusion lymphoma ( PEL ) , often fatal malignancies afflicting HIV-infected patients . Previous research has shown that blockade of the ubiquitin proteasome system ( UPS , a normal quality control pathway that degrades cellular proteins ) is able to kill KSHV-infected lymphoma cells . A large component of the UPS is made up by the protein family known as the cullin-RING ubiquitin ligases ( CRLs ) , which are activated by NEDD8 ( a process known as NEDDylation ) . Recently , an inhibitor of NEDDylation ( MLN4924 ) was developed and is currently in clinical trials as an anti-cancer drug . As NEDDylation has not been investigated for many viruses , we used this to compound examine its importance in KSHV biology . Firstly we show that NEDDylation is essential for the viability of KSHV-infected lymphoma cells , and MLN4924 treatment killed these cells by blocking NF-κB activity ( required for KSHV latency gene expression and KSHV-associated cancer ) . Furthermore , we show that NEDDylation is required for KSHV to replicate its genome , a critical step in the production of new virus particles . Therefore , this research has identified a novel molecular mechanism that governs KSHV replication . Furthermore , it demonstrates that NEDDylation is a viable target for the treatment of KSHV-associated malignancies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | NEDDylation Is Essential for Kaposi’s Sarcoma-Associated Herpesvirus Latency and Lytic Reactivation and Represents a Novel Anti-KSHV Target |
Low oxygen conditions ( hypoxia ) can impair essential physiological processes and cause cellular damage and death . We have shown that specific hypoxic conditions disrupt protein homeostasis in C . elegans , leading to protein aggregation and proteotoxicity . Here , we show that nutritional cues regulate this effect of hypoxia on proteostasis . Animals fasted prior to hypoxic exposure develop dramatically fewer polyglutamine protein aggregates compared to their fed counterparts , indicating that the effect of hypoxia is abrogated . Fasting also reduced the hypoxia-induced exaggeration of proteostasis defects in animals that express Aβ1–42 and in animals with a temperature-sensitive mutation in dyn-1 , suggesting that this effect was not specific to polyglutamine proteins . Our data also demonstrate that the nutritional environment experienced at the onset of hypoxia dictates at least some aspects of the physiological response to hypoxia . We further demonstrate that the insulin/IGF-like signaling pathway plays a role in mediating the protective effects of fasting in hypoxia . Animals with mutations in daf-2 , the C . elegans insulin-like receptor , display wild-type levels of hypoxia-induced protein aggregation upon exposure to hypoxia when fed , but are not protected by fasting . DAF-2 acts independently of the FOXO transcription factor , DAF-16 , to mediate the protective effects of fasting . These results suggest a non-canonical role for the insulin/IGF-like signaling pathway in coordinating the effects of hypoxia and nutritional state on proteostasis .
In order to survive in changing conditions , organisms need to successfully integrate a number of environmental signals and respond appropriately in order to maintain homeostasis . Aerobic heterotrophs must meet their requirements for food and oxygen by taking in these resources from the environment . An inadequate response to low levels of oxygen ( hypoxia ) can lead to cellular damage or death , an unsurprising outcome given oxygen’s central role in cellular metabolism . Like hypoxia , food deprivation presents an obstacle to homeostasis by impinging on cellular metabolism and disturbing anabolic pathways . However , in many cases food restriction can have beneficial effects , such as extending lifespan and delaying the onset of neurodegenerative diseases and their associated pathologies [1] . In a mouse model of Alzheimer’s disease , 12 weeks of caloric restriction reduces Aβ plaque burden [2] , and mice expressing human mutant huntingtin maintained on an alternate-day-feeding diet have reduced brain atrophy and decreased huntingtin aggregate formation [3] . Similarly , depriving C . elegans of their bacterial food source reduces damage associated with expressing polyglutamine proteins [4] . The protective effect of fasting is not limited to symptoms of neurodegeneration–there are many studies that show fasting can protect against damage associated with hypoxia in mammals . For example , mice on an alternate-day feeding regimen have higher survival rates after myocardial ischemia induced via coronary occlusion [5] . Similar results have been obtained with ischemic damage to the liver . Mice on a calorically restricted diet have reduced infarct damage compared to ad-libitum fed controls [6] , and mice that have been fasted for 3 days display reduced hepatocellular apoptosis and damage [7] . Reduced food intake also improves outcomes after cerebral ischemic injury by protecting cortical and striatal neurons [8] and reducing neurological deficits and infarct volume [9] . These observations suggest that understanding the mechanistic basis underlying the protective effects of fasting in hypoxia could provide novel insight into therapeutic strategies to treat pathological conditions associated with ischemia and reperfusion injury . We have previously shown that in C . elegans the cellular response to specific hypoxic conditions involves a disruption of proteostasis–the coordination of protein synthesis , folding , degradation , and quality control required to maintain a functional proteome [10] . Here we show that fasting prevents the hypoxia-induced disruption of proteostasis . Our data indicate that the nutritional context of an animal at the onset of hypoxia has the power to alter hypoxia’s effect on proteostasis and that the insulin-like signaling ( IIS ) pathway plays a role in fasting’s ability to protect against proteostasis decline independently of the canonical downstream transcription factor DAF-16/FOXO .
In order to investigate the effect of nutritional status on proteostasis in hypoxia , we first used transgenic C . elegans that express yellow fluorescent protein ( YFP ) fused to a polyglutamine tract in the body wall muscles [11] . We refer to these animals as QX::YFP , where X refers to the number of glutamine residues fused to YFP , such that Q35::YFP animals express YFP with 35 glutamine residues . In these animals , the number of YFP foci , which correspond to large protein aggregates , can be used as an in vivo measure of cellular proteostasis [12] . Exposing animals to 1000 ppm O2 ( 0 . 1% , with balance N2 ) for 24 hours while fed resulted in an increase in the number of YFP foci ( Fig 1B–1D ) , consistent with previous reports that hypoxia inhibits proteostasis [10] . However , we found that the number of YFP foci that formed in hypoxia was dramatically reduced if the animals were removed from food for six hours before the hypoxic exposure and remained off of food for the duration of hypoxia ( Fig 1A ) . Hypoxia-induced protein aggregation ( HIPA ) was prevented by fasting in fourth-stage larvae ( L4 ) Q35::YFP animals ( Fig 1C ) as well as in first-stage larvae ( L1 ) Q40::YFP ( Fig 1D ) . We verified that the bright fluorescent puncta in animals exposed to hypoxia were aggregated protein using fluorescence recovery after photobleaching ( S1 Fig ) , as had been previously observed for age-associated aggregates [11] . We also determined that the abundance of Q35::YFP was the same in animals exposed to hypoxia when fed and fasted ( S2 Fig ) . This is consistent with previous observations that the expression of Q35::YFP did not change even after nine days without food [4] . In control experiments we found no change in the number of protein aggregates between animals in room air that were fed and fasted ( S3 Fig ) . This is likely because we initiated these experiments before much age-associated protein aggregation had occurred , in order to avoid confounding factors from the effects of fasting and hypoxia on aging . From these data , we conclude that fasting prevents HIPA . We originally chose to fast animals for 6h before exposure to hypoxia to allow animals time to alter gene expression [13] . However , there is no a priori evidence that the protective effects of fasting in hypoxia requires changes in gene expression . Therefore , we measured how long of a fasting period was required to mitigate the effects of hypoxia on aggregation of polyglutamine proteins . To determine the pre-hypoxia fasting duration required to protect against HIPA , we removed Q35::YFP animals from food for varying lengths of time before being exposed to hypoxia ( schematic in Fig 2A ) . We found that animals removed from food immediately before exposure to hypoxia developed significantly fewer YFP foci in hypoxia as compared to controls that remained on food in hypoxia ( Fig 2A , 6h fed compared to fed ) . We conclude that extended fasting before exposure to hypoxia is not required to prevent HIPA . Instead , our data show that the protective effects of fasting occur very rapidly . In fact , the full protection against HIPA is realized with only 2h fasting before exposure to hypoxia ( Fig 2A ) . These results suggest that at least some of the protective effects of fasting do not require a period of adaptation to fasting prior to the hypoxic exposure . Instead , we conclude the environment at the onset of the exposure to hypoxia dictates at least some aspects of the response to hypoxia . Work in other systems has shown that fasting can have a protective effect that persists even after animals are returned to food [14] . To further explore the requirements for fasting to protect against HIPA we next asked whether the protective effects of fasting against HIPA could be reversed . In these experiments ( Fig 2B ) , we began fasting animals 6h before exposure to hypoxia but then returned the animals to food prior to initiation of hypoxia . We observed that animals fasted for a full 6h and then returned to food immediately before exposure to hypoxia ( Fig 2B , 6h fasted ) developed significantly more YFP foci than animals that were fasted for 6h and then exposed to hypoxia in the absence of food ( Fig 2B , fasted ) , suggesting that the nutritional context of an animal as it experiences hypoxia is able to mediate the effect of hypoxia on proteostasis . Furthermore , we found no protection from HIPA if animals were fasted for 4h , but then fed for 2 h before exposure to hypoxia ( Fig 2B , 4h fasted ) , even though 4h of fasting was sufficient for complete protection against HIPA in the absence of food ( Fig 2A , 2h fed ) . This result indicates that the protective effects of fasting are fully reversed within 2h of return to food . We conclude that the protective effects of fasting in hypoxia are rapidly reversed . Shorter exposures to hypoxia , which do not immediately increase the number of polyglutamine protein aggregates , still disrupt long-term proteostasis as evidenced by the increased rate of age-associated protein aggregation after return to room air [10] . We therefore asked whether fasting could protect against these long-term proteostasis deficits in addition to HIPA . We exposed Q35::YFP L4 animals to hypoxia for only 10h either in the fed state or after fasting for 6h ( F = 6 hours , H = 10 hours as per Fig 1A ) . Control animals remained on food in room air . Immediately after this short hypoxic exposure , there was no observed increase in the number of YFP foci in animals exposed to hypoxia regardless of whether food was present ( Fig 3 , 0 hours post-hypoxia ) . As expected , the animals exposed to hypoxia in the fed state accumulate aggregates faster than control animals . In contrast , animals exposed to hypoxia while fasted accumulate YFP foci at the same rate as control animals . Animals that were fasted in room air also accumulated YFP foci at the same rate as room air , fed controls ( S3 Fig ) . These data indicate that fasting both prevents HIPA and protects against the long-term effects on proteostasis induced by a short exposure to hypoxia . The cellular role of protein aggregates is controversial , with some reports finding a protective role and others suggesting a cytotoxic effect [15] . We have previously found that aggregates induced by hypoxia are likely cytotoxic , as they accelerate polyQ-associated paralysis even after animals are returned to room air [10] . We therefore next asked if fasting would protect against increased proteotoxicity in addition to HIPA . To address this , we exposed cohorts of L1 Q40::YFP animals to hypoxia for 24 hours while fed or fasted , then returned the animals to room air and measured the onset of paralysis in each cohort . We found that fasting slowed the rate at which paralysis developed relative to animals exposed to hypoxia while fed ( Fig 4A ) . There was no difference in the rate of paralysis onset if animals were fasted in room air ( S3 Fig ) . This result indicates that fasting protects against hypoxic effects of increased protein aggregation and proteotoxicity . We next sought to determine whether fasting’s protective effects on proteostasis extend to other models of proteotoxicity . Human amyloid β ( Aβ ) 1-42 peptide expressed in the body wall muscles of C . elegans results in cytoplasmic plaque formation , with a subsequent phenotype of progressive paralysis [16] . C . elegans expressing Aβ1–42 in their body wall muscles become paralyzed more quickly when they are exposed to hypoxia [10] . We found that this effect of hypoxia was reversed by fasting , as the rate that paralysis develops is slowed if animals expressing Aβ1-42 are exposed to hypoxia while fasting ( Fig 4B ) . Because Aβ1–42 and Q40::YFP are both expressed in body wall muscles , we also evaluated if fasting protected animals expressing a metastable version of the neuronal dynamin protein DYN-1 from the effects of hypoxia . The dyn-1 ( ky51 ) mutant contains a temperature-sensitive mutation , such that the DYN-1 protein is functional and dyn-1 ( ky51 ) mutant animals exhibit wild-type motility at the permissive temperature ( 20°C ) , but become uncoordinated at the restrictive temperature ( 28°C ) due to improper folding of the DYN-1 protein [17] . Genetic and environmental factors that disrupt proteostasis , including hypoxia , prevent the proper folding of the DYN-1 protein at the permissive temperature , thereby rendering the dyn-1 ( ky51 ) animals uncoordinated [10 , 18] . Similar to our experiments with Q40::YFP and Aβ1–42 , we found that fasting dyn-1 ( ky51 ) mutant animals before exposure to hypoxia results in a partial rescue of hypoxia-induced uncoordination at the permissive temperature ( Fig 4C ) . Together , our results suggest that fasting has a general protective effect against proteostasis defects induced by hypoxia , and that this protective effect is not specific to a particular tissue , developmental stage , or misfolded/aggregation prone model . Dysregulation of insulin-like signaling ( IIS ) has been tied to protein aggregation and neurodegeneration in a number of model organisms [19] . As the IIS pathway links food availability to growth , development , stress resistance , and aging , we hypothesized that changes in IIS could explain how fasting modulates the effect of hypoxia on proteostasis . The IIS pathway is widely conserved in metazoans [20] . We therefore explored the hypothesis that IIS would mediate the effects of fasting to prevent HIPA . We first looked at the localization of DAF-16::GFP in animals exposed to hypoxia to determine if IIS is active in hypoxia . DAF-16 is the C . elegans orthologue of the FOXO transcription factor . When active , the insulin/IGF-like receptor DAF-2 initiates a phosphorylation cascade that results in the phosphorylation and nuclear exclusion of DAF-16 protein [21 , 22] . Conversely , when nutrients are scarce , DAF-16 remains unphosphorylated by upstream kinases and is able to enter the nucleus and bind to its target genes [22 , 23] . We found that DAF-16::GFP remained diffuse and cytoplasmic in control worms maintained in room air on food ( Fig 5B and 5C ) , but accumulated in the nucleus of animals that were removed from food in room air ( Fig 5B and 5C ) or were exposed to hypoxia on food ( Fig 5B and 5C ) . These results suggest that IIS activity is reduced by fasting and hypoxia , consistent with previous reports [24 , 25] . Surprisingly , DAF-16::GFP did not accumulate in the nuclei of animals exposed to hypoxia after fasting ( Fig 5B and 5C ) , despite hypoxia and fasting both individually resulting in nuclear accumulation . These DAF-16::GFP localization patterns led us to interrogate requirements for DAF-16 and the upstream IIS receptor DAF-2 in mediating fasted and fed responses to hypoxia . To this end , we crossed the Q35::YFP transgene into daf-2 ( e1370 ) and daf-16 ( mu86 ) backgrounds . The fact that DAF-16::GFP is localized to the nucleus in fed animals exposed to hypoxia suggests the possibility that DAF-16 facilitates HIPA . However , we found that Q35::YFP; daf-16 ( mu86 ) mutant animals exhibit robust HIPA on food ( Fig 5D ) , indicating that DAF-16 is not required for HIPA despite its nuclear accumulation in fed hypoxic animals . We also asked if there was a genetic requirement for the IIS receptor DAF-2 . Our data indicate that IIS does not mediate the effects of hypoxia on proteostasis in fed animals , as Q35::YFP; daf-2 ( e1370 ) mutant animals exhibit robust HIPA when fed ( Fig 5D ) . Thus , neither DAF-16 nor DAF-2 activities are required for HIPA in fed animals . Given the IIS-independent nature of HIPA in fed animals , we next investigated whether fasting protection requires IIS . We discovered that DAF-2 , but not DAF-16 is required for fasting protection against HIPA . Fasting protects the Q35::YFP; daf-16 ( mu86 ) similar to wild-type ( Fig 5D ) ; however , we observe significant HIPA when Q35; daf-2 ( e1370 ) and Q35; daf-2 ( e1368 ) mutant animals are exposed to hypoxia when fasted ( Figs 5D and S4 ) . These results show that protective effects of fasting in hypoxia require DAF-2 , but not DAF-16 . This is consistent with our observation that DAF-16::GFP is not localized to the nucleus in fasted animals exposed to hypoxia ( Fig 5B and 5C ) . We found that the insulin/IGF-like receptor DAF-2 mediates the protective effects of fasting on HIPA , while the FOXO transcription factor DAF-16 is not required for protection . Given this finding , we also checked the DAF-16::GFP localization pattern in daf-2 ( e1370 ) animals . These mutants have constitutively nuclear DAF-16 in the fed state due to decreased signaling through the IIS pathway [22] . Since DAF-16::GFP is not localized to the nucleus in fasting-protected wild-type animals exposed to hypoxia , we sought to investigate whether the nuclear localization of DAF-16 in daf-2 ( e1370 ) mutants , which are not protected by fasting , would be altered by hypoxia . We found that DAF-16::GFP is fully nuclear in all conditions , including fasted hypoxia , in these animals ( S5 Fig ) . In C . elegans , DAF-16 mediates the effects of decreased signaling through DAF-2 . Mutations in daf-16 suppress most daf-2 mutant phenotypes including increased lifespan , enhanced dauer formation , increased fat storage , reproductive delays , and increased resistance to heat and oxidative stress [26 , 27] . This coupled with the nuclear localization of DAF-16::GFP in daf-2 mutants led us to hypothesize that daf-16 would be required for the HIPA in fasted Q35; daf-2 ( e1370 ) mutant animals . While Q35; daf-16 ( mu86 ) mutant animals were protected from HIPA by fasting similar to wild-type controls , Q35; daf-2 ( e1370 ) ; daf-16 ( mu86 ) animals still exhibit significant HIPA when fasted ( Fig 5D ) . These results indicate that DAF-2 mediates the effects of fasting to prevent HIPA at least partly independently of DAF-16 . We took a candidate approach to attempt to identify factors that act downstream of daf-2 to protect proteostasis in hypoxia . We focused first on hif-1 , the hypoxia inducible transcription factor [28] . We previously demonstrated that HIF-1 activity helps to blunt the effects of hypoxia on proteostasis [10] . Moreover , increased lifespan of C . elegans exposed to hypoxia depends on both hif-1 and daf-16 [25] . To determine if hif-1 is acting downstream of daf-2 to protect proteostasis in hypoxia we compared daf-16; daf-2; Q35::YFP mutant animals with hif-1; daf-16; daf-2; Q35::YFP animals . We find that deletion of hif-1 does not suppress increased protein aggregation in fasted animals exposed to hypoxia ( S6 Fig ) . We similarly investigated the role of skn-1 , which is required for increased lifespan of daf-2 mutants [29] . Our data show that fasted skn-1; daf-16; daf-2; Q35::YFP exhibit HIPA that is indistinguishable from the daf-16; daf-2; Q35::YFP mutant animals ( S6 Fig ) . Finally , we evaluated whether the heat-shock factor hsf-1 was involved . HSF-1 is required downstream of DAF-2 for increased lifespan [30] . However , we find that expression of HSF-1 targets hsp-16 . 2 , hsp-70 , and hsp-4 is induced to the same degree in fed and fasted animals exposed to hypoxia ( S7 Fig ) , suggesting that differential activity of HSF-1 does not underlie the protective effects of fasting . In fact , the expression of a variety of genes induces by proteotoxic stress , including the unfolded protein response and autophagy , are similarly induced in fed and fasted animals ( S7 Fig ) . Together , these results suggest that the mechanism ( s ) by which fasting can protect proteostasis in hypoxia is independent of daf-16 , hif-1 , hsf-1 , and skn-1 . The components of this daf-2 dependent pathway that can modulate proteostasis are , as yet , a mystery .
This study illustrates the power of fasting to ameliorate the deleterious effects of hypoxia on proteostasis . These findings are consistent with phenomena that have been observed in mammals–fasting mice for a single day increases survival after kidney ischemia and also reduces ischemic damage to the liver [31] . Our results suggest that the nutritional milieu present at the onset of hypoxia can dictate the effect of hypoxia on proteostasis , as fasting protection against hypoxia can be induced quite quickly . Animals that are removed from food immediately before hypoxia are protected against HIPA to a significant degree , even after being maintained on food for the entire pre-hypoxic period . This implies that worms are integrating information about their environment , including nutrient availability , concurrently with the perception of hypoxia . The importance of the nutritional environment of the animal as it experiences hypoxia is further supported by the fact that we also see a rapid reversal of fasting protection . Worms fasted for six hours but that are moved onto food immediately preceding hypoxia are not as protected against HIPA compared to worms that were fasted and remained off of food for the duration of hypoxia . The speed with which fasting protection can be induced and reversed indicates that protection cannot be explained solely by changes in gene expression resulting in a hypoxia-resistant pre-adapted state . Furthermore , the rapidity with which fasting protection can be reversed suggests that altered gene expression or metabolism resulting from the fasting period is alone insufficient to protect against HIPA . Although C . elegans enter a reproductive and developmental diapause in 1000 ppm O2 [32] , the protection conferred by fasting does not represent a simple delay in the onset of proteostasis decline due to the time spent in hypoxia . Rather , fasting provides long-term protection against the accrual of protein aggregates and toxicity even after the return to room air . We found that the IIS receptor DAF-2 is required for fasting to prevent HIPA . This is somewhat counterintuitive , as decreased function of daf-2 mutants could be thought of as “phenocopying” the fasted situation . Consistent with this , both fasting and mutation of daf-2 lead to increased nuclear localization of DAF-16 . However , our results show that daf-2 mutant animals do not phenocopy wild-type , fasted animal ( which show little HIPA ) . In contrast , daf-2 mutant animals exhibit robust HIPA regardless of whether they are fed or fasted . Moreover , we found that while hypoxia and fasting individually promote the nuclear localization of DAF-16::GFP , there is no nuclear accumulation in in fasted animals exposed to hypoxia . These results suggest that activation of DAF-2 in fasted animals is required to prevent hypoxia-induced perturbations of proteostasis . Although required for the protective effects of fasting in hypoxia , our data show that IIS is not required for the normal response to hypoxia in fed animals–both daf-16/FOXO and daf-2/IR mutants have relatively normal HIPA when fed . This contrasts with previous studies that show C . elegans daf-2 mutant animals are resistant to anoxia , displaying reduced muscle and neuronal cell death following anoxia [33 , 34] . Similarly , flies with defective insulin signaling due to mutations in the insulin receptor InR , or Chico , the insulin receptor substrate , are protected against anoxia/reoxygenation injury [35] . The discordance between our results and these previous studies may be due to the fact that the phenotypic and genetic responses to hypoxia depend strongly on the precise concentration of O2 available ( reviewed in [36] ) ; in our studies we focused on hypoxic conditions with 1000 ppm O2 , whereas the anoxic conditions used in these previous studies had far less O2 available . Our results suggest that , in fed animals , IIS is not required for nor can it protect against hypoxia-induced disruption of proteostasis . Mammalian systems offer precedents of insulin receptor mutations causing sensitivity to hypoxic stress . Knockdown of neuronal insulin-like growth factor 1 receptor ( IGF-1R ) exacerbates hypoxic injury and increases mortality in mice [37] , and IGF-1R is required in order for IGF-1 to protect myocardial cell exposed to ischemia [38] . However , data on the role of mammalian IIS in response to hypoxia are mixed , and are complicated by the fact that different types of insulin receptors mediate distinct cellular functions [39] . As such , the simplified C . elegans IIS system may be useful for understanding contextual inputs that alter IIS outputs . DAF-16 is believed to be the main nexus of IIS [22 , 30 , 40 , 41] , which makes the DAF-2-dependent , but DAF-16-independent nature of the protective effect of fasting we have described unusual in C . elegans . Decreased DAF-2 activity results in phenotypes such as increased lifespan , reproductive delays , and increased resistance to heat and oxidative stress , all of which require DAF-16 [27] . However , a few other examples exist in the literature of DAF-2 dependent , DAF-16 independent phenomena: dauer formation at 27° , meiotic progression of oocytes , salt chemotaxis learning , and regulation of the dao-3 and hsp-90 genes [40 , 42–45] . Our studies suggest that fasting-mediated protection against HIPA is mediated by factors that act downstream of DAF-2 , but separate from DAF-16 . Understanding the nature of these factors could reveal new aspects of how IIS modulates stress responses and proteostasis in animals .
Animals were maintained on nematode growth media ( NGM ) with OP50 E . coli at 20°C [46] . See S10 Table for worm strains . Strains were obtained from the Caenhorabditis Genetics Center at the University of Minnesota . Double and triple mutants were generated using standard genetic techniques , and genotypes were verified using PCR . Hypoxic conditions were maintained using continuous flow chambers , as described in [49] . Compressed gas tanks ( 1000 ppm O2 balanced with N2 ) were Certified Standard ( within 2% of target concentration ) from Airgas ( Seattle , WA ) . Oxygen flow was regulating using Aalborg rotameters ( Aalborg Intruments and Controls , Inc . , Orangeburg , NY , USA ) . Hypoxic chambers ( and room air controls ) were maintained in a 20°C incubator for the duration of the experiments . Synchronous cohorts of L1 YFP::polyQ40 animals were generated by either bleaching first-day adult animals in a 20% alkaline hypochlorite solution or allowing first-day adult animals to lay eggs for 1–2 hrs on seeded NGM plates . The adults were then removed , and the plates were incubated at 20°C . The next morning , cohorts of hatched L1 larvae were suspended in M9 and mouth-pipetted to new NGM plates for hypoxic exposure . Synchronous cohorts of L4 YFP::polyQ35 animals were generated by picking L4 animals from well-fed , logarithmically growing populations . Cohorts of 25–35 YFP::polyQ animals were exposed to hypoxia for approximately 24 h at 20°C on unseeded 3 cm NGM plates with 40mg/mL carbenicillin or NGM plates seeded with live OP50 food . Plates were ringed with palmitic acid ( 10mg/mL in ethanol ) , creating a physical barrier around the edge of each plate to discourage animals from leaving the surface of the agar . To quantify the number of YFP foci , worms were mounted a 2% agar pad in a drop of 50mM sodium azide as anesthetic . Control experiments showed that azide did not affect the aggregation of YFP::polyQ35 or YFP::polyQ40 [47] . YFP foci were identified and quantified as described in [11] and [48] . A Nikon 90i fluorescence microscope with the YFP filter and 10x objective ( Nikon Instruments Inc . , Melville , NY , USA ) was used to visualize and quantify aggregates . In all experiments , the number of aggregates was counted blind to treatment and genotype . Statistical significance was evaluated by calculating P-values between conditions using a Kruskal-Wallis test and Dunn’s multiple comparisons post hoc analysis in GraphPad Prism version 7 . 0c for Mac OSX ( GraphPad Softare , San Diego , California , USA ) In all cases , P < 0 . 05 was considered statistically significant . Animals expressing Aβ1–42 or YFP::polyQ40 were exposed to 1000 ppm O2 for 24 at 20°C as L4 or L1 , respectively . For both , animals were grown on seeded NGM plates until 6 hrs before hypoxic exposure , at which point fasted animals were transferred to unseeded NGM plates , where they remained until the end of the hypoxic exposure . Fed animals were transferred to new seeded NGM plates . After hypoxic exposure , all animals were returned to food and normoxia , and incubated at 20°C . Paralysis was scored daily . Worms were considered paralyzed if they failed to respond , other than with movement of the nose or pharyngeal pumping , when tapped with a platinum wire pick 3 consecutive times . Dead or bagged worms were censored from the experiment on the day of death/bagging . Paralyzed worms were removed from the plate on the day of paralysis . Live worms that were not paralyzed were moved to a new plate each day until all worms were scored as either paralyzed or dead . Statistical significance was calculated using Kaplan-Meier log-rank ( Mantel-Cox ) tests and a Bonferroni correction for multiple comparisons using GraphPad Prism version 7 . 0c for Mac OSX ( GraphPad Softare , San Diego , California , USA ) . Synchronous cohorts of L2 animals expressing DAF-16::GFP were exposed to hypoxia for 24 h at 20°C on unseeded 3 cm NGM plates with 40mg/mL carbenicillin or NGM plates seeded with live OP50 food . Plates were ringed with palmitic acid ( 10mg/mL in ethanol ) , creating a physical barrier around the edge of each plate to discourage animals from leaving the surface of the agar . To visualize the localization of DAF-16::GFP , worms were mounted a 2% agar pad in a drop of 10mM levamisole as anesthetic . A Nikon 90i fluorescence microscope with the GFP filter and 10x objective ( Nikon Instruments Inc . , Melville , NY , USA ) was used to visualize DAF-16::GFP . For quantification , percent of animals with nuclear GFP was scored immediately after removal from hypoxia . In all experiments , the GFP localization was scored blind to treatment and genotype . Statistical significance was evaluated by calculating P-values between conditions using a Kruskal-Wallis test and Dunn’s multiple comparisons post hoc analysis in GraphPad Prism version 7 . 0c for Mac OSX ( GraphPad Softare , San Diego , California , USA ) . P < 0 . 05 was considered statistically significant . Animals were grown on NGM plates seeded with OP50 E . coli at 20°C . When animals reached gravid adult , synchronized embryos were obtained by a 5-minute bleach in 1:1:5 water:KOH:hypochloric acid solution . For each strain/condition , ~9 , 000 embryos were plated onto a 150 mM NGM plate seeded with live OP50 E . coli . Animals were not allowed to starve out the plate at any time during the experiment . When animals reached L4 , they were exposed to hypoxia with or without a 6 hour fasting period , or were left in room air as controls . Animals were harvested into 1 mL Trizol solution and immediately frozen in liquid nitrogen , as described previously [49] . RNA was isolated from the Trizol preparation as described previously [50] . cDNA was made using Invitrogen SuperScript III First Strand Synthesis System . qPCR was performed using Kappa SYBR FAST qPCR Kit . PCR cycle was as follows: 95C for 3 min , 95C for 15 sec , 55C for 15 sec x40 . 4°C to hold . qRT-PCR values were analyzed as described in [51] . In summary , ΔCt for each gene product was calculated by subtracting Ct values from the geometric mean of the control targets that are not altered in response to fasting or hypoxia ( hil-1 , irs-2 , and tba-1 ) . ΔCt were averaged across experiments . Student’s t-test was used to evaluate differences between ΔCt values of treated samples and untreated controls . For differences between genotypes , p-values were calculated with a one-way ANOVA from summary statistics ( mean , standard error , n ) . Reported fold-changes were calculated as 2^-ΔΔCt where ΔΔCt = ΔCt ( experimental condition ) —ΔCt ( control condition ) . Error bars on graphs represent standard error of the mean . | When blood flow to various parts of the body becomes restricted , those tissues suffer from a lack of oxygen , a condition called hypoxia . Hypoxia can cause cellular damage and death , as in stroke and cardiovascular disease . We have found that in the model organism C . elegans ( a roundworm ) specific concentrations of hypoxia cause aggregation of polyglutamine proteins–the same kind of proteins that are found in an aggregated state in the neurodegenerative disorder Huntington’s disease . Here , we show that that worms can be protected from hypoxia-induced protein aggregation if they are fasted ( removed from their food source ) prior to experiencing hypoxia . Furthermore , we show that the insulin receptor is required for this protection . The insulin receptor is responsible for detecting insulin , a hormone that is released after feeding . Worms with a nonfunctional version of the insulin receptor displayed hypoxia-induced protein aggregation despite being fasted before the hypoxic exposure . Our results highlight a new role for the insulin signaling pathway in coordinating the effects of both hypoxia and nutritional state on protein aggregation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
]
| [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"diabetic",
"endocrinology",
"caenorhabditis",
"gene",
"regulation",
"protein",
"aggregation",
"regulatory",
"proteins",
"dna-binding",
"proteins",
"light",
"microscopy",
"animals",
"pulmonology",
"hormones",
"animal",
"models",
"caenorhabditis",
"elegans",
"luminescent",
"proteins",
"model",
"organisms",
"microscopy",
"yellow",
"fluorescent",
"protein",
"hypoxia",
"experimental",
"organism",
"systems",
"transcription",
"factors",
"research",
"and",
"analysis",
"methods",
"medical",
"hypoxia",
"insulin",
"animal",
"studies",
"proteins",
"endocrinology",
"gene",
"expression",
"fluorescence",
"recovery",
"after",
"photobleaching",
"biochemistry",
"eukaryota",
"cell",
"biology",
"genetics",
"nematoda",
"biology",
"and",
"life",
"sciences",
"organisms"
]
| 2019 | Fasting prevents hypoxia-induced defects of proteostasis in C. elegans |
Significant departures from expected Mendelian inheritance ratios ( transmission ratio distortion , TRD ) are frequently observed in both experimental crosses and natural populations . TRD on mouse Chromosome ( Chr ) 2 has been reported in multiple experimental crosses , including the Collaborative Cross ( CC ) . Among the eight CC founder inbred strains , we found that Chr 2 TRD was exclusive to females that were heterozygous for the WSB/EiJ allele within a 9 . 3 Mb region ( Chr 2 76 . 9 – 86 . 2 Mb ) . A copy number gain of a 127 kb-long DNA segment ( designated as responder to drive , R2d ) emerged as the strongest candidate for the causative allele . We mapped R2d sequences to two loci within the candidate interval . R2d1 is located near the proximal boundary , and contains a single copy of R2d in all strains tested . R2d2 maps to a 900 kb interval , and the number of R2d copies varies from zero in classical strains ( including the mouse reference genome ) to more than 30 in wild-derived strains . Using real-time PCR assays for the copy number , we identified a mutation ( R2d2WSBdel1 ) that eliminates the majority of the R2d2WSB copies without apparent alterations of the surrounding WSB/EiJ haplotype . In a three-generation pedigree segregating for R2d2WSBdel1 , the mutation is transmitted to the progeny and Mendelian segregation is restored in females heterozygous for R2d2WSBdel1 , thus providing direct evidence that the copy number gain is causal for maternal TRD . We found that transmission ratios in R2d2WSB heterozygous females vary between Mendelian segregation and complete distortion depending on the genetic background , and that TRD is under genetic control of unlinked distorter loci . Although the R2d2WSB transmission ratio was inversely correlated with average litter size , several independent lines of evidence support the contention that female meiotic drive is the cause of the distortion . We discuss the implications and potential applications of this novel meiotic drive system .
Mendel’s Laws provide the theoretical foundation of transmission genetics and explain many of the inheritance patterns of biological traits in sexually reproducing organisms . The Laws state that each gamete receives a random collection of alleles—exactly one per pair of homologous loci—and that gametes unite at random . However , reports of exceptions to Mendelian inheritance date back almost to the rediscovery of Mendel’s Laws , and have been instrumental in elucidating the mechanisms of genetic inheritance [1–4] . Transmission ratio distortion ( TRD ) is defined as a significant and reproducible violation of the inheritance ratios expected under Mendel’s Laws [1 , 5–7] . Most observations of TRD are due to selection acting upon the products of meiosis ( gamete selection ) or fertilization ( differential pre- or post-natal survival ) [5–8] . The latter is a relatively common occurrence in experimental crosses in many types of organisms including plants and animals [8 , 9] , and is routinely used to classify the essentiality of genes and alleles [9–14] . However , a small but increasing number of observations of TRD can be ascribed to the differential segregation of alleles during meiosis , a process called meiotic drive [1 , 10–14] . To qualify as such , meiotic drive systems must exhibit three characteristics: 1 ) asymmetry in the meiotic division ( s ) with respect to cell fate; 2 ) functional asymmetry of the meiotic spindle poles; and 3 ) functional heterozygosity at a locus that mediates attachment of a chromosome or a chromatid to the meiotic spindle [1 , 15 , 16] . Meiotic drive is an evolutionary force thought to contribute to karyotypic evolution [15–17] and maintenance of non-essential “B chromosomes” in multiple clades [17 , 18] . The incidence of meiotic drive is unknown , but given that it is a relatively strong evolutionary force that can lead to the rapid fixation of a selfish allele , it should be rare to observe in action [18–20] . In most plant and animal species meiotic drive is restricted to females , which undergo asymmetric meiosis . At the locus where TRD is observed , an allele that is subject to preferential segregation is termed a responder [19–21] . There are examples in many species of meiotic drive responder alleles that , when in heterozygosity , succeed in being transmitted to the functional product of the asymmetric meiosis more than half of the time ( S1 Fig . ) . Responders in known meiotic drive systems typically involve multi-megabase , highly repetitive and heterochromatic sequences , such as the D locus in monkeyflower [11 , 21] , knobs in maize [11 , 22] , homogenously staining regions ( HSRs ) in wild mice [12 , 17 , 22 , 23] and centromeres and B chromosomes in multiple species [12 , 17 , 23 , 24] . Those systems have mostly proven intractable to molecular characterization , and thus the mechanism ( s ) by which they gain a segregation advantage are largely unknown . Meiotic drive may be promoted or suppressed by distorter loci ( alternately referred to in some publications as effectors , modifiers or drivers ) . It is rare for TRD at any single locus to be observed in multiple independent genetic backgrounds . An exception is TRD on mouse Chromosome ( Chr ) 2 , which was reported first in interspecific backcrosses between C57BL/6J ( a classical inbred strain , primarily of Mus musculus domesticus origin [24–28] ) and SPRET/EiJ ( a Mus spretus wild-derived inbred strain ) [25–28] . In offspring from two different ( C57BL/6JxSPRET/EiJ ) xC57BL/6J backcrosses , the SPRET/EiJ allele was overrepresented across a 40 cM region on Chr 2 [25 , 28] and a ~140 Mb region on Chr 2 with a maximum transmission frequency of 0 . 66 [25 , 29] . TRD in Chr 2 was also reported in an F2 cross between two body weight selection lines , one of which ( high body weight; M16i ) was derived from the Hsd:ICR outbred stock ( also known as CD-1 ) [29 , 30] . Additionally , in an advanced intercross between the Hsd:ICR-derived high-running selection line HR8 [30–32] and C57BL/6J , TRD in Chr 2 was present in the primary data but not reported in the corresponding manuscripts [31–35] . And recently , we reported TRD in Chr 2 in the Collaborative Cross ( CC ) [33–35] . The CC is a mouse recombinant inbred panel derived from eight genetically diverse inbred strains: the classical strains A/J , C57BL/6J , 129S1/SvImJ , NOD/ShiLtJ and NZO/HlLtJ and the wild-derived strains PWK/PhJ ( M . m . musculus origin ) , CAST/EiJ ( M . m . castaneus ) and WSB/EiJ ( M . m . domesticus ) [35] . We reported TRD in favor of the WSB/EiJ allele across a ~50 Mb region in the middle of Chr 2 in three largely independent sets of CC lines . In the largest sample , involving 350 genetically independent CC lines , the WSB/EiJ allele was present on 22% of Chr 2 [35 , 36] , a significant over-representation compared to the expected frequency of 12 . 5% ( 1/8 ) . Here we report our extensive genetic characterization of Chr 2 TRD in the CC and in the Diversity Outbred ( DO ) , an outbred population initiated from 144 incompletely inbred CC lines and specifically tailored for high resolution mapping of complex traits [33–36] . Using a combination of classical genetics , whole genome sequence analysis and bioinformatics , we demonstrate conclusively that maternal transmission distortion is caused by a large copy number gain of a 127 kb DNA segment containing a single gene , Cwc22 . We also provide compelling evidence that meiotic drive is required to explain the TRD in the progeny of heterozygous dams . Finally , we show that there exist several genetically determined levels of TRD controlled by unlinked genetic variation , which , to our knowledge , is unique among meiotic drive systems .
To test whether TRD of the WSB/EiJ allele in Chr 2 is present in the DO , we analyzed 1 , 175 animals from DO generation 8 ( G8 ) that were genotyped using two related genotyping arrays ( MUGA or MegaMUGA , see Materials and Methods ) . We sampled the genotypes of each individual at 1 Mb intervals along Chr 2 and then computed the overall frequencies of the eight founder alleles at each position . The WSB/EiJ allele was over-represented relative to the other seven founder alleles across a roughly 100 Mb region in the middle of Chr 2 ( S2 Fig . ) . However , there was a striking difference in the level of distortion observed in the CC and the DO , with the WSB/EiJ allele frequency reaching a maximum of 0 . 22 in the CC compared to 0 . 55 in the DO . This result indicates that the additional outcrossing in the DO is associated with higher levels of TRD . We conclude that TRD favoring the WSB/EiJ allele is a general feature of crosses in the CC genetic background; however , the level of TRD may vary widely depending on the number of generations of outbreeding . To determine the parental origin of the TRD , we analyzed 5 , 499 offspring from 18 experimental crosses in which exactly one parent was heterozygous for the WSB/EiJ allele in an interval spanning the region of maximum distortion on Chr 2 ( 75–90 Mb ) [33–35 , 37 , 38] . In all cases the heterozygous parent was an F1 hybrid derived either from an intercross between the WSB/EiJ inbred strain and one of eight other inbred strains ( the seven founder strains of the CC or PWD/PhJ ) , or from two CC strains , of which one was homozygous for the WSB/EiJ allele on Chr 2 and the other was homozygous for a non-WSB/EiJ allele . F1 hybrids were mated to either C57BL/6J or FVB/NJ mice , and their progenies were euthanized at birth and genotyped using genetic markers located in the region of maximum distortion . For each cross , we computed the TR of the WSB/EiJ allele and the non-WSB/EiJ allele using the aggregate genotypes across all litters from parents with identical genotypes ( Table 1 ) . TRs in six paternally segregating crosses ( rows 1–6 in Table 1 ) were as expected under the null hypothesis of Mendelian segregation ( range 0 . 482–0 . 524 , p ≥ 0 . 37 ) . In contrast , the mean TR in maternally segregating crosses ( rows 7–18 in Table 1 ) was 0 . 666 and deviated significantly from the null hypothesis ( p = 3 . 4x10–89 ) . We conclude that , in the genetic backgrounds tested , TRD in favor of the WSB/EiJ allele on Chr 2 is restricted to the progeny of heterozygous dams . The TRs among maternally segregating crosses were significantly different ( p = 2 . 4x10–90 ) , demonstrating that TRD depends on genetic background ( i . e . , TRD is under genetic control ) . The 12 crosses using F1 hybrid dams can be divided into three classes based on the observed TR ( S3 Fig . ) . F1 hybrid dams derived from crosses between WSB/EiJ and CAST/EiJ or PWD/PhJ showed no distortion ( crosses 7–10 in Table 1; aggregate TR = 0 . 485 , 95% CI = 0 . 46–0 . 51 , p = 0 . 23 ) . Moderate but significant distortion was present in F1 hybrid dams derived from crosses between WSB/EiJ and A/J , 129S1/SvImJ , NZO/HILtJ or NOD/ShiLtJ; and in ( CC042/GeniUncxCC001/Unc ) F1 hybrid dams ( crosses 11–15 in Table 1; aggregate TR = 0 . 645 , 95% CI = 0 . 61–0 . 68 , p = 8 . 3x10–19 ) . Finally , extreme distortion was observed in reciprocal ( WSB/EiJxC57BL/6J ) F1 hybrid dams and in ( CC001/UncxCC039/Unc ) F1 hybrid dams ( crosses 16–18 in Table 1; aggregate TR = 0 . 943 , 95% CI = 0 . 93–0 . 96 , p = 9 . 6x10–193 ) . We conclude that heterozygosity for the WSB/EiJ allele in the central region of Chr 2 is necessary but not sufficient to observe TRD , because TR was consistent with Mendelian inheritance in some dams that met that criterion . We also conclude that the grandparental origin of the WSB/EiJ allele has no influence on TRD because the TR levels were not significantly different between three pairs of reciprocal F1 dams ( compare crosses 7 and 8 , 9 and 10 and 17 and 18 in Table 1; p = 0 . 53 , 0 . 11 and 0 . 59 , respectively ) . To define the boundaries of the locus subject to TRD , we screened 61 CC lines and 378 DO mice that had been genotyped with MegaMUGA for recombinations involving the WSB/EiJ haplotype in the 75–90 Mb interval of Chr 2 . We identified five DO females ( DO-600 , DO-681 , DO-732 , DO-832 and DO-OCA45 ) and two CC strains ( CC039/Unc and CC042/GeniUnc ) that each had at least one informative recombination ( Fig . 1 ) . Next , we mated four of the DO females ( all except DO-OCA45 that was already heterozygous ) and the two CC strains to one of two additional CC lines ( CC001/Unc and CC005/TauUnc ) that had no contribution from WSB/EiJ on Chr 2 , to obtain heterozygous G1 hybrid females . Each hybrid female was genotyped with MegaMUGA and mated to FVB/NJ males ( total of 35 crosses; S1 Table ) . We found that dams carrying eight of the ten recombinant chromosomes exhibited significant TRD in the Chr 2 interval ( TR range 0 . 69–1 . 0 , p ≤ 2 . 1x10–5; Fig . 1 A ) , but dams carrying two other recombinant chromosomes did not ( TR = 0 . 48 and 0 . 37 , p ≥ 0 . 72; Fig . 1 B ) . These results are consistent with our conclusion that heterozygosity on Chr 2 is required but not sufficient for TRD; therefore , dams with Mendelian transmission ratios were not used for mapping the locus subject to TRD . Dams with TRD in favor of the WSB/EiJ allele were all heterozygous for a 9 . 3 Mb interval ( the candidate interval; boxed in Fig . 1 A ) . The proximal boundary of the candidate interval is defined by the recombination found in the CC strain CC039/Unc ( i . e . , the most distal SNP inconsistent with a WSB/EiJ haplotype ) . The distal boundary of the candidate interval is defined by the recombination found in DO-732 and DO-832 females ( i . e . , the most proximal SNP inconsistent with a WSB/EiJ haplotype ) . Those SNPs define the maximum boundaries of the locus subject to TRD , Chr 2 76 , 860 , 362–86 , 117 , 205 ( all positions from NCBI/37 unless otherwise noted ) . Among the eight CC founder strains , the candidate interval has 5 , 018 SNPs , 1 , 286 small insertions/deletions and 35 structural variants that are private to the WSB/EiJ strain [37–39] . Although this very large number of variants would typically make it difficult to confidently identify and prioritize candidates , one large structural variant has several unique features that made it a strong candidate causative allele for the TRD phenotype . That structural variant is a copy number gain of a 127 kb-long genomic DNA segment ( herein referred as R2d for responder to drive ) . In the reference genome , R2d is composed of nine non-contiguous sections that , in total , span 158 kb ( see R2d1 locus; Chr 2 77 , 707 , 014–77 , 865 , 265; Fig . 2 A; S2 Table ) . We used the normalized per-base read depth from whole-genome sequence alignments generated by the Sanger Mouse Genomes Project [31 , 32 , 37 , 39] and the HR8 selection line to estimate the number of copies of R2d in 18 inbred strains ( see Materials and Methods ) . Similar to C57BL/6J , 15 of the 18 strains , including 5 additional CC founder strains ( A/J , 129S1/SvImJ , NOD/ShiLtJ , NZO/HlLtJ and PWK/PhJ ) were copy number one ( i . e . , a single haploid copy ) , and CAST/EiJ was copy number two . In contrast , WSB/EiJ had an estimated copy number of 34 , and SPRET/EiJ had an estimated copy number of 36 , resulting in ~4 . 4 Mb of additional DNA in those strains ( Fig . 2 A ) . We sequenced 10 individuals from the HR8 selection line ( for which Chr 2 TRD was also observed when mated to C57BL/6J [31 , 32 , 40] ) to a total depth of 125x and aligned the reads to the reference genome . All 10 individuals had evidence of a copy number gain with the same boundaries as in WSB/EiJ and SPRET/EiJ ( Fig . 2 A; mean copy number 24 . 5 +/- 1 . 4 , equating to ~3 Mb of additional DNA ) . We used two additional methods to assay the copy number of R2d . First , we identified sets of probes on two different genotyping arrays for which the sum hybridization intensity was highly correlated with the copy numbers estimated from sequencing read depth ( 34 probes in MDA and 3 probes in MegaMUGA; S3 and S4 Tables , respectively ) . Second , we used real-time quantitative PCR to estimate the R2d copy number ( Fig . 2 B ) using TaqMan assays internal to exons of the single protein-coding gene within R2d , Cwc22 ( Fig . 2 C ) . Using that gene as a proxy for the copy number gain , we found that the copy number estimates from all three methods were highly concordant for the 28 sequenced strains/individuals . Using the TaqMan assay , we also found that the M16i inbred strain has a high number of copies of R2d ( Fig . 2 B ) . We conclude that a large increase ( > 20-fold ) in R2d copy number is found exclusively in strains with TRD ( WSB/EiJ , SPRET/EiJ , HR8 and M16i ) and that TRD consistently favors the transmission of the allele with the copy number gain . Many structural variants identified from whole-genome sequencing reads have uncertain genomic positions due to the challenge of mapping large variants that are absent from the reference genome . To determine the position of the copy number gain associated with R2d , we mapped the WSB/EiJ and CAST/EiJ alleles using segregating populations that have been genotyped at medium ( MegaMUGA ) or high ( Mouse Diversity Array , MDA ) density [26 , 40] . In the CC founder strains , probes located in R2d have hybridization intensities correlated with the number of copies estimated from aligned read depth and TaqMan CNV assays ( Fig . 2 A , B ) . The MDA provides robust discrimination between the reference ( one copy ) , CAST/EiJ ( two copies ) and WSB/EiJ alleles ( 34 copies; Fig . 3 A ) . MegaMUGA is able to identify mice carrying the WSB/EiJ allele with little ambiguity ( Fig . 3 B ) . Using the sum intensities of the informative probes as a quantitative trait , we mapped the WSB/EiJ and CAST/EiJ copy number gains in two independent populations and platforms . A genome scan identified a single , broad , highly significant peak on Chr 2 in each population , and those peaks overlap with each other and with the initial candidate interval for TRD ( Fig . 3 C-E ) . We conclude that the copy number gain is closely linked to R2d1 . This location is consistent with the large copy number gain being the causative allele . Note that both genome scans ( Fig . 3 C , D ) demonstrate that all the extra R2d copies found in WSB/EiJ are located in this interval because no other significant peak is observed in either scan . QTL mapping using TaqMan readout as the phenotype confirmed this result ( Fig . 3 D , E ) . Analysis of individual mice with recombinant chromosomes in the candidate interval revealed that the copy number gain maps to a 900 kb interval ( the R2d2 locus; Chr 2 83 , 631 , 096–84 , 541 , 308; Fig . 2; Fig . 3 A , B ) . Specifically , the CAST/EiJ copy number gain ( R2d2CAST; one additional copy of R2d ) is located distal to the transition from the CAST/EiJ to the NZO/HILtJ haplotypes found in mice OR3172m10 and OR3172f9 because both mice have low hybridization intensity consistent with a single copy , hence they lack R2d2CAST ( Fig . 3 A; S4A Fig . ) . Similarly , the WSB/EiJ copy number gain ( R2d2WSB; 33 additional copies of R2d ) is located proximal to the transition from the WSB/EiJ to the CAST/EiJ haplotype found on DO mouse DP2–446 , because it had high hybridization intensity consistent with the presence of R2d2WSB ( Fig . 3 B; S4B Fig . ) . These results demonstrate that R2d2 is not located immediately adjacent to R2d1 but approximately 6 Mb distal to it . The distal location of the copy number gain is confirmed by the analysis of the sum intensity of the three MegaMUGA probes that track R2d in two backcrosses involving the SPRET/EiJ inbred strain [26 , 41] ( S4C Fig . ) . We used the TaqMan assay to confirm R2d copy number in all heterozygous females tested for TRD ( S1 Table; S5 Fig . ) . We identified a dam ( DO-G13–44 ) that was homozygous for the WSB/EiJ haplotype across the entire candidate interval but produced offspring that were segregating for the copy number gain ( Fig . 4 A ) . This was confirmed by estimating R2d copy number in each of 27 G3 females and 16 G4 progeny that were heterozygous for a WSB/EiJ haplotype ( Fig . 4 B; S5 Fig . ) . We determined the TR in 825 progeny of G3 dams mated to FVB/NJ sires . The TRs among the 27 G3 dams were significantly different ( p = 4 . 9x10–12 ) . In the progeny of the 15 G3 dams with high copy number there was significant TRD in favor of the WSB/EiJ allele ( TR = 0 . 78 , p = 2x10–30; Fig . 4 C ) . In contrast , we found absence of TRD in the 12 G3 dams that inherited the low-copy allele ( TR = 0 . 53 , p = 0 . 234 ) . A genome scan for TRD as a binary trait demonstrated that presence or absence of TRD in this pedigree maps uniquely to the candidate interval ( Fig . 4 D , E ) . We were also able to estimate that G3 dams with the low-copy allele had a copy number of ~11 . We conclude that the loss of ~22 copies of R2d was sufficient to rescue Mendelian transmission , thus demonstrating that the copy number gain is causative of TRD . The results presented above demonstrate that TRD at R2d2 is only observed in the progeny of heterozygous dams . This restricts the plausible causes of TRD to meiotic drive , genotype-dependent embryonic lethality ( including genotype-dependent competition between embryos ) or a combination of both . To identify the cause of TRD , we first determined whether TR levels ( S6 Fig . ; S1 Table ) were correlated with litter size in 127 DO dams ( these 56 DO-G13 and 71 DO-G16 females are a random sample from an outbred population ) . We observed a strong inverse correlation between average litter size and TR at R2d2 ( r = -0 . 65 , p = 7 . 2x10–8 and r = -0 . 40 , p = 5x10–4 in the DO-G13 and DO-G16 dams , respectively; Fig . 5 A , B ) . We conclude that the presence and the strength of TRD are significantly associated with reduced litter sizes and thus with some type of embryonic lethality . We determined the relationship between TRD and litter size under the assumption of TRD caused exclusively by embryonic lethality [40 , 41] ( S7 Fig . ) . Under this scenario , in both the DO-G13 and DO-G16 samples the observed average litter size is significantly greater than predicted based on TR ( p = 0 . 021 and 6 . 0x10–5 for DO-G13 and DO-G16 dams , respectively; S7 Fig . ) . We conclude that embryonic death alone could only account for a fraction of the “missing” progeny inheriting a non-WSB/EiJ ( R2d2NotWSB ) allele . We determined directly the levels of embryonic lethality in DO-G13 dams at mid-gestation ( see Materials and Methods ) . We observed that dams with TRD had slightly , but not significantly , higher numbers of resorbed embryos present in utero than did dams with Mendelian segregation ( 1 . 3 ± 1 . 5 and 1 . 1 ± 1 . 2 resorbed embryos , respectively , p = 0 . 66; N = 29 and 19 dams , respectively; S8 Fig . ) . We conclude that embryonic lethality alone is insufficient to explain TRD at R2d2 . Although embryonic lethality can change the proportion of progeny inheriting alternative alleles at R2d2 , only meiotic drive can lead to an increase in the absolute number of progeny inheriting the R2d2WSB allele per litter in dams with TRD compared to dams with Mendelian segregation . To test whether meiotic drive was responsible for TRD , we determined the average absolute number of offspring per litter that inherited the R2d2WSB and R2d2NotWSB alleles in the progenies of the DO-G13 and DO-G16 DO dams with either TRD or Mendelian segregation . In dams with Mendelian segregation , the average numbers of offspring per litter that inherited either allele were not different ( 3 . 80 R2d2WSB versus 3 . 96 R2d2NotWSB , p = 0 . 73 in DO-G13 dams; 4 . 13 R2d2WSB versus 4 . 03 R2d2NotWS , p = 0 . 29 in DO-G16 dams; Fig . 5 A , B ) . In contrast , in the progenies of dams with TRD the average number of offspring per litter that inherited the R2d2WSB allele ( 4 . 51 and 4 . 89 in the DO-G13 and DO-G16 dams , respectively ) was significantly greater than the absolute number of either allele in the offspring of dams without distortion ( p = 0 . 006 and 0 . 049 for the R2d2WSB and R2d2NotWSB alleles in DO-G13; p = 0 . 005 and 4x10–4 for the R2d2WSB and R2d2NotWSB alleles in DO-G16; Fig . 5 A , B ) . The same result holds true for live embryos at mid-gestation: the average numbers of offspring that inherited R2d2WSB and R2d2NotWSB alleles were 5 . 0 ± 2 . 2 and 1 . 6 ± 1 . 8 for dams with TRD versus 4 . 3 ± 1 . 6 and 3 . 4 ± 1 . 8 for dams without TRD . Based on the consistent and significant excess average absolute number of R2d2WSB alleles in the litters of dams with TRD , we conclude again that meiotic drive is required to explain TRD at R2d2 . Further support for meiotic drive was provided by the analysis of the DO-G13–44 pedigree ( Fig . 5 C ) and crosses between ( NZO/HILtJxWSB/EiJ ) F1 dams and FVB/NJ sires ( cross 15 in Table 1; Fig . 5 D ) . The average litter size of DO-G13–44 G3 dams inheriting the mutant R2d2WSB allele ( R2d2WSBdel1 ) was larger than in dams inheriting the standard R2d2WSB allele ( 9 . 4 ± 2 . 9 and 6 . 8 ± 1 . 6 , respectively ) , but the observed average litter size in dams with TRD is significantly greater than predicted based on TR ( p = 0 . 02; S7 Fig . ) . Similarly , in the ( NZO/HILtJxWSB/EiJ ) F1 crosses the average litter size ( 7 . 7 ± 2 . 4; Fig . 5 D ) was comparable to DO-G13 and DO-G16 dams without TRD , and was greater than predicted based on TR ( p = 0 . 09; Fig . 5 ) . There was little direct evidence of embryonic lethality at mid-gestation ( 1 . 8 ± 1 . 6 and 0 . 4 ± 0 . 5 resorbed embryos , respectively; S8 Fig . ) . Furthermore , DO-G13–44 G3 dams with different R2d2 alleles differed significantly in the average absolute number of offspring per litter inheriting the R2d2WSB allele ( in dams with TRD ) compared to the R2d2WSBdel1 allele ( in dams with Mendelian segregation; 5 . 3 ± 2 . 0 and 4 . 64 ± 2 . 4 , respectively , p = 0 . 07; Fig . 5 C ) . Similar results are observed when comparing the absolute number of offspring per litter that inherited the R2d2WSB allele in the ( NZO/HILtJxWSB/EiJ ) F1 crosses to the DO dams without TRD ( 5 . 1 ± 1 . 0 and 4 . 1 ± 1 . 1 , respectively , p = 0 . 03; Fig . 5 D ) . In summary , all data from four independent experimental populations were consistent with an explanation of Chr 2 TRD that requires the joint presence meiotic drive and low-level embryonic lethality .
After demonstrating that TRD occurs only through the germline of F1 female mice , we were faced with two major obstacles in our efforts to map the causative locus . First , although heterozygosity for the WSB/EiJ allele is required , it is not sufficient for meiotic drive ( Table 1; S1 Table ) . Therefore , we initially mapped the responder by determining the minimum region of overlap for the WSB/EiJ haplotype only in dams with TRD ( Fig . 1 ) . This yielded a 9 . 3 Mb candidate interval . Second , the candidate interval spans a recombination-cold region [37 , 40 , 42] , and the frequency of recombination is three-fold lower than expected in the CC ( Fig . 2 D ) . Although this likely contributes to the overall deficit in recombinant chromosomes ( none observed versus an expected 23 in the 378 DO females and 4 in 61 CC lines ) , the complete lack of recombinants involving the WSB/EiJ haplotype is striking , and , for the purposes of this study , a major impediment to the precise mapping the responder . Within the candidate interval , a single variant ( R2d2 ) stands out as the most likely cause of TRD . R2d2 consists of one or more copies of a 127 kb sequence ( R2d ) . High copy number ( ≥ 24 ) is present in all four strains with reported TRD and low copy number ( ≤ 2 ) is present in all eight strains without TRD ( Fig . 2 A , C ) . The expansion in copy number leads to an increase of at least 3 Mb in DNA content within the allele favored by maternal TRD . Among CC founders , only WSB/EiJ has a high copy number allele . As the reference genome is based on a single classical inbred strain , C57BL/6J , copy number gains in other strains or wild mice may be located in a different physical location . Fortunately , the presence of a third allele in CAST/EiJ ( which exhibited a twofold enrichment of sequencing reads ) combined with the fact that recombinations involving the CAST/EiJ haplotype are not suppressed within the 9 . 3 Mb candidate interval , enabled us to map the physical location of R2d2 to a 900 kb region located 6 Mb distal to R2d1 , the locus where the sequencing reads mapped in the reference genome ( Fig . 3 ) . Importantly , the mapping of R2d2 was enabled by the availability of deep sequence data for each of the strains used in our experiments [25–27 , 37 , 42 , 43 and this study] and by combining the results of experiments completed 20 years apart [25–27 , 43 , 44] . We determined the number and spatial distribution of SNPs in the 9 . 3 Mb candidate interval that partition the ten inbred strains with whole genome sequence in a pattern consistent with the TRD phenotype ( three strains with TRD: WSB/EiJ , SPRET/EiJ and HR8; and seven strains without TRD: A/J , C57BL6/J , 129S1/SvImJ , NOD/ShiLtJ , NZO/HILtJ , CAST/EiJ and PWK/PhJ ) . Compared to a genome-wide mean of 1 consistent SNP every ~3 . 2 kb , within the 900 kb region where we mapped R2d2 there was a mean of 1 consistent SNP every 883 bp ( p < 1 . 0x10–4 , one-sided Student’s t-test; Fig . 2 E ) . This reduction in diversity is not due to undercalling of SNPs in the R2d2 candidate interval ( Fig . 2 F ) . The fact that consistent SNPs are rare in most of the genome but are common within the 900 Kb region in which R2d2 maps supports the hypothesis that R2d2 is the causative allele for TRD . Most importantly , we identified a DO female ( DO-G13–44 ) that was homozygous for the WSB/EiJ haplotype across the entire R2d candidate interval but was heterozygous for R2d2 alleles with different copy numbers ( Fig . 4 ) . We generated a three-generation pedigree and analyzed the R2d copy number , the Chr 2 haplotype and TR in the progeny of heterozygous dams with different copy numbers . This analysis revealed perfect correlations between the inheritance of R2d2WSBdel1 and complete absence of TRD in favor of the WSB/EiJ allele , and between the inheritance of R2d2WSB and presence of TRD . This experiment demonstrates that the reduction in copy number from 33 to 11 is sufficient to restore Mendelian segregation , and that R2d2 is the causative allele for maternal TRD . Further evidence that TRD requires an R2d2 allele with copy number of above 11 is provided by the NU/J inbred strain . This strain has intermediate copy number ( 7 , estimated by TaqMan ) but no TRD in the progeny of ( NU/JxC57BL/6J ) F1 female hybrids ( 0 . 55 , p = 0 . 55; S12 Fig . ) . The presence of R2d sequences at two distinct locations ( Fig . 2 G ) indicates an initial duplication of this segment in the ancestor of CAST/Eij , WSB/EiJ , SPRET/EiJ and Hsd:ICR . R2d spans a highly expressed protein coding gene ( Cwc22; Fig . 2 C ) that is implicated in RNA splicing [38 , 44] , a predicted gene of unknown function that overlaps with the last exon of Cwc22 ( Gm13727 ) and a pseudogene ( Gm13726 ) . DNA copy number variation for Cwc22 has been described previously [38 , 45] . Cwc22 is highly expressed in mouse oocytes and fertilized eggs [45 , 46] . The Cwc22 gene is a known eQTL in mouse: allele-specific RNA-seq of brain tissue from reciprocal crosses between WSB/EiJ , PWK/PhJ and CAST/EiJ showed extreme differential expression , with the WSB/EiJ allele more highly expressed than the other two [46 , 47] . Apart from its size and repetitive nature , an important feature of the R2d2 locus is its remarkable uniformity between three divergent genetic backgrounds that are separated by ~1 million years of evolution: WSB/EiJ , SPRET/EiJ and HR8 [47–49] . For example in WSB/EiJ and SPRET/EiJ the genome-wide mean is 1 SNP every ~60 bp [37] and the mean SNP frequency within R2d is significantly reduced to 1 SNP every 1 , 342 bp ( t-test , p = 3 . 9x10–58 ) . Further analysis will be required to determine the respective ages of the duplication and the copy number change ( s ) , and whether interspecific introgression [48–51] is required to explain the unlikely degree of sequence conservation between M . m . domesticus and M . spretus . We note that , while unlikely given the results of our QTL mapping ( Fig . 3 ) , it is possible that there have been additional duplication events that have also inserted R2d in other chromosomes . Additionally , the causal allele may incorporate additional DNA sequences , including some that may be absent in the reference genome ( similar to the origin of the sequence on maize chromosome Ab10 that causes meiotic drive in that species ) . If that is the case , the causal allele may be much larger than 4 . 4 Mb . For example , HSR alleles as large as 200 Mb have been described [50–52] . A second focus of our study was to discriminate among the many mechanisms [29 , 52] that could give rise to TRD at R2d2 , and to rule out as many as possible . First , the fact that TRD is only observed through the maternal germline rules out both spermatogenesis-mediated processes and sperm competition . Second , the presence of TRD at birth rules out differential survival of offspring . Third , the fact that distortion was independent of the maternal granddam precludes cytoplasmic effects . The remaining plausible explanations are differential fertilization based on the oocyte genotype , embryonic lethality and/or meiotic drive . The first two mechanisms should reduce the average litter size proportionally to TR ( black line in S7 Fig . ) , while the average absolute number of offspring inheriting the favored genotype ( R2d2WSB ) per litter remains constant . The number of resorbed embryos observed in pregnant females could distinguish the two mechanisms because it should be greater in the second than in the first scenario . In contrast , if meiotic drive is solely responsible for TRD then the following should be true: 1 ) average litter size is independent of TRD , 2 ) the average absolute number of offspring inheriting the favored genotype ( R2d2WSB ) per litter is higher in dams with TRD than in dams with Mendelian segregation , and 3 ) the level of embryonic lethality is independent of the presence and level of distortion . The data shown in the Results section are most consistent with the combined action of embryonic lethality and meiotic drive . Specifically , meiotic drive is required to explain both the fact that the observed average litter size in the DO-G13 and DO-G16 dams , in the DO-G13–44 pedigree and in the ( NZO/HILtJxWSB/EiJ ) F1 dams is greater than predicted based on TR ( S7 Fig . ) , and that the average absolute number of offspring inheriting the R2d2WSB genotype per litter is greater in dams with TRD ( Fig . 5 ) . Note that some p-values in comparisons involving ( NZO/HILtJxWSB/EiJ ) F1 crosses failed to reach statistical significance due to the small sample size , but the trends were always consistent with those in DO dams with TRD . An alternative explanation that does not involve meiotic drive would require the combined presence of increased ovulation in dams with TRD and pre- or post-implantation genotype-dependent competition between embryos favoring the allele with the high copy number at R2d2 . Genotyping at R2d2 and re-analysis of 159 F2 females from the M16ixL6 intercross [29] confirms an overdominant effect of the R2d2 genotype in the number of live and dead embryos at day 16 of gestation , as predicted under the meiotic drive and embryo competition scenarios , but shows no effect of the R2d2 locus on ovulation rates ( S9 Fig . ) . This result is not due to a lack of power , as we have 80% power ( at a = 0 . 5 ) to detect a difference in the mean ovulation rate du to an effect of the R2d2 genotype and QTLs for ovulation rate were identified in the original study [29 , 53–55] . In summary , the effect of the R2d2 genotype on reproductive phenotypes is most consistent with the meiotic drive hypothesis . However , the possibility remains that the genotype-associated difference in number of live embryos may be due to differential fertilization or implantation . Additional breeding experiments and genotyping of pre-implantation embryos will resolve the remaining questions concerning the mechanisms involved in TRD at R2d2 . It is interesting to speculate about the types of embryonic lethality that are consistent with our data and with previous reports of TRD on Chr 2 . Lethality is associated with distortion at R2d2 , and thus the simplest explanation is preferential death of embryos inheriting maternal R2d2NotWSB alleles . However , such a scenario would require parent-of-origin-dependent death of embryos with maternal C57BL/6J , 129S1/SvImJ , NOD/ShiLtJ and NZO/HILtJ R2d2 alleles in crosses involving F1 females ( Table 1 ) and CAST/EiJ , PWK/PhJ and A/J R2d2 alleles in the CC/DO females ( S10 Fig . ) . The lack of evidence of TRD and parent-of-origin lethality in dozens of crosses involving these alleles [53–55] , combined with the lack of evidence for imprinted genes in the central region of Chr 2 [2–4 , 46 , 56] , appears to rule out this explanation . Specifically , the Cwc22 gene present in R2d is not imprinted in brain , kidney , lung and liver in crosses involving the WSB/EiJ , PWK/PhJ and CAST/EiJ strains [46] . A more likely explanation for the joint and correlated presence of meiotic drive and lethality is that the unequal segregation of chromosomes and/or chromatids that leads to TRD in euploid embryos may also lead to increased Chr 2 aneuploidy , and thus to embryonic death ( all autosomal aneuploidy is embryonic-lethal in the mouse ) . This would also explain the slight increase in the number of resorbed embryos observed at mid-gestation ( S8 Fig . ; S1 Table ) . This hypothesis makes the testable prediction that Chr 2 should be especially affected by aneuploidy in some dams with TRD . Importantly , co-segregation of a deletion allele of R2d2 and increased litter size in the DO-G13–44 pedigree demonstrates that lethality is mediated by an element within the R2d repeat . Overall , we assessed TR at R2d2 in hundreds of females carrying a single WSB/EiJ allele in at least nine distinct genetic backgrounds ( Table 1; S1 Table ) . The presence of significantly different TR levels among F1 hybrid dams , combined with the fact that we observe both extreme TRD and no distortion in the progeny of females with A/J , C57BL/6J , 129S1/SvImJ , NOD/ShiLtJ , CAST/EiJ and PWK/PhJ alleles in trans at R2d2 ( S10 Fig . ) , demonstrates that TRD is under genetic control of at least one additional locus ( i . e . , there is at least one unlinked distorter locus that is genetically variable in the CC and DO mice ) . Furthermore , the presence of at least two significantly different levels of distortion among F1 hybrid dams ( Table 1; S3 Fig . ) indicates either that more than one distorter locus is involved or that an allelic series exists at a single distorter locus . Further evidence that TRD is under control of one or more unlinked distorters was provided by 15 female DO-G13–44 G1 offspring that inherited the high-copy allele . Those dams had significantly different levels of TRD ( p = 9 . 8x10–5 ) . Note that there was no correlation between the presence or level of TRD and the paternally inherited allele ( one-way ANOVA , F = 2 . 21 on 1 and 23 df , p = 0 . 15; Fig . 3 ) . In the DO-G13–44 pedigree , females that inherited the R2d2WSBdel1 allele had copy number 11 ( S11 Fig . ) , indicating a partial rather than complete deletion of the expansion . Using the TaqMan assay , we identified two additional DO females ( DO-G13–49 and DO-G16–107; S4 Fig . ; S11 Fig . ) that had results consistent with a copy number loss in the WSB/EiJ haplotype . The presence of the deletion in the respective germlines was confirmed by the TaqMan assay in their progenies ( S4 Fig . ) . Importantly , each one of the three deletions appears to be independent because these females are not closely related , their WSB/EiJ haplotypes in Chr 2 are different and the copy number present in each female is also different ( S12 Fig . ) . The deletions appear to be internal to R2d2 based on the analysis of the MegaMUGA genotypes and intensities [57] at all surrounding markers . The repeated observation of independent deletions indicates that R2d2 is rather unstable and may explain the fact that , despite its presence in laboratory strains and wild mice , it has not led ( yet ) to a complete selective sweep . Known meiotic drive systems ( S1 Fig . ) consist of one or more responder loci ( a locus subject to preferential segregation during meiosis ) and a single distorter ( the effector locus required for drive at the responder ) . In meiotic drive systems that are stable in natural populations , responder and distorter loci are tightly linked and are typically protected from decoupling by factors that inhibit recombination , such as structural variation [7 , 11 , 14] . Although R2d2 resides within a recombination-cold region , the distorter is not closely linked to R2d2 based on the TR observed and the diplotypes present in F1 hybrid and DO dams ( Fig . 1; S8 Fig . ) . Therefore , at least one unlinked distorter is required to explain the observed variability in TRD . These observations indicate that the maternal TRD phenotype has a complex genetic architecture . Specifically , a minimum number of copies of R2d are required in heterozygosity at R2d2 for TRD to be observed . Therefore , it can be classified as overdominant , restricted to the female germline and caused by structural variation . Similar characteristics have been recently reported for the Xce locus that controls X-inactivation choice; notably , characterization of Xce relied on the analysis of a genetically diverse set of F1 hybrid mice [58] . In addition , multiple alleles at unlinked loci interact to determine whether distortion occurs at R2d2 , and to what extent . This is unique among meiotic drive systems ( S1 Fig . ) and has important implications for the natural history of the system and for the ease of genetic dissection . We hypothesize that variation in TR levels at R2d2 results from the interaction of alleles originating from multiple taxa , and thus the use of inter-specifc and inter-subspecific mouse populations was key to the characterization of this system . Wild-derived strains and wild-caught mice have enabled important biological discoveries [4 , 59] , and we echo previous encouragements of a more prominent role for these resources in biological and biomedical research [60 , 61] . Centromeres ( i . e . , the site of kinetochore formation ) are remarkable loci that control , in cis , proper segregation of chromosomes during mitosis and meiosis . It is easy to envision how a responder at , or tightly linked to , a centromere can influence chromosome segregation . Recent evidence shows that kinetochore protein levels and microtubule binding are positively correlated with preferential segregation to the oocyte in mice that are heterozygous for Robertsonian fusions [62] , indicating that differences in centromere “strength” lead to meiotic drive . Responders located far away from centromeres are thought to influence their own segregation in cis by becoming “neocentromeres” and taking advantage of the inherited functional polarity of the female meiotic spindle [63] . We hypothesize that R2d2 may act as a neocentromere after epigenetic activation mediated by C57BL/6J , NZO/ShiLtJ , 129S1/SvImJ , and NOD/HILtJ alleles at the distorter ( s ) . The discovery of multiple R2d2 alleles with different copy numbers demonstrates that the presence of the distal insertion of R2d is not sufficient for meiotic drive; rather , some minimum copy number ( > 11 ) is required for TRD . This raises the possibility that meiotic drive at R2d2 is dosage-dependent , such that fine-scale control over the level of TRD is possible by adjusting the number of copies of R2d . If R2d2 is acting as a neocentromere , this may also indicate that some minimum size and/or number of repeats is required for recognition and activation by the epigenetic machinery . The Ab10 system of maize provides examples of responders that function as neocentromeres and for which the level of meiotic drive depends on the size of the responder ( i . e . , knob size ) [11] . The effect on the Chr 2 centromere of activating an ectopic neocentromere at R2d2 is unknown , but it might explain the moderate levels of lethality caused by aneuploidy and suggests that some coordination between the two loci is required to achieve chromosome segregation . Meiosis involving chromosomes with neocentromeres may lead to an increased rate of non-disjunction and a reduced rate of recombination . The conclusion that a genetically complex meiotic drive system is responsible for TRD favoring the WSB/EiJ allele at R2d2 is fully consistent with the initial observations of TRD in the CC , with our prediction that positive selection of the WSB/EiJ allele occurred during outcrossing or in early inbreeding generations [35] , with the presence of similar levels of TRD in extinct and extant CC lines at intermediate generations of the CC ( S5 Table ) and with the fact that C57BL/6J , 129S1/SvImJ , NOD/ShiLtJ and NZO/HILtJ haplotypes at R2d2 are not underrepresented among the currently completed CC strains ( http://csbio . unc . edu/CCstatus/index . py ) . The observed levels of TRD in crosses that use DO females are consistent with presence of different alleles at the distorter ( s ) ( S7 Fig . ; S1 Table ) . Although the discovery and identification of TRD that emerged from the DO pseudo-randomized mating scheme offered the opportunity to characterize a novel meiotic drive responder , the existence of such a locus could negatively impact the utility of this population for genetic studies . Fortunately , the locus was discovered before complete fixation of the R2d2WSB allele . Although the candidate interval spans 900 kb , TRD affects a much larger region in the DO because the strength of selection in favor of the WSB/EiJ allele is outpacing the rate at which recombination can degrade linkage disequilibrium in the region . Ultimately , this region would become an actual or statistical ‘blind-spot’ in the DO , such that the non-WSB/EiJ allele frequencies would become too small to detect allelic effects on phenotypic variation . Efforts are underway to purge the WSB/EiJ allele from the DO breeding population at this locus or to select for mice carrying a WSB/EiJ haplotype with a low copy number for R2d2 , rather than allow the region to become fixed . Using marker-assisted selection , progeny of heterozygous WSB/EiJ carrier crosses are excluded from subsequent generations . Allele frequencies and random segregation on all other chromosomes are being preserved ( EJC unpublished ) . The SPRET/EiJ and WSB/EiJ strains and the Hsd:ICR outbred stocks are among the most extensively characterized and utilized mouse populations . Resources involving those populations include whole-genome sequencing and genotyping [24 , 37] , development of linkage maps of the mouse [40 , 64 , 65] , creation of genetic reference populations [35 , 36 , 66] , experimental crosses to map a diverse collection of biomedical and evolutionary traits [33 , 48 , 53 , 61] and selection lines derived from Hsd:ICR ( such as M16i and HR ) that have been widely used for genetic analyses [30–32 , 67–69] . The potential for distorted allele frequencies in crosses involving those populations may affect the interpretation of results from a wide range of genetic , behavioral and physiological studies . The R2d2 system has attributes that make its genetic and mechanistic characterization a tractable problem . Identification of several distorters would allow assembling the pathway ( s ) responsible for centromere function and spindle polarity . This may open the way to explore at the molecular and mechanistic levels an evolutionary force ( meiotic drive ) thought to be responsible for karyotype evolution in mammals and in many other organisms [15] . With the advent of genome engineering tools such as CRISPR/Cas9 [70] , we also anticipate practical applications of a strong , modulable meiotic drive system with only modest levels of lethality . For example , meiotic drive could be used to increase the efficacy of gene drives for introducing new genes into experimental or natural populations [71] .
All animal work was performed according to one of the following protocols: 1 ) the Guide for the Care and Use of Laboratory Animals under approved IACUC animal use protocols within the AAALAC accredited program at the University of North Carolina at Chapel Hill ( Animal Welfare Assurance Number: A-3410–01 ) ; 2 ) the requirements of The Jackson Laboratory Animal Ethics Committees under approved protocol #JAX10001; 3 ) an animal protocol approved by the North Carolina State University Institutional Animal Care and Use Committee ( 09–0133-B ) ; or 4 ) an animal study protocol approved by the NCI Animal Care and Use Committee ( ASP# LCBG-013 ) . All animals were euthanized according to the regulations of the governing protocol . The G2:F1 population has been previously reported and was genotyped on the Mouse Diversity Array [72] ( MDA ) . A population of 96 ( FVB/NJx ( WSB/EiJxPWK/PhJ ) F1 ) G2 mice was previously reported and was genotyped on the MegaMUGA array [40 , 53] . DNAs from selected progeny from previously published ( C57BL/6JxSPRET/EiJ ) xC57BL/6J and ( A/JxSPRET/EiJ ) xA/J backcrosses [26 , 43] were regenotyped on the MegaMUGA array . The SPRET/EiJ strain designation had not yet been assigned to the inbred strain at the time the backcross was performed [26] . Finally , DNA from multiple samples from the ( M16ixL6 ) F2 intercrosses and from generations 4 and 10 of the ( HR8xC57BL/6J ) advanced intercross line [29 , 31 , 32] were genotyped at markers closely linked to R2d2 . Crosses 1–2 , 7–10 and 16–17 ( Table 1 ) . WSB/EiJ and C57BL/6J were used in reciprocal combinations . Male F1 hybrids were backcrossed to C57BL/6J to produce the progeny of crosses 1 and 2 . Female F1 hybrids were backcrossed to C57BL/6J to produce the progeny of crosses 16 and 17 . The progeny of crosses 7–10 was produced in a similar way to crosses 16 and 17 , except that female F1 of reciprocal matings of WSB/EiJ and CAST/EiJ were used for crosses 7 and 8 , and female F1 of reciprocal matings of WSB/EiJ and PWD/PhJ were used for crosses 9 and 10 . All breeding was done at the Jackson Laboratory ( Bar Harbor , ME ) . All other crosses . DO mice and standard mouse inbred strains ( 129S1/SvImJ , A/J , C57BL/6J , CAST/EiJ , FVB/NJ , NU/J , NOD/ShiLtJ , NZO/H1LtJ , PWK/PhJ and WSB/EiJ ) were obtained from The Jackson Laboratory ( Bar Harbor , ME ) . CC mice were obtained from the Systems Genetics Core Facility colony at UNC Chapel Hill [73] ( http://csbio . unc . edu/CCstatus/index . py ) . Those mice were used to generate the following number and types of hybrid mice: nine ( 129S1/SvImJxWSB/EiJ ) F1 females; two ( A/JxWSB/EiJ ) F1 females; seven ( NOD/ShiLtJxWSB/EiJ ) F1 females; six ( NZO/HILtJxWSB/EiJ ) F1 females; 10 ( CC042/GeniUncxCC001/Unc ) F1 females; three ( CC001/UncxCC039/Unc ) F1; nine ( DOxCC001/Unc ) F1 females , 13 ( DOxCC005/Tau Unc ) F1 females and five ( NU/JxC57BL/6J ) . F1 females were mated to FVB/NJ males and cages were surveyed three to five times per week . Litter sizes were recorded and pups were sacrificed at birth , and tissue was collected for DNA isolation . The same breeding schema was followed with 127 DO R2d heterozygous females used to determine the origin of maternal TRD . All breeding was done at UNC Chapel Hill ( Chapel Hill , NC ) . A single G13 DO female ( DO-G13–44 ) was mated to a male that was the result of an intercross between four CC lines ( CC013/GeniUnc , CC053/Unc , CC065/Unc and CC008Geni/Unc; Fig . 4 ) . G3 female progeny were weaned , single housed and mated to FVB/NJ males . Cages were surveyed three to five times per week . Litter sizes were recorded and G4 pups were sacrificed at birth , and tissue was collected for DNA isolation . TR was measured in G3 dams as described above . Each dam was classified as having TRD ( p < 0 . 05 for 1-df Χ2 test of null hypothesis TR = 0 . 5 ) or not having TRD ( p ≥ 0 . 05 ) . Both G2 parents and G3 dams were genotyped on MegaMUGA and phased haplotypes at R2d2 were inferred by manual inspection of haplotype reconstructions . In order to isolate the contribution of maternal and paternal alleles to TRD , MegaMUGA markers called as H in the G2 dam and homozygous in the G2 sire were retained for mapping , and presence of TRD was mapped as a binary phenotype using a logistic regression analog to the Haley-Knott method . The procedure was repeated using only markers called as H in the father of the G3 dams and homozygous in the mother . Significance thresholds for LOD scores were obtained by unrestricted permutation . Crosses 1–2 , 7–10 and 16–17 ( Table 1 ) . DNA was prepared from spleens of 21-day old mice . DNA extraction and SNP genotyping were carried out as described previously [74] . All other samples . DNA for PCR-based genotyping was performed on crude whole genomic DNA extracted by heating tissue in 100ul of 25mM NaOH/0 . 2mM EDTA at 95°C for 60 minutes followed by the addition of 100ul of 40mM Tris-HCl . The samples were then spun at 2000 rpm for 10 minutes and the supernatant collected for use as PCR template . All primers ( S6 Table ) used in this study were designed using PrimerQuest software ( https://www . idtdna . com/Primerquest ) . PCR reactions contained 1 . 5–2 mM MgCl2 , 0 . 2–0 . 25 mM dNTPs , 0 . 2–1 . 8 μM of each primer and 0 . 5–1 units of GoTaq polymerase ( Promega ) in a final volume of 10–50 μL . Cycling conditions were 95°C , 2 min , 35 cycles at 95° , 55° and 72°C for 30 sec each , with a final extension at 72°C , 7 min . PCR products were loaded into a 2% agarose gel and run at 200 V for 40–120 minutes ( depending on the marker ) . Genotypes were scored and recorded . DNA for MegaMUGA genotyping was isolated as described previously [40 , 53] . Briefly , ~2 mm of mouse tail ( 5 mg ) was harvested , flash-frozen on dry ice and digested with proteinase K overnight at 65°C . The following day , DNA was extracted using the QIAGEN Puregene Gentra kit ( kit no . 158389; QIAGEN GmbH , Hilden Germany ) . Genotyping was performed with the MegaMUGA genotyping microarray ( Neogen/GeneSeek , Lincoln , NE ) , a 78 , 000-probe array based on the Illumina Infinium platform . Genotyping by TaqMan . After R2d2 was established as the causal variant for TRD , a subset of DO-G16 progeny and all ( M16i x L6 ) F2 intercross progeny were genotyped using TaqMan real-time PCR assays for Cwc22 . Samples heterozygous for a high-copy allele at R2d2 can be readily distinguished from samples homozygous for a low-copy allele based on the normalized cycle threshold value estimated from the assay ( see section “Copy-number validation” below ) . Deviation from Mendelian transmission . TR is reported as the ratio of the WSB/EiJ genotype to the total number of genotypes: WSB / ( WSB + nonWSB ) . P values for aggregate data were calculated using a Χ2 goodness-of-fit test of the observed number of WSB/EiJ genotypes compared to the number of WSB/EiJ genotypes expected under the null hypothesis of equal transmission: X2= ( WSB-WSB+nonWSB2 ) 2WSB+nonWSB2 For individual dams , the small sample sizes ( typically fewer than 50 total offspring ) would lead to type II error; therefore , p-values were calculated using an exact binomial test . Confidence intervals for TRs were calculated using the binom R package ( http://cran . r-project . org/web/packages/binom/ ) . Average litter size . Average litter size was calculated as the mean number of offspring counted soon after birth per litter per dam ( ± standard deviation ) , including the number of viable embryos counted in utero in mid-gestation DO dams ( unless otherwise noted ) . The expected average litter size ( ALS ) of a dam under a model in which lethality is the sole explanation for TRD is: ALSObs=ALSExp ( 1-2TR-12TR ) , where ALSExp is the mean ALS in dams with no TRD [41] . Significance of the deviation of ALSObs from ALSExp was determined using a Wilcox signed rank test . Inheritance of R2d2 alleles . Similarly , the average absolute number of offspring inheriting each R2d2 allele was calculated as the mean number of offspring per litter per dam having each of the possible genotypes . Significance was determined using a one-tailed Student t-test . DO and F1 dams were euthanized by CO2 asphyxiation 12–18 days after delivery of the previous litter and the uterus was dissected . The number of live embryos and reabsorbed ( dead ) embryos was recorded . Each live embryo was dissected to isolate DNA for genotyping . Tissue from each live embryo was harvested for DNA extraction and genotyping . All MDA arrays were genotyped using MouseDivGeno [57] , and all MegaMUGA arrays were genotyped using Illumina BeadStudio . We plotted number of H and N calls ( as a fraction of the total number of genotypes ) for each group of similar samples and excluded outliers from further analysis . For CC lines , DO animals , CCxCC F1 females and DOxCC F1 females , we inferred haplotypes using probabilistic methods [40 , 75] . As an additional QC step , we grouped DO samples by generation and plotted the number of recombinations ( counted as unique transitions in haplotype reconstructions ) and removed outliers . CAST/EiJ allele in the CC G2:F1 . Thirty-four MDA SNP probe sets were identified within R2d in the GRCm38 reference sequence ( S3 Table ) . We ensured that these probes were unique using BLAT [76] to map them to the reference genome . In order to map the expansion allele present in the CAST/EiJ strain , phenotypes and genotypes were coded as follows . First , we applied a CCS transform [77] to the mean intensity of all probes in each probe set using MouseDivGeno [57] and summed the values for each sample to obtain the final phenotype value . Next , the genome was divided into a set of disjoint intervals whose boundaries were defined by the 21 , 933 unique recombination events inferred in the population [40] , so that no individual would be recombinant within any of the resulting intervals . Then , using haplotype reconstructions , individuals were coded as either heterozygous ( CAST/not-CAST ) or homozygous ( not-CAST/not-CAST ) within each interval ( there are no CAST homozygous individuals in this population ) . Of 474 individuals , 144 with a WSB/EiJ allele in the middle of chromosome 2 were excluded to yield a final sample size of 330 . A single-locus QTL scan was then performed via Haley-Knott regression [78] , treating the population as a backcross . WSB/EiJ allele in an intercross population . Three MegaMUGA SNP probes were identified within R2d in the GRCm38 reference ( S4 Table ) . Again , uniqueness was verified using BLAT . In order to map the expansion allele in WSB/EiJ , the sum intensity of these probes was used as a phenotype and genotypes were coded as follows . First the genome was divided into a grid of 1 , 000 disjoint intervals of approximately equal size , and one MegaMUGA SNP marker segregating between WSB/EiJ and PWK/PhJ was selected per interval . Individuals were coded as heterozygous ( WSB/not-WSB ) or homozygous ( not-WSB/not-WSB ) at each marker . A single-locus QTL scan was then performed using Haley-Knott regression as implemented in R/qtl [79] , treating the population as a backcross . In order to refine the location of R2d2 , we identified individual mice with recombinant chromosomes within the candidate interval defined by linkage mapping . These critical recombinants define the proximal and distal boundaries of the refined candidate interval . CAST/EiJ allele . We partitioned the 330 G2:F1 individuals without a WSB/EiJ allele in the R2d locus into two groups according to MDA sum-intensity values . From those with sum-intensity consistent with a non-CAST/EiJ expansion allele , we selected the most distal recombinants from CAST/EiJ to another haplotype . From those with sum-intensity consistent with the CAST/EiJ expansion allele , we selected the most distal recombinant from another haplotype to CAST/EiJ . Together these recombinants define the proximal boundary of the candidate interval in CAST/EiJ . Similarly , in order to define the distal boundary of the candidate interval , we selected the most proximal recombinants from CAST/EiJ to another haplotype that still had sum-intensity consistent with the CAST/EiJ expansion allele . WSB/EiJ allele . The boundaries of the WSB/EiJ candidate interval were mapped in the same fashion using 229 individuals spanning generations 10 through 14 of the DO , all of which have been genotyped on MegaMUGA and are recombinant for WSB/EiJ in the initial candidate interval . We first excluded individuals homozygous for WSB/EiJ over any interval with in the interval . Then we selected the most distal recombinants from another haplotype to WSB/EiJ , which also had MegaMUGA sum-intensity values consistent with a non-WSB/EiJ expansion allele . These recombinants define the distal boundary of the candidate interval . We mapped the proximal boundary similarly . SPRET/EiJ allele . ( C57BL/6JxSPRET/EiJ ) xC57BL/6J ( n = 12 ) and ( A/JxSPRET/EiJ ) xA/J progeny ( n = 17 ) [26 , 43] genotyped on the MegaMUGA array were used to refine the candidate interval for the expansion allele in SPRET/EiJ . Haplotypes in the relevant region of Chr 2 were inferred by manual inspection of genotype calls . Samples were partitioned according to sum-intensity at the three MegaMUGA SNP probes tracking the expansion allele . Among individuals with sum-intensity consistent with the expansion allele , the most proximal recombinant from SPRET/EiJ to another haplotype defines the distal boundary of the candidate interval . Likewise the most distal recombinant from a non-SPRET/EiJ haplotype to SPRET/EiJ defines the proximal boundary of the candidate interval . Ten individuals from the HR8 selection line were selected for whole-genome sequencing . Five micrograms of high-molecular-weight DNA were used to construct TruSeq Illumina libraries , using 0 . 5 μg starting material , with 300- to 400- and 400- to 500-bp fragment sizes . Each library was sequenced on one lane of an Illumina HiSeq2000 flowcell , as paired-end reads , with 100-bp read lengths . We aligned the sequences to the University of California at Santa Cruz Mouse Build mm9 . HR8 sequenced reads were aligned to the mouse genome ( mm9 ) using bowtie 2 . 2 . 3 [80] with default options . We removed PCR duplicates and filtered low-quality SNPs using samtools 0 . 1 . 19 [81] and Picard 1 . 88 ( http://picard . sourceforge . net/ ) . We retrieved BAM files of aligned reads ( Oct 2012 release ) from the Sanger Mouse Genomes Project FTP site ( ftp-mouse . sanger . ac . uk ) . We used the mpileup function of samtools [81] to call sequence variants on the HR8 and Sanger BAM files jointly and to output the read depth at each base . We counted a SNP as private to WSB/EiJ , SPRET/EiJ and the 10 HR8 individuals if those samples all shared a genotype that was different from the seven other CC founder strains . We defined the boundaries of the copy number expansion by identifying consecutive 100bp windows in which the average read depth was at least twice the genome-wide average read depth . We estimated the number of copies of the expansion as the modal per-base read depth . We used commercially-available TaqMan assays for Cwc22 to estimate the copy number of R2d2 . We used two copy number assays ( Life Technologies catalog numbers Mm00644079_cn , Mm00053048_cn ) to target the number of Cwc22 copies ( proximal and distal ) . We also used two reference assays ( Tfrc , cat . no . 4458366 , for target Mm00053048_cn; Tert , cat . no . 4458368 , for target Mm00644079_cn ) , for genes known to exist in a single haploid copy in the mouse , to calibrate the amplification curve . Assays were performed according to the manufacturer’s protocol on an ABI StepOne Plus Real-Time PCR System ( Life Technologies , Carlsbad , CA ) . Cycle thresholds ( Ct ) for each assay were determined using the ABI CopyCaller v2 . 0 software with default settings . For each target-reference pair , relative cycle threshold ( ΔCt ) was calculated as The ΔCt value is proportional to copy-number of the target gene on the log scale but is subject to batch effects . In order to account such effects , normalized ΔCt values for each sample were calculated as follows . A standard set of control samples ( from C57BL/6J , WSB/EiJ , CAST/EiJ and ( WSB/EiJxC57BL/6J ) F1 mice ) , spanning the expected copy-number range for Cwc22 , were included in duplicate or triplicate in every assay batch . A linear mixed model was fit to raw ΔCt values for these control samples , with target-reference pair and batch as random effects , using the lme4 package ( http://lme4 . r-forge . r-project . org/ ) for R ( http://www . R-project . org/ ) . Predicted values ( best linear unbiased predictors , BLUPs ) from this model capture technical variation orthogonal to variation due to genotype . BLUPs calculated from control samples were subtracted from raw ΔCt values for all samples , and the residual was used as the normalized ΔCt for copy-number estimation . In this manuscript we chose in most cases to present ΔCt , rather than extrapolated absolute copy number , because ΔCt is the natural scale of the data ( i . e . , the log scale ) . Constant variance ( with respect to mean ) on the log scale grows exponentially on the linear scale so that estimates of absolute copy number become increasingly uncertain as copy-number grows . The use of TaqMan assays for Cwc22 as a proxy for copy number at R2d2 was validated by mapping normalized ΔCt for target Mm00644079_cn as a quantitative phenotype in 64 members of the ( FVB/NJx ( WSB/EiJxPWK/PhJ ) F1 ) G2 intercross population described above . The marker selection and mapping procedure were the same as described above for mapping MegaMUGA sum-intensity values . Chr 2 genotypes and whole-genome sequence that have not been published elsewhere are available at http://csbio . unc . edu/r2d2 . | One of the strongest expectations in genetics is that chromosomes segregate randomly during meiosis . However , genetic loci that exhibit transmission ratio distortion ( TRD ) are sometimes observed in offspring of F1 hybrids . Meiotic drive is a type of non-Mendelian inheritance in which a “selfish” genetic element exploits asymmetric female meiotic cell division to promote its preferential inclusion in ova . We previously reported TRD on Chr 2 in the CC , a mouse recombinant inbred panel with contributions from three Mus musculus subspecies . Here we show that maternal TRD consistent with a novel meiotic drive system is caused by a copy number gain . This mutation is similar in size and structure to other known meiotic drive responders , such as the knobs of maize . A deletion of most of the copies is sufficient to restore Mendelian segregation , proving that the copy number variant is causative of the observed TRD . In the CC , and also the related DO population , the transmission frequency of the favored allele varies dependent on genetic background , demonstrating that this system is under genetic control . In conclusion , we describe a novel wild-derived meiotic drive locus on mouse Chr 2 that exploits female meiosis asymmetry to violate the Laws of Mendelian inheritance . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | A Multi-Megabase Copy Number Gain Causes Maternal Transmission Ratio Distortion on Mouse Chromosome 2 |
The human membrane cofactor protein ( MCP , CD46 ) is a central component of the innate immune system . CD46 protects autologous cells from complement attack by binding to complement proteins C3b and C4b and serving as a cofactor for their cleavage . Recent data show that CD46 also plays a role in mediating acquired immune responses , and in triggering autophagy . In addition to these physiologic functions , a significant number of pathogens , including select adenoviruses , measles virus , human herpes virus 6 ( HHV-6 ) , Streptococci , and Neisseria , use CD46 as a cell attachment receptor . We have determined the crystal structure of the extracellular region of CD46 in complex with the human adenovirus type 11 fiber knob . Extracellular CD46 comprises four short consensus repeats ( SCR1-SCR4 ) that form an elongated structure resembling a hockey stick , with a long shaft and a short blade . Domains SCR1 , SCR2 and SCR3 are arranged in a nearly linear fashion . Unexpectedly , however , the structure reveals a profound bend between domains SCR3 and SCR4 , which has implications for the interactions with ligands as well as the orientation of the protein at the cell surface . This bend can be attributed to an insertion of five hydrophobic residues in a SCR3 surface loop . Residues in this loop have been implicated in interactions with complement , indicating that the bend participates in binding to C3b and C4b . The structure provides an accurate framework for mapping all known ligand binding sites onto the surface of CD46 , thereby advancing an understanding of how CD46 acts as a receptor for pathogens and physiologic ligands of the immune system .
The human CD46 receptor , also known as membrane cofactor protein ( MCP ) , is present on all nucleated cells [1] . It belongs to a family of proteins known as the regulators of complement activation ( RCA ) , which cluster on chromosome 1q32 [2] , [3] . In addition to CD46 , the RCA family includes decay-accelerating factor ( CD55/DAF ) , complement receptors 1 ( CR1/CD35 ) and 2 ( CR2/CD21 ) , the C4-binding protein , and factor H ( FH ) . CD46 acts as a key regulator in the classical and alternative complement activation cascades of the innate immune system by serving as a cofactor for the factor I - mediated cleavage of C3b and C4b [4] . This process protects host cells from inadvertent lysis by the complement system [3] . The relevance of CD46 has expanded beyond the innate immune system in recent years as it has become clear that CD46 can regulate T-cell immunity , and is in particular able to control inflammation [5] . Consequently , reproductive processes , multiple sclerosis , and inflammatory responses in the brain have all been functionally linked to CD46 [5] , [6] , [7] , [8] . In addition to its role in complement activation and regulation of the adaptive immune response , CD46 is used as a cellular receptor by several viruses and bacteria . Some measles virus ( MV ) [9] , [10] and adenovirus ( Adv ) [11] , [12] , [13] strains attach to cells by engaging CD46 . In addition , group A Streptococci [14] , [15] , some Neisseria strains [16] , [17] and human herpes virus 6 ( HHV6 ) [18] , [19] all use CD46 as a receptor . While other members of the RCA-cluster are also targeted by viruses [20] , [21] , the number of pathogens that attach to cells by using CD46 remains unsurpassed . This has led to the description of CD46 as a “pathogen's magnet” [22] . The prominence of CD46 in pathogen interactions may be attributed , at least in part , to the protein's ubiquitous expression in the host . In some cases , interactions with pathogens have also been shown to down-regulate cellular levels of CD46 , thereby increasing complement sensitivity of infected cells [23] , [24] , [25] . A recent study provides evidence for a direct link between CD46 and components of the autophagy machinery [26] . Recognition of pathogens by CD46 is thought to trigger autophagy , which serves as a critical step to control infection . However , some pathogens are known to exploit autophagy in host cells . Common to all the proteins expressed from the RCA cluster is their modular construction , which is primarily based on concatenated short consensus repeats ( SCR ) [3] . Each SCR module contains about 60 amino acids that fold into a compact β-barrel domain surrounded by flexible loops [27] . While the modules display high sequence variability , they all contain four conserved cysteine residues that form two disulfide bridges at the top and bottom of the repeat . The number of repeats present in the members of the RCA family ranges from four in CD55 and CD46 to 30 in CD35 . Many structures of fragments of RCA family members are known , and they exhibit significant diversity both in their loop structures and also in their interdomain orientation [28] , [29] . The four SCRs in CD46 constitute the bulk of its extracellular region . The repeats are connected to a short linker region rich in serines , threonines and prolines ( STP region ) , a single membrane-spanning segment , and a cytoplasmic tail . Alternative splicing generates multiple isoforms of CD46 that all have identical N-terminal repeats but exhibit variation in the STP region and the cytoplasmic tail [30] . The crystal structure of the N-terminal two repeats , SCR1 and SCR2 , of CD46 ( CD46-2D ) revealed essential features of this region , including a pronounced bend between the two repeats and significant flexibility at the interdomain interface [31] . Although CD46-2D is heavily glycosylated , one side of the two-domain fragment was found to be entirely devoid of glycans . Subsequent crystal structures of CD46-2D in complex with the Adv fiber knob [32] , [33] and with the MV hemagglutinin [34] demonstrated that both viral attachment proteins bind to this glycan-free surface . In both cases , engagement by the virus leads to “straightening” of the CD46-2D protein into a linear conformation . Furthermore , both viral attachment proteins form contacts with CD46-2D that predominantly involve residues at the SCR1-SCR2 interface . The implications of the structural rearrangement of CD46 upon ligand binding are not understood . Structural information about the binding of complement proteins C3b and C4b to CD46 is not available . However , biochemical mapping studies strongly implicate domains SCR2 , SCR3 and SCR4 in this interaction , with most of the predicted contacts located on SCR3 and SCR4 [35] , [36] , [37] . Notably , the regions of CD46 that are thought to interact with C3b and C4b overlap but are not identical [37] . As the cellular C3b and C4b proteins as well as HHV6 engage regions that include the SCR3 and/or SCR4 domains , modeling studies have aimed to predict the structure of unknown portions of CD46 in order to provide a basis for the mapping of binding epitopes [37] , [38] . Although some features of the SCR domains are conserved and can be predicted with reasonable accuracy , loop regions and interdomain orientations are notoriously difficult to model . These latter features are however central components of the protein and , to a large extent , determine its overall conformation and interaction properties . In order to advance an understanding of how CD46 interacts with its many ligands , we determined the three dimensional structure of an extracellular segment of CD46 that comprises all four SCR domains ( CD46-4D ) . The structure provides a basis for identifying binding sites for several CD46 ligands that bind to the C-terminal region of the protein . It also reveals an unexpected kink between domains SCR3 and SCR4 , which has profound implications for the conformation of CD46 on the cell surface , and for the recognition of its ligands .
Glycosylation of CD46 plays an important role in mediating its interactions , at least with some proteins [39] . Proper glycosylation probably also helps to stabilize the overall conformation of the CD46-2D fragment [31] . In order to preserve the glycosylation of CD46-4D , we produced the protein in a mammalian cell line ( see Methods ) . However , efforts to determine the crystal structure of unliganded CD46-4D were unsuccessful , perhaps due to the heavy glycosylation and the known flexibility between domains SCR1 and SCR2 [31] . Although several crystal forms could be obtained , none of these diffracted beyond 15 Å ( M . Larvie and T . Stehle , unpublished results ) . The Ad11 knob , which can easily be crystallized in its unbound form and engages in a high-affinity complex with the SCR1 and SCR2 domains of CD46-2D [32] , was then used to form a complex with CD46-4D for crystallization . This strategy produced crystals that diffracted to 2 . 84 Å resolution , allowing us to trace the polypeptide chains for the entire complex ( Table 1 and Methods ) . The Ad11 knob is a trimeric complex composed of three protomers . The asymmetric unit of the crystals contains two Ad11 knob protomers that are located in different trimers , and are each complexed with a single CD46-4D molecule . For each protomer , crystallographic three-fold rotation axes in the P63 space group then generate a trimeric knob structure ligated with three CD46-4D molecules ( Fig . 1 ) . At the center of the complex lies the trimeric Ad11 knob structure , which , in support of previous findings [32] , [40] , engages domains SCR1 and SCR2 but does not interact with domains SCR3 and SCR4 of CD46 ( Fig . 1 ) . The SCR1-SCR2 segment adopts a rod-like conformation that is similar but not identical to the one seen in the earlier crystal structure of Ad11 knob in complex with CD46-2D [32] ( Fig . 2A ) . The SCR1 domain and the SCR1-SCR2 interface make nearly identical contacts with the Ad11 knob in both structures , including the central salt bridge between CD46 residue Glu63 and Ad11 knob residue Arg280 ( Figs . 2B , C ) . However , the position and orientation of SCR2 is quite different in the two complexes ( Figs . 2A , D ) . In the Ad11 knob - CD46-2D complex , the SCR2 domain rests on the IJ loop of a second Ad11 knob protomer , forming several contacts including two hydrogen bonds , with the knob . By contrast , SCR2 has moved away from the knob in the Ad11 knob - CD46-4D complex , and the number of contacts have been reduced significantly . As the SCR2 domains are involved in different crystal contacts in the CD46-2D and CD46-4D complexes , we conclude that the interactions of this domain with the Ad11 knob are at least partially determined by crystal packing effects and not crucial for binding . Our data therefore suggest that SCR2 merely rests above the Ad11 knob but does not engage in critical interactions , in line with mutational studies that show that contacts between Ad11 knob and the base of SCR1 are most critical for contact formation [40] , [41] . We also conclude that CD46 retains some flexibility at its SCR1-SCR2 interface even when bound to the Ad11 knob . The CD46-4D chain folds into an elongated structure that is about 115 Å long ( Fig . 1 ) . Domains SCR1 , SCR2 and SCR3 are arranged in nearly linear fashion , with interdomain angles of 148 and 149 degrees , respectively . However , with an interdomain angle of only 120 degrees between domains SCR3 and SCR4 , the SCR4 domain deviates profoundly from the long axis of the protein ( Fig . 1 ) . The overall structure of CD46-4D can therefore best be described as resembling a hockey stick , with the N-terminal three domains forming the “shaft” and SCR4 forming the “blade” ( Fig . 1 ) . The observed conformation of CD46-4D is nearly identical in both copies of the protein , despite differing crystal contacts . Sequence analysis predicts that CD46-4D carries three N-linked glycans ( at Asn49 in SCR1 , Asn80 in SCR2 , and Asn239 in SCR4 ) . Structures of CD46-2D had shown that Asn49 and Asn80 are glycosylated [31] , [32] , [33] , [34] . In accordance with this , we observe clear electron density for single N-acetyl glucosamine ( NAG ) residues at both positions , allowing us to incorporate these moieties into the model . Although the electron density at Asn239 is not clear enough to accurately model a carbohydrate into it , its shape and location strongly suggests the presence of a NAG . Thus , all three potential N-linked glycosylation sites of CD46-4D are utilized . Modeling a physiologic glycan structure onto the protein shows that all three glycans would face into the same direction , and would likely shield the concave “inner” side of CD46 entirely from interactions ( Fig . 3A and Methods ) . The STP region of CD46 comprises about 30 amino acids that are not included in our structure . These residues feature sites of O-linked glycosylation and likely serve as a spacer between the base of SCR4 and the membrane . To date , no structural information about this region is available . The prototypical SCR module is primarily composed of four longer β-strands ( B , C , D and E ) that form a barrel-like structure . The barrel is augmented with a set of smaller β-strands ( A , B' , D' and E' ) ( Fig . 3B ) , although not all strands are always present in an SCR . Structural features of SCR1 and SCR2 of CD46 , including the domain interface , have been described previously [31] , [32] . As expected , the overall folds of the SCR3 and SCR4 modules are quite similar to those of other SCRs such as SCR1 ( Figs . 3C–E ) . The two domains can be superimposed onto SCR1 with low r . m . s . deviations ( ranging from 2 . 2 to 2 . 8 Å ) , resulting in nearly identical locations of key features such as the conserved tryptophan side chains and the disulfide bonds that are hallmarks of each SCR ( Figs . 3B–E ) . We note that SCR3 carries a long , almost entirely hydrophobic insertion in its CD' loop ( connecting β-strands C and D' ) , causing this loop to protrude markedly from the domain ( Fig . 3D ) . The interdomain interfaces determine the overall conformation of the protein . Domains SCR2 and SCR3 are stacked together head-to-tail , producing a nearly linear two-domain fragment with interdomain contacts mostly involving the CD' loop of SCR2 and the B'C and DE loops of SCR3 ( Fig . 4A ) . Lys125 makes contacts primarily with SCR2 residues , and Val126 is involved in interactions with SCR3 . The interface is stabilized by a hydrogen bond between SCR3 residue Asp178 and the main chain nitrogen of Gly96 in SCR2 , and by non-polar contacts between Gly96 and the Tyr149 side chain in SCR3 . The interface buries an area of about 480 Å2 from solvent , which is comparable to the area buried between domains SCR1 and SCR2 in unliganded CD46 ( 340 Å2 ) [31] . The interface between SCR3 and SCR4 ( Fig . 4B ) is unique among the three CD46 interdomain interfaces as it has by far the largest buried surface area ( 729 Å2 ) and features a profound kink . These characteristics can be directly attributed to the protruding , hydrophobic CD' loop at the base of the SCR3 domain ( Fig . 3B , D ) . Since this loop contains four proline residues , we term it the “proline-rich loop” . The interface is generated by two tyrosines , Tyr213 and Tyr214 at the top of SCR4 , that form a cradle-like platform on which the proline-rich loop of SCR3 rests . There are numerous contacts between residues in the proline-rich loop and hydrophobic portions of the two tyrosine side chains as well as SCR4 residue Lys193 . The only polar residue in the proline-rich loop , Asp164 , lies close to two lysine residues in SCR4 , Lys193 and Lys211 , and forms weak charge-charge interactions with both . The conformation of the proline-rich loop is incompatible with a more linear arrangement of the SCR3 and SCR4 modules , and since it mediates a large number of interdomain contacts we conclude that this loop is responsible for the profound kink between these two domains . Its unusual length , proline-rich sequence , and key role in interdomain contacts suggest an important function , perhaps by serving as a contact point for complement proteins [37] or by helping to orient the CD46-4D protein at the cell surface ( see Discussion ) . The crystal structure of C3b in complex with the N-terminal 4 repeats of FH has been reported recently [42] . As C3b serves as a ligand for both CD46 and FH , a comparison of the CD46-4D and FH structures offers useful insights into the location of contact surfaces and overall conformations of proteins constructed from SCR domains . In the C3b-FH complex , domains SCR2 , SCR3 and SCR4 of FH engage a large surface that spans the entire side of C3b [42] ( Fig . 5 ) . Interestingly , the FH structure also revealed a kink between domains SCR3 and SCR4 at a region that mediates contacts with C3b . With an r . m . s . deviation of 1 . 43 Å ( 60 residue pairs ) , the SCR3 domains of FH and CD46-4D superimpose well . However , this superposition clearly shows that the overall conformations of the four domain segments of FH and CD46 are rather different . The CD46-4D structure is significantly more bent both at the SCR2-SCR3 and SCR3-SCR4 interfaces . It is not known whether the SCR3-SCR4 region is also bent in unliganded FH , or whether the observed bend is caused by contacts with C3b . However , the bend at the SCR3-SCR4 interface of CD46 clearly exists in the absence of ligand and is stabilized by an elongated CD' loop that is unique to the SCR3 domain of CD46 ( Fig . 5 ) . As discussed below , the preformed bent CD46 conformation could facilitate binding to C3b . Information on C3b and C4b binding to CD46 is primarily based on epitope mapping and mutagenesis experiments , as well as the analysis of molecules lacking specific SCR domains [35] , [36] , [37] . Taken together , these data indicate ( i ) that SCR1 is not required for binding C3b or C4b , ( ii ) that both complement proteins interact with a large portion of the remaining CD46 structure , and ( iii ) that the binding sites for C3b and C4b are overlapping but distinct . We have mapped all sites that were previously identified as important for binding to C3b and C4b ( see Figure 7 in reference [37] ) onto the protein surface , excluding amino acids that play a role in function but not direct binding . Intriguingly , the sites for the natural ligands C3b and C4b mostly involve the glycan-free aspects of CD46 and cluster in several smaller areas on SCR2 and SCR3 as well as a large region of SCR4 , near the SCR3-SCR4 interface ( Fig . 6A ) . Thus , as was seen in the interactions of CD46 with Adv and MV [32] , [33] , [34] , complement binding appears to be limited to glycan-free regions of CD46 . The CD46 sequence contains three unique regions that are rich in proline residues and that were predicted earlier to interact with C3b/C4b: residues 127-LCTPPPKI-135 at the SCR2-SCR3 interface , residues 159-PAPGPDP-165 in SCR3 , and 243-DPPVPKCL-250 in SCR4 [37] . All three regions are partially surface-exposed and available for interactions . The second sequence is especially intriguing as part of it corresponds to the unique insertion in the CD' loop of SCR3 ( Fig . 3B , D ) . This loop is an integral part of the bent SCR3-SCR4 interface ( Fig . 4B ) , and it may therefore play a central role both in determining the overall conformation of CD46 and in mediating interactions with C3b and C4b . Few amino acid mutations affected binding of C4b to CD46 , and not cofactor activity [37] . Amino acids Asn94 , Leu95 and Gly96 were found to be relevant only for interactions with C4b , and not C3b . These residues are located within the CD' loop at the base of SCR2 , near the SCR2-SCR3 interface , and Gly96 does in fact participate in contacts with SCR3 ( Fig . 4A ) . Thus C4b appears to engage a region closer to the N-terminus of CD46 , while also making contact with SCR4 residues . The extensive C3b-binding epitope covering a large area on SCR4 ( Fig . 6A ) partially overlaps with a positively-charged region involving a large number of lysine and arginine residues that all lie on one side of SCR4 or near the SCR3-SCR4 interface ( Lys190 , Lys193 , Arg195 , Lys203 , Lys210 , Lys211 , Lys224 , and Lys251 ) . It is conceivable that some of the basic residues towards the base of SCR4 that are not implicated in C3b binding ( e . g . , Lys224 , Lys251 ) mediate interactions with negatively-charged membrane lipids . Binding sites of Adv and MVH on CD46 have been well characterized by cocrystallization of complexes [32] , [33] , [34] . Both viruses bind to a similar region of CD46 , but they do so by making distinct contacts , with different amino acids . In each case , contacts are limited to SCR1 and SCR2 , and they are thus spatially separated from the C3b and C4b binding sites , which do not involve SCR1 at all and are located near the base of the CD46-4D protein ( Fig . 6B ) . Given the large size of the complement proteins , it is nevertheless likely that interaction with either viral protein will directly compete with complement binding . CD46 also serves as a receptor for Streptococcus on keratinocytes [14] . Interactions are mediated by the streptococcal surface protein , M , a long , filamentous protein that is also able to engage other members of the RCA family . Using domain exchange experiments and chimeric CD46/CD55 molecules , Giannakis et al . [38] showed that binding of the M protein is dependent only on domains SCR3 and SCR4 of CD46 . Sequence comparison of CD46 with other RCA family members for which M protein binding has been mapped to individual residues [43] , [44] suggests that M protein primarily interacts with a region of SCR4 that partially overlaps with binding sites for C3b and C4b ( compare Fig . 6B with Fig . 6A ) . However , C3b-mediated complement activity was detectable even after addition of M protein [38] , indicating that the binding sites for C3b and M protein are not identical . The binding sites of Neisseria and HHV-6 have been mapped to individual domains only . The SCR3 and STP domains of CD46 are required to mediate adherence of Neisseria [17] , while interactions of HHV-6 with CD46 depend on repeats SCR2 and SCR3 [19] . In both cases , therefore , interactions appear to be distant from the binding sites for Adv and MV , and they are also expected to compete with the binding of C3b or C4b to CD46 .
Precise regulation of immune defense mechanisms is essential to protect host tissue from injury . This is achieved in part by mechanisms that prevent the inappropriate activation of complement on autologous tissues . The RCA family of proteins plays a key role in this process by interacting with fragments of complement proteins C3 and C4 . The CD46 protein inhibits complement activation by binding separately to C3b and C4b and promoting their proteolytic inactivation by factor I [4] . In addition , CD46 also serves as the cell attachment receptor for a number of human pathogens [22] . We have determined the three-dimensional structure of all four SCR domains of CD46 , which constitutes the bulk of the extracellular region of this cell surface receptor protein , in complex with the Ad11 knob . The conformation of CD46-4D resembles a hockey stick , with an unexpected bend between domains SCR3 and SCR4 . This bend can be attributed to a unique five-residue insertion into the CD' loop of SCR3 ( Figs . 3B , D ) . The insertion is not compatible with a linear arrangement of the SCR3-SCR4 interface but instead provides a platform that stabilizes the bent structure . The smaller SCR1-SCR2 interface possesses some flexibility [31] , and flexibility may also be a feature of the similarly-sized SCR2-SCR3 interface . However , our structure suggests that the SCR3-SCR4 interface has little , if any , flexibility as it has a much larger buried surface area , exhibits low temperature factors , and contains many rigid amino acids . The role of the CD' loop in SCR3 thus appears to be in forming a brace that molds the SCR3-SCR4 unit of CD46 into a bent conformation . Inspection of sequences of RCA family members shows that no homolog of CD46 contains a similarly elongated and hydrophobic loop [45] . Previous studies identified a number of residues that play a role in mediating interactions of CD46 with its many ligands . Our structure now places these data in proper context by displaying CD46 ligand binding surfaces on the extracellular portion of the molecule . Interactions of CD46 with Adv and MV are exclusively mediated by the SCR1-SCR2 region , and these interactions have been described earlier [32] , [33] , [34] . Our analysis indicates that interactions with C3b and C4b involve several regions on domains SCR2 , SCR3 and SCR4 . These regions are located on the convex surface of the curved receptor molecule , and they are devoid of glycosylation . The most extensive binding site for C3b is located on one side of SCR4 , and appears to depend on a number of charged residues . This region partially overlaps with a binding site for C4b . Smaller contact regions for both C3b and C4b are located at the base of SCR2 , near the SCR2/SCR3 interface . Thus , the large C3b and C4b proteins probably contact a significant portion of the surface of CD46 that is defined by the SCR2-SCR4 fragment , similar to the contacts observed in the recent crystal structure of the N-terminal four SCR domains of FH in complex with C3b [42] ( Fig . 5 ) . Therefore , the bent conformation of CD46 could be a highly significant determinant for the recognition of complement proteins . In contrast to soluble FH , which contains 20 SCR domains , the CD46 protein is much smaller and attached to the membrane . The presence of a preformed bend in the protein conformation could facilitate association of complement proteins to CD46 on the cell surface , reducing , or perhaps eliminating , the requirement for domain rearrangements during C3b and C4b binding . The structure reported here does not include the short STP region , which connects SCR4 to the single transmembrane spanning sequence of CD46 . We can therefore not provide a definitive view of how the CD46 molecule is arranged on the cell surface . The proline-rich nature of the STP region suggests that it has limited flexibility , perhaps serving as a stalk that provides some distance between SCR4 and the membrane surface . Two extreme possibilities for the conformation of CD46 on the cell surface can be envisaged ( Fig . 7 ) . In one of these , the SCR4 domain and the STP region project vertically from the cell surface , generating a protein arrangement in which the glycans face toward the membrane and the N-terminal SCR1 domain is near the cell surface ( Fig . 7A ) . Interactions of the glycans with the membrane could help to orient the molecule on the cell surface , with the glycan-free region being highly accessible for interactions with even large ligands such as complement proteins C3b and C4b . Moreover , the proximity of the SCR1 domain to the membrane , which serves as the main contact point for Adv and MV , would facilitate penetration of the cell membrane by those viruses , and in particular fusion of MV and cell membranes . In the second scenario , the STP region is bent , and the SCR4 lies more or less parallel to the cell surface ( Fig . 7B ) . The SCR1-SCR2 region would project into solution , and would readily be available for interactions with Adv and MV , but also more distant from the cell surface . If such an arrangement were to exist at the cell surface , it might preclude binding of C3b ( and perhaps C4b ) to CD46 as the predicted sites for C3b binding on SCR4 would face towards the membrane , and thus would not be easily accessible to the large C3b protein . In order to expose the complement binding sites on SCR4 , CD46 would need to adopt a conformation in which the SCR1 domain would be close to the membrane ( Fig . 7A ) . Multivalent interaction of the Adv knob with CD46 in this conformation would require either movements within the STP region toward an alternative CD46 conformation ( Fig . 7B ) , which could be limited by the proline rich nature of this region , or some plasticity in the cell membrane for virus binding to multiple receptor molecules . Trimeric binding of the knob to CD46 molecules adopting a conformation similar to that shown in Fig . 7B could be accomplished in concave membrane microdomains . Alternative splicing variants of the STP region could influence the orientation of the CD46 molecule on the cell surface . It has been shown that alternative splicing in this region has significant implications for complement regulatory function [46] , [47] as well as MV binding and fusion [47] , [48] . The overall structure of the CD46 extracellular region presented here differs drastically from earlier models that pictured CD46 as an elongated , rod-like structure , and suggests a more dynamic conformation of this receptor molecule on the cell surface .
A cDNA encoding residues 1 to 286 of the CD46 precursor protein was subcloned into the expression vector pBJ5-GS [49] . This vector was transfected into CHO Lec 3 . 2 . 8 . 1 cells [50] , and stable cell clone transfectants secreting the CD46-4D protein to the culture medium were selected with methionine sulfoximine , a glutamine synthetase inhibitor . Transfected cells were cultured in Ex-Cell 302 medium ( JRH Biosciences ) supplemented with 100 µM methionine sulfoximine , GS supplement ( JRH Biosciences ) , 50 units/ml penicillin G , 50 mg/ml streptomycin , 7 . 5 mM HEPES at pH 7 . 3 and 1% dialyzed fetal bovine serum . After harvesting , the culture supernatant was centrifuged and filtered . CD46-4D was then purified by Concanavalin A affinity chromatography ( Con A Sepharose , GE Healthcare ) , gel filtration ( Superdex 200 , GE Healthcare ) , and anion exchange chromatography ( MonoQ , GE Healthcare ) . Ad11 fiber knob amino acids 118–325 were expressed in E . coli Rosetta2 ( DE3 ) cells and purified via nickel affinity chromatography and gel filtration , as described earlier [32] . The complex was formed by incubating both proteins at 4°C for 2 hrs . A 1 . 2 molar excess of CD46-4D was used , based on the earlier observation that one trimeric knob can bind three CD46 ligands [32] . Separation of the complex from excess , unbound CD46-4D was performed by size exclusion chromatography ( Superdex 200 HR column ( GE Healthcare , Uppsala , Sweden ) in gel filtration buffer containing 20 mM HEPES , 150 mM NaCl at pH 7 . 4 . Well-diffracting plate-like crystals of Ad11 knob in complex with CD46-4D were obtained at 4°C using a precipitant solution containing 20% polyethylene glycol 1000 , 0 . 2 M ammonium phosphate at pH 8 . 0 with the use of a microseeding protocol [51] . Poorly diffracting crystals grown at 20°C in 20% PEG 6000 , 200 mM ammonium phosphate pH 8 . 0 were used for seeding . Crystals belong to space group P63 , with two copies of Ad11 knob protomers and two CD46-4D chains present in the asymmetric unit . The crystals were flash frozen in liquid nitrogen using precipitant solution supplemented with 25% PEG 200 for cryogenic protection . Diffraction data were collected at the Swiss Light Source ( beam line X06SA ) and ESRF ( beam line BM14 ) . Diffraction images were processed using XDS [52] and SCALA [53] , producing a data set that extends to 2 . 84 Å with good statistics . The structure determination was carried out by molecular replacement with Phaser [54] . Coordinates for the Ad11-knob protomer as well as the SCR1 and SCR2 domains of CD46 [32] were used independently as search models , after removal of surface loops that had elevated temperature factors in each case . This strategy produced two clear solutions for the complex , indicating the presence of two copies of Ad11 knob protomers and two CD46 molecules in the asymmetric unit . After initial rigid body refinement using NCS restraints in Phenix [55] , 2Fo-Fc and Fo-Fc difference electron density maps revealed the location of the SCR3-SCR4 portions in both copies of CD46 . These domains were then included into the model . The structure was built using Coot [56] and O [57] , and refined using REFMAC5 [58] , Phenix [55] and Autobuster [59] . The entire model could be built with the exception of residues 81–84 of SCR2 in chain D , which probably have multiple conformations in the crystal . Coordinates and structure factor amplitudes have been deposited with the Protein Data Bank ( PBD ID code 3O8E ) . Figure 5 was prepared using Molscript [60] , all other figures were made with PyMol [61] . Superpositions were done with LSQKAB [53] and the SSM routine in Coot [56] . CD46-4D has three N-linked glycosylation sites , at Asn49 , Asn80 and Asn239 . In order to produce a realistic estimate of size and distribution of the glycan structure of native human CD46-4D , we used the GlyProt online server [62] and modeled hybrid and complex glycans linked to the three Asn residues with NAG electron density . | The human membrane cofactor protein ( MCP , CD46 ) is expressed on all nucleated cells and serves as a marker that prevents host cells from destruction by the immune system . It functions as a cofactor that helps to inactivate the C3b and C4b molecules , which are central components of the complement system . In addition to its role in regulation complement activation , CD46 is also used by a large number of pathogens , including measles virus and adenovirus , as a receptor to allow these pathogens to attach to the cell surface and initiate an infection . We have determined the three-dimensional structure of the bulk of the extracellular region of CD46 using X-ray crystallography . This structure provides detailed information about the location of previously identified residues that play a role in the interactions with C3b , C4b , and several pathogens , advancing an understanding of the function of the CD46 protein as a host and pathogen receptor . Moreover , the structure also reveals an unexpected , bent conformation of the protein that has implications for how the binding sites are presented at the cell surface . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"immunology/immune",
"response",
"immunology/innate",
"immunity",
"cell",
"biology/cell",
"adhesion",
"microbiology/cellular",
"microbiology",
"and",
"pathogenesis",
"virology/host",
"invasion",
"and",
"cell",
"entry"
]
| 2010 | Structure of the Extracellular Portion of CD46 Provides Insights into Its Interactions with Complement Proteins and Pathogens |
The ability to induce a defense response after pathogen attack is a critical feature of the immune system of any organism . Nucleotide-binding leucine-rich repeat receptors ( NLRs ) are key players in this process and perceive the occurrence of nonself-activities or foreign molecules . In plants , coevolution with a variety of pests and pathogens has resulted in repertoires of several hundred diverse NLRs in single individuals and many more in populations as a whole . However , the mechanism by which defense signaling is triggered by these NLRs in plants is poorly understood . Here , we show that upon pathogen perception , NLRs use their N-terminal domains to transactivate other receptors . Their N-terminal domains homo- and heterodimerize , suggesting that plant NLRs oligomerize upon activation , similar to the vertebrate NLRs; however , consistent with their large number in plants , the complexes are highly heterometric . Also , in contrast to metazoan NLRs , the N-terminus , rather than their centrally located nucleotide-binding ( NB ) domain , can mediate initial partner selection . The highly redundant network of NLR interactions in plants is proposed to provide resilience to perturbation by pathogens .
Signal Transduction ATPases ( STAND proteins ) comprise an ancient group of modular proteins sharing a conserved nucleotide-binding ( NB ) domain [1] . STAND proteins are present in Archaea , Bacteria , and Eukaryota , implying a common , ancient evolutionary origin [1 , 2] . Duplications and associations of the NB domain with other functional domains have driven their divergent evolution , allowing them to participate in multiple signaling processes . Typically , STAND proteins act as intracellular receptors triggering cellular signaling responses upon elicitation . In animals , members of two major groups of STAND proteins , the nucleotide-binding oligomerization domain ( NOD ) -like receptors ( also referred to as NACHT [1] or animal nucleotide-binding leucine-rich repeat receptors [NLR] ) and the nucleotide-binding ARC [1] domain ( NB–ARC or simply NB ) -containing apoptotic proteins , include some of the key players involved in the induction of immune responses or programmed cell death ( pcd ) , respectively [1 , 3] . The NB domain controls the transition from a resting to an activated state through its involvement in differential adenosine diphosphate ( ADP ) /ATP ( or guanosine triphosphate [GTP] ) binding and nucleotide hydrolysis [1 , 4] . The best studied STAND protein , the Apoptotic Protease Activating Factor 1 ( APAF1 ) , induces pcd in human cells upon perception of cytochrome C released from mitochondria [5] . APAF1 activation triggers a conformational change that frees its C-terminal caspase-recruitment domain ( CARD ) and exposes its NB domain , enabling interactions with other APAF1 monomers [5] . Subsequently , intermolecular interactions are formed between the NB domains of adjacent monomers allowing formation of a circular heptamer called the apoptosome . The apoptosome is the active form of the protein and can initiate a caspase-signaling cascade resulting in pcd [6 , 7] . Similarly , APAF1 orthologs in Drosophila and Caenorhabditis elegans ( DARK1 and CED-4 , respectively ) form multimeric assemblies upon their activation and trigger pcd [8 , 9] . The NOD domains of metazoan NOD-like receptors also interact to form oligomeric assemblies of nine or more subunits [10 , 11] . The bacterial transcription factor MalT , which is evolutionarily related to ancestral STAND proteins , similarly oligomerizes to form a curved homopolymer upon its activation [12] . In all of these cases , oligomerization of the central NB or NOD domain serves to bring the N-terminal domains in close proximity , allowing their partners to interact and induce downstream signaling [13] . Hence , formation of apoptosome-like complexes facilitating the induced proximity of N-terminal domains may represent a common feature of STAND proteins [13] . In plants , numerous STAND receptors are present , and those that have been functionally characterized are mostly involved in innate immunity , conferring protection against diverse pests and pathogens [14 , 15] . At least part of their recognition specificity can be attributed to highly variable leucine-rich repeats ( LRRs ) , defining the C-terminal portion of plant NB-LRR receptors or plant NLRs . The majority of hundreds of genetically characterized disease resistance traits in plants map to genes encoding NLRs; the large numbers of such sequences in the genome and their high diversity reflect dynamic interactions between hosts and rapidly evolving pathogens [16] . NLRs are integral to effector-triggered immunity ( ETI ) [14] through direct or indirect recognition of effectors ( virulence-enhancing proteins secreted by pathogens during infection ) . ETI complements the less specific microbe-associated molecular pattern ( MAMP ) -triggered immunity ( MTI ) mediated by extracellular receptor-like kinases ( RLKs ) [14] . The NB–ARC domains of APAF1 ( a NB–ARC-WD40 type of receptor ) and its orthologs are the most similar at the sequence level of the NLRs outside the plant kingdom . Instead of a CARD , the N-termini of plant NLRs contain ( with some exceptions ) sequences similar to either Toll/interleukin-1 receptor ( TIR ) or a coiled–coil ( CC ) domains , allowing subclassification into TIR–NLRs ( TNLs ) and CC–NLRs ( CNLs ) [17] . In dicotyledonous plants , TNLs are sometimes more abundant than CNLs; however , in monocots , CNLs provide the core repertoire of receptors mediating ETI [18] . NLR activation most likely releases the N-terminal CC or TIR domain ( similarly to the release of CARD upon APAF1 activation ) to induce defense responses , which are often concomitant with pcd [19] . This type of ETI response is often referred to as the hypersensitive response ( HR ) [20] . Even though plant NLRs trigger pcd , universal signaling mediators ( such as caspases in case of APAF1 ) activated by plant NLRs have not been identified to date [3] . Unlike in vertebrates , which typically have fewer than two dozen NB–ARC or NOD proteins encoded in their genomes , the number of NLRs expressed in a single plant may exceed several hundred [21] . This large number provides potential for homo/heteromerization , but formation of multimeric complexes in plants following NLR activation has so far not been conclusively demonstrated [22 , 23] . Hence , it was currently an open question whether plant STAND proteins oligomerize upon activation . Consistent with their involvement in downstream signaling , in planta expression of TIR or CC domains alone can induce HR [24–27] , but their potential role in NLR oligomerization is unclear . Some TIR domains form dimers or even homotypic heteromultimers ( implying the existence of higher-order complexes ) when expressed without the adjacent NB and LRR domains , suggesting their involvement in NLR multimerization [23 , 28–30] . The extended CC domain of the barley powdery mildew resistance 10 ( Mla10 ) receptor ( Mla10-CC ) forms a helix-loop-helix rod-shaped homodimer , and mutations affecting dimerization compromise Mla-mediated resistance against powdery mildew [26 , 31] . Subsequent structural studies proposed that only the extended Mla10-CC dimerizes and folds into a monomeric four-helix bundle structure , a structure similar to that reported for CCs of wheat stem rust resistance 33 ( Sr33 ) and potato virus X resistance ( Rx ) receptors [32 , 33] . Full-length CC domains of Mla10 and Sr33 form homomeric and heteromeric associations , disruption of which compromises induction of cell death [32 , 34] . The observed monomeric and dimeric CC structures may reflect different states in a receptor’s activation [35] . Similarly , self-association of CC corresponding to Arabidopsis resistance to Pseudomonas syringae ( avrRpm1 ) ( RPM1 ) receptor appeared to be required for its activity [36] . Furthermore , heterodimer formation between the CC domains of two rice CNLs , RGA4 and RGA5 , is required to respond to the Avr-Pia effector from the fungus Magnaporthe oryzae [22] . So although homo- and heterodimerization of N-terminal domains had been shown for some CNL proteins , it was unknown whether the full-length proteins form higher-order complexes in plants and , if so , what the role of the CC was in this process [37] . NLRs in plants can be categorized into two functional groups , the sensors and the actors ( also referred to as helpers or activators ) , in which sensor NLRs are proposed to detect the pathogen-derived effectors , and an evolutionary conserved downstream-signaling partner NLR triggers defense [38–40] . The sensor/actor concept emerged after the discovery of a conserved class of CNLs referred to as CCR–CNLs , RPW8- , or NRG-like CNLs that share a distinct consensus sequence of their CC ( CCR ) domains [24 , 38] . CCR–CNLs are required for the activity of some canonical CNLs and TNLs , and expression of the CCR domain alone triggers extensive pcd , consistent with their proposed actor role in downstream immune–signaling [24 , 39] . Physical association between putative sensors and CCR–CNLs had not been demonstrated and how the phylogenetic diversity of CC domains reflects their roles in CNL cross-activation and signaling remains unknown [37] . To investigate the network of CNLs mediating ETI in plants and to elucidate the role ( s ) of CC domains in induction defense , we performed extensive genome-wide functional and in silico analyses of N-termini containing a predicted CC domain of nearly all of the CNLs in Arabidopsis thaliana ecotype Columbia-0 ( At-Col-0 ) . By combining data on their sequence variation with their ability to homo-/heterodimerize and induce cell death and/or disease resistance in three different plant species , we identified regions required for their function . Subsequent genetic mapping and reverse complementation confirmed the involvement of canonical receptors in CNL signaling in other plant species , implying that the role of CC domains in downstream signaling involves transactivation of other CNLs . Surprisingly , a highly variable part of the CC domain is required for this transactivation . Accordingly , we present two lines of evidence that NLR receptors form a network mediated by physical and functional associations .
CC domains are defined by heptad repeats of hydrophobic residues ( L , I , or V ) [41]; these form a binding interface of α-helical secondary structure that is involved in helix-to-helix binding [42] . We aligned the sequences of N-terminal fragments representing all CNL receptors predicted in At-Col-0 , including sequences encoding truncated receptor proteins referred to as CC–NBs ( CNs ) [17] . The alignment was refined using structure-based words ( patterns such as hydrophobic heptad repeats along with their predicted accessibility ) devised from crystallographic/NMR data of Mla10-CC , Sr33-CC , and Rx-CC [26 , 32 , 33] . The alignment ( S1 Fig ) revealed four major Groups that we designated Group A , B , C , and D and an E outgroup ( S2 Fig ) . The naming of these Groups is based on a previous study in which At-Col-0 CNLs were clustered based on the topology of their NB–ARC domains and the intron/exon features of the encoding genes [17] . A classical sequence alignment ( Clustal Omega ) [43] using the CC and NB–ARC sequences resulted in a very similar cladogram as compared to the NB–ARC domain sequence alone ( S3 Fig ) , implying that the CC domains follow a similar pattern of diversification as the remaining part of the receptor . However , the obtained trees were not identical; whereas , for instance , Groups B and D are well-defined in both studies , three members of Group C ( AT4G19060 , AT5G45440 , and AT5G45490 ) were placed with Group A in the CC–NB–ARC-based cladogram ( S3 Fig ) . These differences can be attributed to high diversity and rather ambiguous alignments of the CC-containing fragment . However , for consistency with the existing literature , we applied the same letter designations as before [17] . Group E gathered N-terminal fragments of the CN homologs lacking a clearly defined CC domain precluding their structure-based alignment ( S1 Fig , S2 Fig ) . Group D had the highest average sequence identity ( id . ) /similarity ( sim . ) of the Groups to sequences of Mla10 or Sr33 ( Mla10: id . 18 . 0% , sim . 44 . 3%; Sr33: id . 17 . 5% , sim . 39 . 2%; Rx: id . 11 . 1% , sim . 30 . 2% ) . Groups C and D , comprising eight and 14 members , respectively , were the most similar and shared the four predicted α helices of the CC domains of Mla10 , Sr33 , and Rx: H1a , H1b , H2a , and H2b [26 , 32] ( Fig 1 , S2 Fig ) . Even though all members of Group B also contain four predicted α helices , these helices did not align with helices predicted for Groups C and D ( S2 Fig ) . Detailed secondary structural analysis of Groups C and D revealed a profile that resembled Mla10–CC and Sr33–CC more than Rx–CC ( S1 File ) . This is seen especially in the H1a–H1b turn region; for these Groups , the separation of the first two α helices ( H1a and H1b ) was not consistently predicted , which is in agreement with crystallographic data for Mla10–CC , in which helices H1a and H1b form one single helix H1 [26] . However , the molecules may adopt different secondary structures in the presence of an interacting partner . In contrast , to the H1a–H1b region , the predicted helices H1b , H2a , and H2b were clearly separated from each other by areas of flexibility , which corresponds to turns in the three resolved CC structures [26 , 32 , 33] ( Fig 1 , S1 Fig , S2 Fig ) . The strongest sequence identity within Groups C and D occurs around the five-amino acid–long EDVID motif in helix H2a; in addition , Group D contains a conserved stretch of 10 polar amino acids immediately preceding this motif ( S2 Fig ) . The CC–NB–ARC-based phylogeny ( Fig 2A , S3 Fig ) places Group D within Group C; thus , both can be considered a merged C/D Group . In accordance with earlier reports , we did not find a clearly distinguishable EDVID motif in Group A and Group B , yet in Group B the corresponding area of eight amino acids showed some conservation , including a hydrophobicity distribution similar to the EDVID motif ( Fig 1 , S1 Fig ) . In Group B , representing more than a third of all analyzed CNLs , structure-based alignments predicted the existence of two short β-strands immediately preceding the first H1a helix and immediately following the H2b helix ( Fig 1 , S2 Fig ) . Also , only CNLs in Group B carry the previously described [44] putative myristoylation ( Gly-2 , Gly-3 ) and palmitoylation motifs ( e . g . , Cys-4 , Ser-4 , and others ) in their N-termini . In contrast to the N-terminal domain of Mla10 , Sr33 , and Sr50 , in which all four α helices appear to form CC structures , available algorithms [45] only predict formation of CC structures among N-termini of At-Col-0 CNLs for helices H1a and H1b . The smallest group , Group A ( comprising CCR–CNLs ) include the three previously described activated disease resistance 1 ( ADR1 ) homologs: ADR1 , ADR1-L1 , and ADR1-L2 [47 , 48]; two NRG1 homologs ( AT5G66900 and AT5G66910 ) [24]; and a third NRG1-like protein annotated as DAR5 ( AT5G66630 ) ( Fig 1 ) . Each of these six homologs contain RPW8-defined consensus sequences and four predicted α helices , yet these did not align with those of Groups B , C , and D ( Fig 1 , S1 Fig ) . In contrast to canonical CNLs , formation of CCs by N-termini of CCR–CNLs was predicted for three C-terminal α-helices but not the first α-helix . Heteromeric interaction between CC domains of the rice RGA4 and RGA5 CNLs and homomeric association of Mla10-CC have been previously detected using yeast 2 hybrid ( Y2H ) assays [22 , 26] . Therefore , we used Y2H to assess homo- and heteromeric interactions of the N-terminal CNL fragments . We cloned 56 DNA fragments encoding the N-terminal regions of CNLs , referred to hereafter as extended CC domains ( ECCs ) ( Fig 1 , S2 Fig , S2 File ) . Each ECC contains the entire N-terminus up to the predicted P-loop in the NB–ARC domain . This region includes the amino acids immediately preceding the predicted α-helices and the linker encompassing the pre-P–loop motif ( Fig 1 ) . Next , we tested their interactions in Y2H assays using initially two and later one pair of vectors , creating fusions to binding and activation domains ( see Materials and methods ) . Five ECC fragments showed homomeric associations ( Fig 2B ) , while 123 heteromeric interactions were observed involving 39 ECCs . Interactions occurred at similar frequencies between sequence-related and sequence-unrelated ECCs . There seems to have been no evolutionary selection toward homomeric associations because homomerization was observed only for ECCs that also interacted frequently with other partners . If ECCs had evolved to facilitate CNL homodimerization ( like previous studies have implied [22 , 26 , 32] ) , homomeric interactions should be prevalent , and heteromeric interactions between ECCs corresponding to close paralogs should be infrequent . However , 34 ECCs showed higher affinity toward at least one partner other than themselves . The large number of heteromeric interactions suggests that N-termini–mediated heteromerization is a common feature of CNLs . To better understand the molecular features within ECC domains that may be required or involved in the interaction , we correlated their sequence variation and the ability to interact in Y2H with their predicted CC monomeric and dimeric structures . First , all possible models of monomeric and dimeric structures that could be derived from available crystal structures were built . Based on knowledge and physical binding free energy calculations , the most probable models were retained ( S1 File ) . The monomeric structural model ( mono4α ) resembles the four-helix bundle observed in Sr33–CC [32] , whereas the dimer model resembles the intertwined CC structure ( 2α ) observed in Mla10–CC [26] . Binding free energy calculations ( S1 File ) suggests that dimer configurations based on two mono4α domains have a higher binding free energy than the 2α monomers and are less likely . Second , we identified sequence variants that could be correlated with the ability to interact . For this , we focused on members of Group D , as these share overall high-sequence similarity but differ in their ability to interact ( Fig 2B ) . As compared to rarely interacting ECCs in Group D , four amino acids were found to be conserved and unique to frequently interacting ECC domains of AT1G58390 , AT1G58848 , AT1G59218 , AT1G58807 , and AT1G59124: C21 , S42 , V57 , and R107 ( S1 Fig ) . To resolve their putative location on the protein surface , the positions of these residues were mapped on both the monomeric and dimeric 3D models ( Fig 3 ) . In the dimer , C21 , S42 , and R107 add up to six amino acids located on the H1 helices . All six residues are surface exposed and on the same face of the protein , while V57 has a more lateral location . In the monomer structure , C21 resides in a highly flexible region and may reach the proximity of the EDVID motif ( Fig 3 ) , whereas in the 2α dimer configuration , it resides in a rigid region , and the EDVID motif is no longer within reach . The genetic variation in Group D enabled correlating oligomerization potential to four amino acids residues that likely form a patch on the CC surface . However , the exact role of these residues in this process remains to be resolved in future studies and awaits elucidation of the protein structure . Transient expression of a CC domain can induce necrosis and activate defense-related genes , thereby recapitulating induction of HR and defenses triggered by activated full-length CNLs [24 , 26 , 34] . To obtain a comprehensive assessment of abilities to induce cell death across CNLs , we transiently expressed all ECCs in At-Col-0 plants using the Tobacco Rattle Virus ( TRV ) system [49 , 50] . Arabidopsis is a good host for TRV supporting its systemic spread [51 , 52]; therefore , utilization of TRV allowed us to monitor not only necrosis but also disease resistance by assessing viral spread using reverse transcription PCR ( RT-PCR ) ( Fig 2 ) . Expression of ECCs in At-Col-0 plants induced various phenotypes . These responses varied from very mild leaf deformations to more severe deformations of the entire plant and , in extreme cases , necrosis or even death of infected plants ( Fig 2C ) . In a separate category of response , full resistance was observed in which viral movement was blocked without any development of symptoms , and accordingly , TRV was not detected in leaves after inoculation . Different colors and numbers ( zero through six; legend in Fig 2C ) were assigned to these distinct categories for visualization and correlation analysis . None of the members of Group D elicited a macroscopically visible response ( Fig 2B ) . Expression of ECCs corresponding to the CN genes [17] belonging to Group E also did not induce any visible response in At-Col-0 plants . However , within Group C , two ECCs induced necrosis ( AT3G46710 and AT3G46530 ) , and one triggered mild plant deformations ( AT3G14470 ) ( Fig 2 ) . In Group B , 14 of the 23 ECCs induced either necrosis or resistance in infected plants . Notably , all ECCs corresponding to CCR–CNLs ( Group A ) induced either necrosis or resistance . This Group includes AT5G66630 , which our functional analysis classified as a third functional NRG1 homolog , raising the total number of CCR–CNLs in At-Col-0 to six . Besides CCR–CNLs , the ability to induce cell death was more prevalent ( but not exclusive ) for ECCs in Group B lacking the conserved EDVID motif ( Fig 1 ) . In conclusion , the ability of ECCs to trigger immunity varied over the different Groups from none ( Groups D and E ) to all members ( Group A ) . Notably , whereas some ECCs triggered necrosis to various extents , others had the ability to fully block viral movement , showing that expression of an ECC alone can be sufficient to trigger disease resistance . CC-mediated oligomerization may be involved in communication between sensor and actor CNLs at either activation or downstream-signaling stages . To assess the requirement of these interactions for immune/HR signaling , we investigated whether these properties were correlated within and between NLR Groups . Therefore , we mapped all known resistance specificities onto the ECC interactome in order to link known functions to the ( in ) ability of the respective ECCs to interact and/or induce necrosis/resistance . The number of interactions , including self-association , and the ability to induce necrosis or resistance between ECCs were not significantly correlated ( r = −0 . 035; p ≤ 0 . 1 ) . Apparently , the ability of ECCs to interact is independent of their ability to trigger immune responses . Accordingly , we found that many noninteracting ECC fragments triggered necrosis ( e . g . , ECC corresponding to AT1G12290 , AT1G61180 , and AT4G27190 or ECCs corresponding to CCR–CNLs: AT5G66900 or AT5G04720 ) , while several ECCs showing extensive interactions lacked the ability to induce cell death ( e . g . , ECC corresponding to AT5G63020 or ECCs corresponding to the bottom clade within Group D; Fig 2C ) . Four out of five ECC fragments capable of homomerization triggered only mild or moderate plant responses . Notably , the same five fragments were also among the most frequent interactors with other partners , implying common features required for both associations and promiscuity in these interactions . CCR–CNLs have been proposed to function as actors [24] for sensor CNLs to induce defenses following pathogen perception . This hypothesis corresponds with the ability of their N-termini to induce cell death [38 , 53] . In our screens , the frequency of interactions between CCR–ECCs ( putative actors ) and ECCs corresponding to canonical CNLs ( putative sensors ) were not different ( p ≤ 0 . 05 ) from frequencies of interactions between ECCs of canonical CNLs . Because ECCs of several canonical CNLs induced cell death similarly to CCR–ECCs ( implying that they also function as actors ) , we examined whether necrosis-inducing ECCs interacted more frequently than those that did not . Again , the frequencies of such interactions were not higher than expected from random distribution . Thus , despite the fact that in our assays , ECCs of several putative sensors , such as RPS5 , ZAR1 , or RPP13 homolog ( note that the ECCs of RPS5 and ZAR1 did not induce cell death ) , interacted with several CCR–ECCs , we did not find evidence for higher prevalence of such interactions as compared to associations of ECCs corresponding to putative sensors . Moreover , the ECC of RPS2 , a receptor whose function depends on three CCR–CNL homologs ( ADRs ) [38] , did not interact with any of the ADR ECCs in our screens . Accordingly , based on the ECC interactome , we did not find evidence for preferential CC-mediated communication between putative sensors and actors; however , this does not preclude the possibility that sensors and actors may interact transiently in planta or via other domains , as shown for the RGA4/RGA5 pair in rice [22] . The extensive heteromeric interactions between ECC members from different classes is suggestive for such a signaling network . Genetic support for a CNL-signaling network comes from mutating hubs that compromise its activity . Despite mutating many putative interacting NLRs , no loss of necrosis was obtained upon ECC expression ( Fig 2B , S3 File ) . This result is consistent with the resilience of a network in that other hubs can take over the function of the mutated node . To obtain evidence for a CNL-signaling network activated by At-Col-0 ECCs , we assessed the activity of these fragments in heterologous species . The rationale was that in a heterologous species , redundancy might be lower , as the node ( representing a CNL ) did not coevolve with the ECC that triggers the response . Occurrence of an ECC-induced response in evolutionarily distant species would imply conservation and compatibility of the immune-signaling network in these species . Furthermore , if a polymorphic response is obtained in a heterologous species , it might allow identification of the ECC-interacting partner ( s ) . To evaluate the ability of CNLs to induce defenses in other species , we expressed all At-Col-0 ECCs in Nicotiana benthamiana ( Nb ) and in lettuce , Lactuca sativa ( Ls ) cultivar ( cv . ) Ninja ( Fig 4 ) . ECCs corresponding to CCR–CNLs induced strong responses not only in Nb , as reported previously [24] , but also in lettuce and the source species Arabidopsis . Out of the 16 Group B ECCs that triggered necrosis or resistance in At-Col-0 , 13 induced necrosis in Nb and five did in lettuce , implying greater compatibility with downstream-signaling components in the source species and in Nb than in lettuce . This response is unlikely to be triggered by general toxicity because for some ECCs induction of necrosis was clone-specific and sometimes limited to one or two of the three species tested . The finding that many ( and often the same ) CC fragments triggered responses in distantly related species indicates conservation of compatibility with specific signaling partners . Induction of cell death by ECCs in the three plant species allowed us to screen for an ECC/species combination that showed a polymorphic response upon At-Col-0 ECC expression . Interestingly , transient expression of ECC of AT4G14610 in lettuce cv . Ninja triggered clear cell death , whereas it did not elicit necrosis in cv . Valmaine ( Fig 4 , Fig 5A ) . This differential response allowed genetic mapping to determine the genomic location of the potential AT1G14610-signaling partner in lettuce . In F1 hybrids between cv . Ninja and Valmaine and derived F2 plants , we observed intermediate phenotypes , indicating that either the ability to induce necrosis in cv . Ninja or the lack of a response in cv . Valmaine is due to incomplete dominance . Variation in plant response to the ECC of AT4G14610 expression was assessed in 75 individual F2 plants on a scale from 1 to 5 . All plants were genotyped , and subsequent QTL mapping linked the variation to a single locus on linkage group 3 ( LG3 ) of lettuce ( Fig 5A ) . This locus contains candidate disease resistance genes , including sequences encoding TNLs and a large CNL family previously described as the Resistance Gene Candidate 21 family ( RGC21 ) [55] . In a previous project , transgenic lettuce plants of cv . Cobham Green ( CG ) , referred thereafter as CGsil-RGC21 , were generated expressing an interfering hairpin RNA ( ihpRNA ) designated to trigger posttranscriptional gene silencing ( PTGS ) of RGC21 family members [55] . The ECC of ATG14610 did not trigger necrosis in cv . CG; consequently , cell death assays could not be performed in CGsil-RGC21 . Therefore , these silenced CGsil-RGC21 plants were outcrossed to cv . Ninja and to cv . Valmaine . Among 30 F1 hybrids derived from the cross between cv . Ninja and CGRNAi-RGC21 , 14 were identified as silenced and 16 as not silenced for RGC21 , which is as expected due to the hemizygous state of the transgene in CGRNAi-RGC21 ( Fig 5B ) . All 16 nonsilenced hybrids showed a similar weak necrosis following expression of ATG14610–ECC as F1 cv . Ninja x cv . Valmaine F1 hybrids and many F2 plants derived from the same cross ( Fig 5B ) . In contrast , similar to wild-type cv . CG plants , none of the 14 RGC21-silenced hybrids showed necrosis , indicating that one or more RGC21 member ( s ) are required for induction of ATG14610-mediated cell death . Furthermore , none of the cv . Valmaine x CGRNAi-RGC21 hybrids showed necrosis despite many being identified as silenced . Notably , all accessions and hybrids described above responded with moderate necrosis to transient expression of bacterial effector AvrPto [58] , indicating that the activity of RGC21 member ( s ) in induction of cell death is specific to the ECC of ATG14610 . The sequences and repertoire of RGC21 paralogs in cv . Ninja are unknown , and the exact identity of the RGC21 member ( s ) acting downstream of ATG14610 remains undetermined . However , reverse BLAST of RGC21 sequences ( Gene Bank accession number EU889315 . 1 ) to At-Col-0 sequences specifically identifies multiple members of Group C/D as closest homologs , indicating that ATG14610 requires member ( s ) of this Group for cell death induction in lettuce . From this experiment , we concluded that an At-Col-0 ECC requires CNL partner ( s ) to induce cell death and that despite the large diversity among NLR receptors , compatibility between CNLs can be retained across distantly related plant species . To identify the region/motifs within ECCs required for their activity , we searched for correlations between all 56 sequence variants and their ability to trigger cell death . Following the predicted four α helices , each ECC fragment contains a variable linker that separates the last α helix ( H2b ) from the pre-P–loop at the beginning of the NB–ARC domain . Alignments of all ECCs refined the consensus of a pre-P–loop motif among CNLs in Arabidopsis to V/IG x ( 8 ) L/I x ( 3 ) L and disclosed a cluster of charged amino acids within the linker region . This cluster , which we refer to as the “charged motif , ” maps to positions −11 to −3 relative to the highly conserved VG residues of the pre-P–loop motif ( S1 Fig ) . To determine exactly which regions in the ECC are required for At CNLs to trigger signaling , we generated a series of deletions , swaps , and point mutants involving cell death–inducing and non-cell death–inducing ECCs to delineate the region responsible for induction of cell death ( Fig 6 , S4 File ) . We focused on four ECCs in Group B because their high homology and accurate alignment allowed precise swaps , yet they had different phenotypes to differentiate the output . The following amino acids residues were used as break points to create chimaeras: the last conserved hydrophobic residue of heptad repeat of H2b ( referred thereafter to as CC-END ) , the start ( VG residues ) of the pre-P–loop , and the charged motif ( Fig 6 ) . Because the selected ECCs induced necrosis in more than one plant species ( Fig 4 ) , we examined the plant response in At-Col-0 , Nb , and in lettuce cv . Ninja . Four wild-type clones and all chimeras were fused to C-terminal hemagglutinin ( HA ) tag to evaluate their expression in Nb ( Fig 6A ) . Reciprocal swaps at CC-END between ECCs ( corresponding to two pairs of CNLs: AT1G63360 and AT1G15890 , and AT1G62630 and AT4G14610 ) surprisingly resulted in chimaeras that lost their ability to induce cell death in any species ( A1 and A2 , B1 and B2; Fig 6A ) . Accordingly , clones that previously did not induce necrosis ( like ECC of AT1G63360 in Nb and in lettuce or ECC of AT1G62630 in lettuce ) did not gain this ability after introducing the CC-END-linker-pre-P–loop fragment from a necrosis-inducing clone ( A1 and B1 , respectively; Fig 6 ) . This indicated that regions upstream and downstream of CC-END were required but alone were insufficient for induction of cell death . Furthermore , these results showed that compatibility between both regions was essential for induction of cell death . A reciprocal swap at the VG residues ( pre-P–loop ) between the same two pairs of ECCs ( A3 , A4 , B3 , and B4; Fig 6A ) did not affect their ability to induce cell death nor their patterns across the three plant species , implying that the pre-P–loop itself is not required for elicitation of cell death . Deletions in ECCs at the charged motif did also not affect the ability or patterns of necrosis induction in three clones ( clones A5 , A6 , and B5 ) but eliminated growth deformations triggered by the ECC of AT1G14610 in At-Col-0 ( clone B6; Fig 6A ) . This implied requirement of the fragment between CC-END and the charged motif for cell death induction . The charged motif itself was not essential for cell death induction but clearly modulated the response . Indeed , substitutions G158A at the beginning of pre-P–loop and E/E148/149A/A within the charged motif in the ECC of AT1G12290 did not compromise cell death induction ( M1–M3; Fig 6B ) in At-Col-0 plants , but substituting the same EE residues with positively charged KK potentiated the immune response and made At-Col-0 fully resistant to viral infection ( clone M4; Fig 6B ) . Reciprocal swaps at the charged motif between ECCs corresponding to AT1G63360 and to AT1G15690 ( the exact position of the swaps is shown in the left panel in Fig 6A ) resulted in the elimination of plant deformations induced by the former in At-Col-0 and weakened the response induced in all three plant species by the latter ( A7 through A10; Fig 6 ) . A similar swap between ECCs corresponding to AT1G62630 and AT4G14610 eliminated the response induced by the latter in At-Col-0 ( B8 and B10 , Fig 6A ) . Swapping the entire charged motif between the ECC of AT4G14610 with the ECC of AT1G62630 compromised cell death induced in lettuce and At-Col-0 but not in Nb ( B10 , Fig 6A ) . This implied that the charged motif , besides modulating the strength of the response , may also determine host specificity . To exclude the possibility that a lack of responses was due to a lack of stability of the chimaeras , protein accumulation of wild-type ECCs and their derived variants was assessed in Nb leaves using immunoblotting ( Fig 6C ) . Full-length ECCs migrated at an apparent size of 25 to 35 kDa , which is slightly larger than the predicted sizes of 22–23 kDa ( including HA tag ) and might be attributed to the high content of hydrophobic residues ( V , I , and L ) . Protein accumulation levels varied between constructs and appeared to be slightly reduced for four clones swapped at CC-END ( A1 , A2 , B1 , and B2 ) . Notably , A3 or A10 triggered plant responses despite comparably low accumulation levels; this implied that the amount of protein in the aforementioned swaps should have sufficed to trigger a response in the three plant species tested . Interestingly , ECC At1G63360 and the A4 , A8 , and A9 proteins ran at an apparent increased molecular weight , suggesting an unknown posttranslational modification of these fragments . The reduced accumulation of A1 , A2 , and A5 through A7 , A10 , B1 , and B2 , as compared to the WT protein , may indicate a stabilizing role for this region and necessity for compatibility between this motif and the N-terminal portion of the fragment . To conclude , the lack of responses following expression of the chimaeras was likely due to a loss of activity rather than a reduced accumulation or stability of the produced protein . From these experiments , we concluded that the integrity of a variable stretch of 16 to 18 amino acids ( referred hereafter to as CC-variable amino acids residues [CCVX] ) following the last hydrophobic residue in helix H2b is required ( but insufficient ) for induction of cell death ( Fig 6 and Fig 7 ) . The potency of the CCVX fragment in induction of cell death can be modulated by its charge , implying the involvement in electrostatic intra- or intermolecular interactions . Because of its involvement in cell death elicitation , the charged motif may be analogous to the hydrophilic motif identified in three homologous CNLs in monocots [34] despite being localized in a slightly different position relative to H2b ( Fig 7 ) .
Among STAND receptors , the presence of an N-terminal CC domain is exclusive to CNL pathogen receptors in plants [2 , 37] . Our interactome analyses revealed a high tendency of ECCs to form heteromeric interactions as compared to homomerization , suggesting functional importance of associations involving different CNLs . Approximately two-thirds of the ECCs associated with more than one ECC , which implies functional redundancy among the CNL interactome . Consistently , knockouts of multiple NLR interacting partners in Arabidopsis did not compromise the ability of a specific At-Col-0 ECC to trigger an immune response ( S3 File ) . In lettuce , however , a single dominant CNL locus ( RGC21 locus ) could be associated with the ability of AT4G14610–ECC to trigger HR . Existence of a CNL network in Arabidopsis is consistent with the availability of the many putative sensor- and actor-type receptors [39] . In tomato , an NLR network is defined [39] in which sensors signal via downstream actors . Whether the tomato sensor NLRs form similar heteromeric interactions as the At-Col-0 ECCs ( Fig 2A ) needs to be resolved . Unlike in tomato , involving CCR–CNLs as actors ( or helpers ) [39] , the partnering receptor of AT4G14610–ECC in lettuce appeared to be an EDVID-type CNL ( Fig 2B ) . Nevertheless , the ability of NLRs to transduce their signal via various partners seems a recurring theme in plant immune signaling because not only NLRs involved in ETI form networks but also At-Col-0 LRR receptor kinases involved in MTI show an extensive network of interactions and functional dependencies [59] . A network provides resilience to perturbation and manipulation by pathogens and facilitates compatibility to adjust to new recognition specificities [40] . Identifying a lettuce CNL as a downstream-signaling partner for AT4G14610–ECC also has implications for CNL evolution . Canonical NLRs evolve rapidly , being highly diverse across different taxa , yet despite approximately 100 million years following the separation of superrosid ( Arabidopsis ) and superasterid ( lettuce ) lineages [60] , the identified NLR pair retained a functional relationship . Interactions between Group B and Group C/D members were frequent ( Fig 2B ) . Because compatibility between the AT4G14610–ECC and RGC21 CNL were retained , specific molecular features must be conserved within the N-terminal fragment . For Mla10 homologs , both the predicted α-helices and the hydrophilic motif was proposed to be involved in CC–CC interactions [26 , 32]; hence , the evolutionarily conserved feature retaining compatibility between ECCs may simply be the ability to oligomerize . Adaptation of the TRV-based expression system enabled functional analysis of the ECCs in the source plant , Arabidopsis , in N . benthamiana , and in lettuce . Induction of cell death by ECCs was not unique to CCR–CNLs ( the number of which we increased to six by the identification of AT5G66630 ) but could also be triggered by approximately half of the ECCs present in the repertoire of CNLs in At-Col-0 . Notably , five ECC fragments were able to trigger full immunity in At-Col-0 to TRV infection . It is unclear whether these fragments themselves function as actors or represent sensors signaling via an actor . Overexpression of some confirmed sensor ECCs , such as the Group B members RPS2 and RPS5 , triggered a mild reaction but no resistance . Overexpression of ECC domains from putative sensors from Group C ( RPM1 and ZAR1 ) did not trigger any response like most other members of this Group and from Group D . Together , these findings show that an unambiguous distinction between actors and sensors based on the ability of their ECC to trigger immune responses may not be possible . Nevertheless , these observations together demonstrate that the ECC represent the “effector” domain of a CNL that is sufficient to trigger full resistance , supporting recent conclusions of Jacobs and coworkers [27] . Furthermore , our findings are in agreement with loosening the association between the CC domain and the remainder of the CNL following conformational rearrangements upon activation [15] . Such a mechanism is analogous to release of the CARD of APAF1 or metazoan NLR receptors following their activation [5 , 10 , 61] ( S3 Fig ) . However , in contrast to the CARD in metazoan receptors , the CC is capable to signal and interact autonomously and apparently does not require an NB–ARC to trigger defense responses when heterologously expressed in planta . A recent study revealed an increased binding affinity of the CC domain of RPM1 toward CC–NB–ARC domain variants that harbored autoactivating mutations [36] . Notably , this increased affinity was only observed using extended CC domains containing the NB–ARC linker encompassing the charged motif identified in our study . This finding implies that ECCs may transactivate full-length downstream CNLs in at least two distinct ways . First , ( as indicated by our Y2H data ) , the CC of one receptor may interact with the CC of the downstream partner , thereby relieving the negative regulation of its NB , allowing it to adopt an activated ATP-bound conformation . Second , the ECC linker may directly interact with the NB domain of a partner CNL , thereby activating the receptor . In either scenario , a cascade of transactivation may be initiated because the released ECCs could potentially activate additional CNL receptors . The presence of two interaction surfaces ( CC–CC and CC–NB ) would stabilize the complex formed , facilitating the formation of multimeric ( yet self-terminating ) complexes similar to heteromeric inflammasomes described for mammalian NLR family apoptosis inhibitory protein 2 ( NAIP2 ) and NLRC4 [7 , 10] ( S4 Fig ) . In contrast to the metazoan NLRs in which the NB triggers transactivation and multimerization , CNLs appear to be able to also employ the CC domain to transactivate other CNLs and possibly nucleate NLR multimerization ( S4 Fig ) . After activation and reaching initial proximity , further association of monomers could involve the NB–ARC domain of the full-length receptor . Such a mode of association is consistent with oligomerization of other STAND receptors , such as APAF1 , in which the interface between monomeric subunits is complex and involves different parts of the receptor [6] . We demonstrated that variable area ( CCVX ) ( Fig 6 , S2 Fig ) located in CC–NB linker is involved but not sufficient for induction of cell death . Chimeras swapped at the last hydrophobic residue of the CC lose their ability to induce necrosis , implying that compatibility between the C-terminal part of the H2b helix and the CCVX with the remainder of the CC is required to induce cell death . This requirement implies that certain structural features are needed to trigger cell death , which is consistent with the hypothesis that the CC may share its origin with metazoan death folds [37] . Accordingly , the hydrophilic motif [34] and the CCVX may be structural motifs required for CC folding and function . The observation that substitution of EE residues within the charged motif to positively charged lysines enhanced the plant immune response opens up an interesting possibility to engineer CNL receptors to enhance their performance after activation in a more subtle manner than mutations in the NB–ARC domain that typically trigger autoactivity or loss of function [62 , 63] . Possibly , a potentiated ECC might confer a stronger defense against pathogens . The transfer of resistance genes ( encoding NLRs ) from one species to another is believed to require compatibility of upstream ( such as decoys ) and downstream-signaling partners [64] . If endogenous CNLs are immediate signaling partners of receptors , then the observed interspecies compatibility is surprisingly high , as suggested by the similar patterns of HR induction in unrelated plant species ( Fig 4 ) . Notably , CCs of Mla10 homologs from a monocot induce cell death in the dicot Nb [26 , 32 , 34] . Therefore , it is surprising that their structurally closest homologs in At-Col-0 ( Groups C and D ) ( Fig 2B , S1 Fig ) rarely induced necrosis . Whether this lack of functional homology can be attributed to convergent evolution or represents diversification in the ability of the ECCs to trigger defenses among domains of a common origin remains a question for future study . It will also be interesting to test whether compatibility of ECC signaling in heterologous species solely involves compatibility to the corresponding CNL , or whether there are other requirements . If not , this would suggest that CC-mediated oligomerization is the main factor in CNL immune signaling , as proposed in our model in S4 Fig . ECC-driven formation of multimeric CNL complexes ( similar to apoptosomes or inflammosomes in metazoa ) [65] results in the formation of the active resistasome ( or NLRsome ) , thereby providing a platform for communication between sensor and actor NLRs . The ability of ECCs to recruit other receptors and the self-terminating nature of a circular complex may enable precise control of defense induction , which is consistent with the quantitative nature of plant immune responses .
Protein sequences were profiled for their predicted physicochemical profiles , as previously described [66–68] . Profiles were raised for linker , coil–coil , intrinsic disorder , secondary structure , contact , and turn-forming propensities [69–73] . For each profile , several methods were used , and the consensus was built to increase the prediction reliability , as described in the references above . The alignment was refined using Mla10–CC , Sr33 , and Rx–CC structural data [26 , 32 , 33] . Sequence-clustering/phylogenetic tree–building was carried out using structural words weighting and variability analysis . Remote homology 3D models of CC domain structures were built , as described [26 , 74] . In essence , the alignments of target sequences were optimized , incorporating predicted secondary structure profile data as well as other predicted physicochemical profiles , followed by threading with SLIDE [75] . Molecular modeling was performed using Discovery Studio software ( Accelrys-Dassault Systèmes ) and Modeler v9 . 18 [76] . Along the conserved regions of the proteins , coordinates were assigned using standard Modeler coordinate transfer functions , while insertion loops were generated randomly and chosen by energy- and steric-based procedures . Generated loops were brought to local minima using a divide and conquer strategy , including recursive rounds of energy minimization and/or simulated annealing . The global model generated was further subjected to energy minimization , followed by global and local quality check using MetaMQAP [77] , MolProbity [78] , and PROCHECK V . 3 . 4 . 4 [79] for crystallographic standards compliance . The overall structural optimization was performed with NAMD [80] . Molecular dynamics simulation experiments were then performed with Amber16 [81] using ff14SB force field at 300 K and 1 bar . The standard protonation state at physiological pH ( 7 . 4 ) was assigned to the ionizable residues using H++ server [82] . Dimer structures were solvated with TIP3P waters in an octahedral box . Periodic boundary conditions and Ewald sums ( grid spacing of 1 Å ) were used to treat long-range electrostatic interactions . The nonbonded cut-off distance was maintained at 12 Å , and the temperature and pressure were controlled by Langevin thermostat and Berendsen barostat with coupling constant of 1 ps . The quality of the simulations was assessed by analyzing the potential energy , root-mean-square deviations , and root-mean-square fluctuations profiles from molecular dynamics simulations . Binding free energies were calculated using molecular mechanics-based MM/PB ( GB ) SA methods , as implemented in Amber16 package specialized scripts [81 , 83] and knowledge-based Prodigy method [84] . Visual inspection and protein structure graphics were performed with PyMol ( PyMOL Molecular Graphics System , v1 . 8 Schrödinger , LLC ) and VMD [85] . Hydrophobicity profiles ( Fig 1 ) were calculated using the Protscale server ( web . expasy . org ) using hydropathicity scale [86] and sliding window size of three residues . ECCs were expressed in planta plants using the TRV system [89] , in which pTRV2-attR2-attR1 was modified to enable expression of recombinant sequences by adding a coat protein promoter from Pea Early-Browning Virus ( PEBV ) [50 , 90] . The PEBV CP promoter [91] was amplified using primers 5′ATATGGTTACCGCACACAAGGTTAAAAACGCTG and 5′ATCTCGAGTTAGCTAGTTAGGCCTCTCGTTAACTCGGGTAAGTGA ( restriction sites are underlined ) and after digestion with BstEII and XhoI ligated to the pTRV2-attR2-attR1 vector digested with the same enzymes . Subsequently , the vector was converted into a Gateway compatible vector by introduction of a ccdB Frame B cassette ( Gateway Conversion System; Cat . No . 11828029 , ThermoFisher Scientific , https://www . thermofisher . com/ ) into the StuI site . To facilitate protein detection , a derived vector was created that contained the sequences encoding a human influenza HA tag introduced between StuI and XhoI sites . The vector contains a PEBV CP promoter , followed by a Gateway conversion site , as described above . Clones containing swap- and substitution-mutations were custom synthesized as gBlocks ( IDT; https://www . idtdna . com/ ) flanked by attB1 and attB2 recombination sites , and cloned to pENTR207 vector via a BP reaction . All ECCs and their mutants were recombined into this modified pTRV2-attR2-attR1 vector and transferred to Agrobacterium tumefaciens C58 ( RifR ) [92] . Inoculation assays were performed as described for At-Col-0 [93] , Nb , and lettuce [94] . Prior to infiltrations , suspensions of A . tumefaciens ( OD600 = 0 . 5 ) harboring plasmids encoding TRV RNA1 [89] or TRV RNA2 were mixed in a 1:1 ratio . Col-0 plants were evaluated 4 , 6 , and 8 days post inoculation ( dpi ) , whereas the reactions in Nb and lettuce were scored 2 and 3 dpi . Phenotypes reported in Fig 1C , Fig 4 , and Fig 6 reflect the final score at the last day of observation . The assays in Col-0 were replicated at least twice for each clone using at least three plants in each replicate , the assays in Nb and lettuce were replicated at least twice and involved at least two leaves , each inoculated in two places . The presence of TRV , indicative of systemic infection , in At-Col-0 plants was detected by standard RT-PCR using primers design to amplify a fragment of the At-Col-0 actin-2 gene ( control; AT3G18780 ) and the sequences encoding the coat protein of TRV: TAACCCAAAGGCCAACAGAG and GGGCATCTGAATCTCTCAGC for actin-2 and ACGATTCTTGGGTGGAATCA and CGGTGCAGATGAACTAGCAG for TRV CP ( AF406991 ) . Total RNA was extracted from At-Col-0 leaves using RNeasy Plant Mini Kit ( Cat . No 74904 , Qiagen; https://www . qiagen . com/ ) , and cDNA used for PCR was synthesized using SuperScript II Reverse Transcriptase ( Cat . No 18064014 ) and random hexamers . Detection of HA-labelled proteins was performed as described [95] . Standard statistical tests involving Pearson Correlation and Chi-Square tests were applied to data analysis . For segregation analysis , a population of 75 F2 individuals derived from the cross between cvs . Ninja and Valmaine was tested for the response to the ECC of AT4G14610 . This population was genotyped using next-generation sequencing ( GBS ) [96] . In brief , DNA was extracted from each individual , digested with AvaII to reduce the genomic complexity , and ligated to unique barcoded adapters ( Truco and colleagues , in preparation ) . All samples were pooled and sequenced using Illumina HiSeq 4000 . After sequencing , TASSEL [97] was used for demultiplexing , read mapping against the lettuce reference assembly , and SNP calling . Custom scripts ( https://github . com/alex-kozik/atgc-xyz . ) were used to obtain single haplotypes per scaffold . Scaffold-based haplotypes were used to construct a genetic map using MSTmap [98] . QTL analysis was conducted using WinQTL Cartographer and Composite Interval Mapping [99] . Significance threshold at p ≤ 0 . 05 was calculated by permutation analysis ( 1 , 000 permutations ) . The graph in Fig 7A was created using CIRCOS ( http://circos . ca ) . Cv . CG was transformed with transgene LserNBS02_NB_RNAi ( chr 3 ) to producing ihpRNA corresponding to fragments of a RGC21 family member and the uidA gene [55] . Seedlings derived from two independent transgenic plants were tested for silencing by assessing a decreased transient GUS expression , as described previously [54] . Six individuals exhibiting silencing were crossed to cvs . Ninja and Valmaine . Hybrids were identified based on distinct morphology as compared to cv . Ninja used as a maternal parent and repeatedly tested for silencing using transient GUS expression . Progenies of two transgenic CG plants showed 1:1 segregation for the silencing phenotype and were tested for the response to ECC of AT4G14610 , as shown in Fig 7B . | The ability to induce defenses in response to pathogen attack is a critical feature of immunity in any organism . Nucleotide-binding leucine-rich repeat receptors ( NLRs ) are key players in this process and have evolved to perceive the occurrence of nonself-activities or foreign molecules . In plants , coevolution with a variety of pests and pathogens has resulted in repertoires of several hundred diverse NLRs in single individuals and many more in populations as a whole . The mechanism by which defense signaling is triggered by these NLRs is poorly understood . Here , we show that upon pathogen perception , NLRs use their N-terminal domains to transactivate other receptors . Their N-terminal domains homo- and heterodimerize , suggesting that plant NLRs oligomerize upon activation , similar to the vertebrate NLRs; however , consistent with their large number in plants , their complexes are highly heterometric . Also , in contrast to metazoan NLRs , their N-terminus , rather than their centrally located nucleotide-binding ( NB ) domain , can mediate initial partner selection . The highly redundant network of NLR interactions is proposed to provide resilience to perturbation by pathogens . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"and",
"conclusions",
"Materials",
"and",
"methods"
]
| [
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"cell",
"processes",
"brassica",
"immune",
"receptor",
"signaling",
"model",
"organisms",
"signs",
"and",
"symptoms",
"membrane",
"receptor",
"signaling",
"experimental",
"organism",
"systems",
"sequence",
"motif",
"analysis",
"plants",
"flowering",
"plants",
"lettuce",
"arabidopsis",
"thaliana",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"vegetables",
"necrotic",
"cell",
"death",
"animal",
"studies",
"bioinformatics",
"sequence",
"alignment",
"necrosis",
"signal",
"transduction",
"eukaryota",
"diagnostic",
"medicine",
"plant",
"and",
"algal",
"models",
"cell",
"biology",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"cell",
"signaling",
"organisms"
]
| 2018 | Genome-wide functional analyses of plant coiled–coil NLR-type pathogen receptors reveal essential roles of their N-terminal domain in oligomerization, networking, and immunity |
We have previously described the generation of a novel Ebola virus ( EBOV ) vaccine platform based on ( a ) replication-competent rabies virus ( RABV ) , ( b ) replication-deficient RABV , or ( c ) chemically inactivated RABV expressing EBOV glycoprotein ( GP ) . Mouse studies demonstrated safety , immunogenicity , and protective efficacy of these live or inactivated RABV/EBOV vaccines . Here , we evaluated these vaccines in nonhuman primates . Our results indicate that all three vaccines do induce potent immune responses against both RABV and EBOV , while the protection of immunized animals against EBOV was largely dependent on the quality of humoral immune response against EBOV GP . We also determined if the induced antibodies against EBOV GP differ in their target , affinity , or the isotype . Our results show that IgG1-biased humoral responses as well as high levels of GP-specific antibodies were beneficial for the control of EBOV infection after immunization . These results further support the concept that a successful EBOV vaccine needs to induce strong antibodies against EBOV . We also showed that a dual vaccine against RABV and filoviruses is achievable; therefore addressing concerns for the marketability of this urgently needed vaccine .
Several members of the Ebolavirus genus and Marburgvirus genus , Family Filoviridae , cause severe and often fatal viral hemorrhagic fever in humans and nonhuman primates [1] . While the public health burden of filovirus infections remains low relative to other public health threats in Africa , outbreaks continue to affect the Central African region including recent outbreaks in Uganda and the Democratic Republic of the Congo in 2012 . The high case fatality rate , the increasing public health threat in Africa , and the biodefense concerns associated with these viruses have resulted in considerable activity in filovirus vaccine development [2] , [3] . Several vaccination strategies , including DNA , adenovirus , recombinant vesicular stomatitis virus ( rVSV ) , virus-like particles ( VLPs ) and recombinant parainfluenza virus vectored vaccines , have been developed to deliver primarily the EBOV glycoprotein ( GP ) as antigen and have been shown to confer protection in animal models [2] , [4]–[7] . While each vaccine strategy has shown promising results and is protective in macaques , concerns such as vaccine safety , preexisting vector immunity , manufacturing , or lack of commercial interest have slowed progress . Recent investigations have focused on the identification of immune parameters that might serve as correlates of protection in vaccinated nonhuman primates ( NHPs ) . The majority of evidence suggests that IgG antibody levels are important for protection in immunized macaques ( adenovirus or rVSV-vectored EBOV GP ) although the contribution of neutralizing antibodies to protection is unclear [8] , [9] . Further support for the potential contribution of antibodies to protection was recently provided by two studies demonstrating that passive transfer of purified IgG from NHP survivor sera or neutralizing monoclonal antibody cocktails could confer protection from Ebola or Marburg virus infections [10] , [11] . In addition Marzi et . al . showed that the rVSV mechanism of protection for EBOV is mediated by antibodies [12] . EBOV-specific cellular immune responses have also been characterized after several immunization strategies including DNA/adenovirus and VLPs [7] . Using T cell depletion experiments , Sullivan et al . recently concluded that EBOV-specific CD8+ T cells and not humoral immunity mediated protection from EBOV infection upon adenovirus/EBOV-GP immunization [13] . Collectively , these studies suggest that immune parameters that correlate with and/or confer protection may be multi-factorial and vary by vaccination platform . However , we also need to consider that there are likely different requirements for the induction of anti-EBOV immunity and the recall response after exposure to the pathogen . It is not likely that long-lived immunity can be achieved without T-helper cells . In the case of GP-specific antibodies it needs to be shown that they are maintained over time or CD4+ T helper cells will be required to mount fast responses after infection . A filovirus vaccine would be directed for use in humans at risk of infection in Africa as well as for laboratory workers , healthcare providers , first responders , soldiers , or travelers . Furthermore , EBOV vaccines could be utilized in endangered wildlife species such as gorillas and chimpanzees in Central Africa where they are at risk of lethal EBOV disease . Epidemiologic studies have indicated that EBOV outbreaks have resulted in numerous deaths of these animals in Gabon and the Democratic Republic of Congo , hindering conservation efforts to protect these populations [14]–[16] . A vaccine to protect these at risk NHPs would have a second critical benefit to humans . As EBOV is a zoonotic disease with documented human outbreaks , which can arise from contact with diseased NHPs [17] , prevention of disease in these animals might reduce the frequency of EBOV transmission into humans resulting in reduced frequency of outbreaks . Our goal is to identify a vaccine platform for EBOV and other filoviruses of public health importance that would ( a ) produce promising candidates for use in both humans and endangered wildlife species and ( b ) yield multiple vaccine candidates increasing the likelihood that an optimal balance between reactogenicity and immunogenicity might be achieved . To this end , we have utilized the rabies virus ( RABV ) vaccine platform to develop ( a ) replication-competent , ( b ) replication-deficient , and ( c ) chemically inactivated vaccines expressing EBOV GP ( strain Mayinga ) [18] . As RABV is still a considerable public health issue in Africa with an estimated 24 , 000 deaths reported yearly [19]–[21] , a bivalent vaccine that confers protection from RABV and EBOV would be an economical and efficient public health tool . The RABV vaccine platform has proven to be an excellent vaccine vector for safe induction of immunity to HIV , SARS-CoV , and hepatitis C virus [22]–[26] . Further attenuated RABV-vectored vaccines have been generated by the deletion of the RABV glycoprotein ( G ) gene and propagation of viruses on trans-complementing cell lines that express RABV G [25] , [27] , [28] . Additionally , beta-propiolactone-mediated inactivation of RABV-vectored vaccines has been used to generate killed vaccine candidates against hepatitis C virus and bacillus anthracis with optimal safety profiles [22] , [29] . Our primary focus is the development of an inactivated vaccine for use in humans based on the potential for superior safety and the successful history of the existing beta propiolactone-inactivated RABV vaccine that is widely used in humans . However , in addition to the development of inactivated RABV/EBOV vaccines , the parental recombinant RABV vaccine used to generate the RABV/EBOV vaccine candidates is derived from the SAD B19 strain which is used for wildlife vaccination by baiting in Europe suggesting additional applications of our vaccine candidates [30] , [31] . Therefore , live attenuated RABV/EBOV vaccines could be considered for use in Africa in an analogous campaign to protect at risk NHPs from lethal EBOV infections .
Our previous research with RABV-based vaccine vectors expressing HIV-1 antigens indicated that such vaccines are highly immunogenic in NHPs against both the RABV-based vector and foreign antigens [25] , [32] . However , only replication-competent vaccine vectors expressing HIV-1 GagPol or Env have been analyzed in NHPs , and immunogenicity against filovirus antigens expressed in the RABV vector needs to be evaluated in the NHP model . Here , we analyzed the immunogenicity of three different RABV/EBOV vaccine vectors in NHPs , namely replication-competent ( BNSP333-GP ) , replication-deficient ( BNSPΔG-GP ) and inactivated virions ( INAC-BNSP333-GP ) expressing or carrying EBOV GP . The empty , replication-competent vector ( BNSP333 ) served as a control ( Figure 1A , [18] ) . As outlined in the immunization schedule in Figure 1B , four groups of rhesus macaques were immunized intramuscularly ( i . m . ) in the caudal thigh muscle as follows: group 1 , three NHPs , 5×107 focus-forming units ( FFU ) BNSP333 , black; group 2 , four NHPs , 5×107 FFU BNSP333-GP , red; group 3 , four NHPs , 1×107 FFU BNSPΔG-GP , blue; group 4 , four NHPs , 250 µg purified INAC-BNSP333-GP , green . We followed the immune response of the vaccinated animals over time after vaccination as well as after challenge ( Figure 1B ) . Notably , the goal of this novel vaccine approach was to develop a vaccine that protects from two different highly lethal diseases , rabies and filovirus induced hemorrhagic fever . Therefore , we followed both RABV and EBOV GP-specific immune responses . As shown in Figure 2A , all three vaccines and the empty control vector induced seroconversion against RABV G as early as day 7 after immunization , with increasing IgG levels at day 14 and slightly decreasing levels for the sera collected at day 28 . In contrast , at day 7 , EBOV GP-specific humoral responses were only detected in sera from animals vaccinated with the replication-deficient vaccine ( Figure 2B ) . On day 14 , all groups ( groups 2–4 ) showed a positive signal in the EBOV GP-specific ELISA whereas for the control animals ( group 1 ) only background signals were detected ( Figure 2B ) . Interestingly , the replication-deficient ( RABV G-deleted ) vector expressing EBOV GP induced the highest EBOV GP-specific responses but the lowest RABV G responses . This is most likely due to the fact that this virus does not encode RABV G and the RABV G-specific immune response results from the G protein contained in the initial vaccine particle preparation [27] . Our previous research on live RABV-based vaccines indicated that pre-existing anti-RABV antibodies prevent a successful secondary immunization; therefore , only the group of rhesus macaques primed with the inactivated RABV virions containing EBOV GP received a boost at day 28 with the same vaccine ( Figure 1B ) . The boost increased the humoral responses against EBOV GP as well as RABV G for group 4 significantly ( Figure 2A and B , day 35 and 42 ) . Remarkably , even the animals that were not boosted showed an increase in humoral responses directed against RABV G and EBOV GP from days 28 to 35 , indicating that the vaccines were still stimulating the immune system . High ELISA titers of anti-RABV G antibodies are predictive of protection of the immunized host , but we still wanted to confirm the humoral response against RABV by virus neutralization assays ( VNA ) . The result presented in Figure 2C indicates that all four vaccines induced virus-neutralizing antibodies as early as 7 days after immunization . Notably , the VNA titers were well above the critical level of 0 . 5 international units ( I . U . ) , which is considered protective from RABV infection in humans [33] . Similar to the total IgG levels against RABV G , we detected an increase of the RABV-specific immune responses by VNA for all vaccine groups . In contrast to RABV , we were unable to detect significant levels of virus-neutralizing antibodies directed against EBOV for groups 2–4 compared to the controls of group 1 ( data not shown ) . We also analyzed the cellular responses utilizing an IFN-γ specific ELISPOT from larger blood samples collected at day 14 and 42 . As shown in Table 1 , animals of the control group did not mount any cellular responses when stimulated with EBOV GP-specific peptide pools . The highest responses were detected for animals immunized with the replication-competent vaccine , followed by the replication-deficient and the killed viral particles at day 14 . However , in each group we failed to detect cellular responses in one or two animals . At day 42 , cellular responses were only detected in two animals , which previously had the highest responses . All other animals showed only a background level of EBOV GP-specific cellular responses . After day 42 , all animals were transferred to the NIAID BSL-4 facility at the Rocky Mountain Laboratories for EBOV challenge . Since the challenge virus stock had never been utilized in rhesus macaques , we infected two of the three control animals ( NHP1 and NHP2 ) on day 56 with 1000 PFU of EBOV ( strain Mayinga ) prior to the other animals to ensure the pathogenicity of the virus stock . NHP1 and NHP2 rapidly developed disease and reached the hemorrhagic state ( rash ) on day 6 and 7 post challenge , respectively , at which point animals had to be euthanized according to approved protocol . Based on this finding , we infected the remaining 13 animals on day 70 with the same challenge virus stock and dose . For each challenge experiment , physical exams and blood draws were performed at day 0 , 3 , 6 , 12 , 16 , 22 , and 28 post-challenge . The outcome of the EBOV challenge and the clinical findings are presented in Figure 3 . As show in Figure 3A , all animals immunized with the live replication-competent vaccine , BNSP333-GP ( group 2 ) , survived the challenge . As expected , all animals in the control group ( NHP 1–3 ) had to be humanely euthanized according to approved protocol mainly based on high viremia at day 6 ( Figure 3C ) and rash . Two of four animals from group 3 , which were immunized with the replication-deficient vaccine ( NHP8 and NHP9 ) , and two of four animals from group 4 , which received the inactivated RABV/EBOV particles ( NHP14 and NHP15 ) , had to be euthanized . The viral loads detected in the blood of these animals ( NHP8 , NHP9 , NHP14 and NHP15 ) did not differ from those of the three control animals NHP1-3 ( Figure 3C ) . Interestingly , five out of the twelve animals ( NHP4 , NHP5 , NHP6 , NHP11 and NHP12 ) controlled challenge virus replication with undetectable viremia , whereas for three animals ( NHP7 , NHP10 and NHP13 ) the challenge virus was detected transiently at one ( group 2 and 3 ) or two time points ( group 4 ) ( Figure 3C ) . The lack of protection was also reflected clinically ( Figure 3B , D–F ) . Animal body temperature , on average , increased within the first few days after challenge but returned to the original temperatures near day 12 post challenge ( Figure 3B ) . Platelet count significantly decreased after challenge for the unprotected animals whereas protected animals regained normal platelet levels by day 12 post challenge ( Figure 3D ) . Serum alanine aminotransferase ( ALT ) ( Figure 3E ) and serum aspartate aminotransferase ( AST ) levels ( Figure 3F ) monitored liver function . Elevated levels , as seen by days 3–6 , indicated liver damage as a result of EBOV infection . We also monitored the immune responses of vaccinated animals after challenge during the acute and convalescent phases of disease . As shown in Figure 4 , the three animals of the control group failed to mount any EBOV GP-specific responses during the course of the EBOV infection . In contrast , all three groups of vaccinated NHPs ( groups 2 , 3 and 4 ) had similar levels of anti-EBOV GP antibodies at the day of challenge ( Figure 4A , challenge day 0 ) . The EBOV GP-specific antibodies remained at these levels 3 days after challenge ( Figure 4B , challenge day 3 ) . Interestingly , a rapid increase to high levels of EBOV GP-specific IgG was observed on day 6 after challenge in the serum of the five animals that survived ( NHP4 , NHP5 , NHP6 , NHP11 , and NHP12 ) . Notably , these high antibody levels were detected in those animals that did not have detectable levels of EBOV RNA in their blood during the course of challenge . The three animals ( NHP7 , NHP10 , and NHP13 ) , which did show a lower but significant increase in EBOV GP-specific antibodies 6 days after challenge , demonstrated transient levels of EBOV viremia but survived ( figure 3C ) . Lastly , we failed to detect an increase in the humoral responses against EBOV GP in the serum of four animals on day 6 post challenge ( NHP8 , NHP9 , NHP14 , and NHP15 ) . These were the animals that had to be humanely euthanized . Based on these results , we concluded that one requirement for a successful RABV-based EBOV vaccine is the rapid recall response of humoral immunity against EBOV GP after EBOV challenge . The results presented above also indicate that virus specific antibodies are important to control EBOV infection . However , the antibody titers against EBOV GP at the day of challenge were similar for all three groups and , more importantly , the same within groups 3 and 4 where two animals in each group were protected while the other two were not . Therefore , we decided to analyze the humoral immune responses in greater detail . First , we performed an EBOV GP specific ELISA utilizing full-length EBOV GP as well as a “mucin-like domain” ( MLD ) deleted version ( EBOV GP-ΔMLD ) . The MLD is a heavily glycosylated region of the EBOV GP ectodomain . Previous research indicates that antibodies directed against the MLD not only fail to neutralize EBOV but can even enhance the infection [34] . Results shown in Figure 5 demonstrate that similar humoral responses were detected for sera from all animals utilizing both full-length EBOV GP ( Figure 5A ) and EBOV GP-ΔMLD ( Figure 5B ) ELISAs . We concluded from these findings that there are no significant differences in the target of the induced antibodies for these three vaccines , at least not in regard to the MLD . As there seem to be no obvious differences in the EBOV GP-target of the antibodies within the different groups of vaccines , we analyzed qualitative differences of the antibodies . Th1 dependent IgG1 antibodies in NHPs have been shown to possess better antiviral properties by mediating antibody-dependent cellular cytotoxicity and complement activation [35] . Therefore , we performed an isotype-specific ELISA utilizing full-length EBOV GP to determine if any differences in the antibody isotypes occurred between the vaccines . As shown in Figure 6 , the positive control NHP ratio of IgG2/IgG1 on day 0 post-challenge was around 1 . 0 , whereas this ratio was an average of about 2 . 0 for all the vaccinated animals from groups 2–4 . However , six days after challenge this IgG2/IgG1 ratio changed to ∼1 . 0 for the sera collected for the animals from group 2 , which is the group where all four animals survived . In contrast , the average of the IgG2/IgG1 ratio for group 3 was 2 . 0 and for group 4 was 1 . 5 . In both of these groups two animals were not protected . These data suggest that IgG1-biased humoral responses might be beneficial for the control of EBOV infection . This contention was further supported by the analysis of the individual animals in groups 2 and 4 . As shown in Figures 3 and 6 , NHP7 ( group 2 ) had detectable EBOV RNA in the blood on day 6 and also had the highest IgG2/IgG1 ratio of 1 . 6 . The same is true for group 4 where both protected animals had an IgG2/IgG1 ratio below 1 . 0 , whereas the unprotected animals were above 2 . 0 . The only exception from this observation is one protected animal in group 3 ( NHP10 ) , which had a clear bias towards an IgG2 response indicated by an IgG2/IgG1 of almost 3 . 0 . However , this animal had a very low antibody response even on day 6 after challenge ( Figure 4 ) . Lastly , the final outcome at four weeks after challenge was similar for all surviving animals with an IgG1-biased response and an IgG2/IgG1 ratio of about 0 . 8 ( similar to the positive control ) . In general , this novel finding strongly indicates that an IgG1-biased immune response against EBOV GP is advantageous for protection and should be further evaluated with larger animal numbers and for other vaccine approaches . Lastly , qualitative antibody differences were also analyzed by measuring the avidity of the antibodies before , during , and after challenge in the vaccinated animals of groups 2–4 . As shown in figure 7 on day 42 , the avidity of the antibodies in the serum of the vaccinated animals was similar ( group 2 ) or below that of the control animals ( group 3 and group 4 ) . However , the avidity of EBOV GP-specific antibodies significantly increased until challenge day 0 and was above the level of the control animals for most of the vaccines , indicating that RABV-vector induced immune responses mature over a long period of time . However , we did not find a direct correlation between avidity of the EBOV GP-specific antibody induced by the vaccines and protection from disease . Two additional interesting findings are worth mentioning . As listed above , it seems that the avidity of the EBOV GP-specific immune response is not complete until after day 42 , since the levels continue to increase at least until challenge day 0 ( also referred to as day 70 ) , which is the last time point analyzed before the challenge with EBOV ( strain Mayinga ) . Secondly , and even more interestingly , challenge with EBOV does increase the avidity of the antibodies directed to EBOV GP ( see Figure 7 , challenge day 0 and post challenge day 3 ) . In order to investigate if this is an antigen-specific effect , we analyzed the avidity of the RABV G-specific antibodies and observed a similar increase in avidity of the RABV G-specific antibodies compared to EBOV GP-specific antibodies , which was also confirmed by VNA ( supplemental Figure S1 ) . The VNA showed a significant increase in the surviving animals of about 3 to 4-fold . Even though we can only speculate about the mechanism of protection , there is a clear increase of the immune response even against an antigen ( e . g . RABV G ) not expressed during the challenge virus infection .
We have previously described the generation , propagation , safety , immunogenicity , and protective efficacy of RABV/EBOV vaccine candidates [18] , [36] , [37] . Two live vaccine candidates , BNSP333-GP and BNSPΔG-GP with a deletion of the entire RABV G gene , were found to be avirulent upon peripheral administration in mice . Based on the efficient incorporation of EBOV GP into the virion , an inactivated vaccine ( INAC-BNSP333-GP ) was also produced by treatment of the RABV/EBOV vaccine with beta-propiolactone , the standard method utilized for the current human RABV vaccine . Each bivalent vaccine candidate induced strong humoral immunity to RABV G and EBOV GP , and conferred protection from both lethal RABV and mouse-adapted EBOV challenge in mice . Based on the demonstration of promising safety , immunogenicity , and protective efficacy of the live and inactivated RABV/EBOV vaccines in mice , we sought to evaluate these vaccines in nonhuman primates . All three vaccines did not induce any clinical sings including fever or weight loss after vaccination and we were not able to detect any of the vaccine vectors in the blood of the vaccinated NHP by RT-PCR ( data not shown ) . However , further safety studies are necessary for the replication-competent vector to analyze any impact of EBOV GP for this vaccine . Immunogenicity was examined following challenge with EBOV ( strain Mayinga ) . Each vaccine candidate was found to induce potent humoral immunity and 50% to 100% protection from lethal challenge . Our results indicate that the protection of immunized animals was largely dependent on the induced humoral immune response against EBOV GP . This finding is not too surprising because acute viral infections are often controlled by antibodies rather than cytotoxic T-cells , which in general are more important for the control of chronic infections [38] . However , previous research did suggest CD8+ T-cells as the major player for protection from EBOV infection in a single vaccination strategy [13] . This has been challenged by recent studies indicating that , in general , protection of NHPs by different EBOV vaccines seems to depend on the presence of anti-EBOV GP antibodies as well as EBOV GP-specific CD4+ T-helper cells [9] , [12] . Moreover , it cannot be excluded that CD8+ T-cells play a role in viral clearance but no CD8+ memory T-cells are needed . Lastly , the relatively high background in our ELISPOT assay might have prevented us to detect low cellular responses . In any case , our results indicated a major role of EBOV GP-specific antibodies to control the challenge virus replication , and we therefore focused on these responses in greater detail . In this regard , the finding that only 50% of NHPs in groups 3 and 4 were protected was an advantage , because all three groups of vaccinated NHPs did have similar levels of anti-EBOV GP antibodies as analyzed using an EBOV GP-specific ELISA . This suggested a qualitative difference in the humoral responses for the three vaccines . First we investigated if the anti-EBOV GP antibodies were directed against different regions within the glycoprotein . Previous work by others indicates that antibodies directed against the MLD within EBOV GP can enhance the infection with EBOV [34] . Therefore , we analyzed the anti-EBOV GP humoral responses from all three vaccines utilizing full-length EBOV GP and EBOV GP-ΔMLD . However , there was no significant difference in the ELISA signal for each serum sample to the two versions of EBOV GP , and we concluded that MLD-directed antibodies do not explain the difference in the observed protection . Secondly , we analyzed if we could detect differences in the avidity of the antibodies for the three groups of vaccinees , but such differences were not detected . Surprisingly , we found that the avidity of the anti-EBOV GP-specific antibodies greatly increased during challenge at a similar percentage for all tested sera . Whereas we cannot explain this increase in antibody avidity , it was transient and not specific to antibodies against EBOV GP , as the avidity of RABV G-specific antibodies also increased . It is well established that EBOV GP-specific antibodies induced by different EBOV vaccine candidates may have no or only weak VNA activity , but are still protective [8] , [9] . These findings indicate that antibody-dependent cell-mediated cytotoxicity ( ADCC ) might play a major role in protection from EBOV infection . Because ADCC depends on IgG1 antibody responses , we analyzed the total IgG response and also the IgG subtype responses ( e . g . IgG1 and IgG2 ) . Our results indicated that all protected animals , independent of the group , had an IgG2/IgG1 ratio of ∼0 . 8 whereas the unprotected animals had a higher ratio of ∼2 . 5 . Interestingly , two animals from group 3 and group 4 , NHP10 and NHP13 , transiently had the highest viral loads and also had IgG2/IgG1 ratios that fell between those of unprotected and protected animals . These results clearly indicate that the quality of the antibodies in regard to the isotypes is very important for a successful EBOV vaccine based on EBOV GP . However , the protected animals from group 2 and 4 had also the highest total antibody levels and therefore we conclude that in the case of a rabies-based EBOV vaccine , high level of GP-specific antibodies that are IgG1-bias are very likely to be significant as shown for groups 2 and 4 in Figure 6B . In summary , the results presented in this study clearly indicate that the RABV-based vector induced an immune response sufficient to protect from lethal EBOV infection . In the case of replication-competent RABV vectors expressing EBOV GP , no further improvements are necessary and such a vector could be used to protect NHPs from EBOV in the endemic setting . Of note , it would be best to establish efficacy via an oral application , which is already well established for live RABV in wildlife [30] , [31] . The replication-deficient and the inactivated RABV particles did not protect all animals; therefore , the responses induced by these vaccines need to be improved to resemble the responses detected for the replication-competent vaccine , BNSP333-GP , which did protect 100% of the animals . In any case , for the replication-deficient vector , the virus could be concentrated so higher titers such as the once used for the replication competent vaccine can be used for the immunization . We cannot exclude the possibility that using a five-fold lower titer for the immunizations than was used for BNSP333-GP was responsible for this difference in protection . Moreover , another vector choice could be the matrix protein ( M ) -deleted replication-deficient RABV vector expressing EBOV GP . Studies with such a vector as a RABV vaccine indicated that it is superior , even compared to a replication-competent RABV [39] . In the case of the inactivated virions containing EBOV GP , a new construct containing an exact fusion of the RABV G cytoplasmic domain to EBOV GP increased the incorporation level about two-fold ( the previous construct contained two foreign amino acids between the EBOV GP transmembrane domain and the RABV G cytoplasmic domain ) and showed better responses than the current construct in mice ( Willet and Schnell , unpublished data ) . Moreover , we discovered that the glycosylation pattern of EBOV GP was different for RABV particles grown on BSR cells than particles grown on VERO cells ( data not shown ) . Therefore , we believe that the utilization of particles containing higher levels of EBOV GP and perhaps an additional immunization dose would bring the protection rate to 100% of the animals , a reachable goal for a safe and promising dual vaccine .
This study was carried out in strict accordance with the recommendations described in the Guide for the Care and Use of Laboratory Animals of the National Institute of Health , the Office of Animal Welfare and the United States Department of Agriculture . All animal work was approved by the NIAID Division of Intramural Research Animal Care and Use Committees ( IACUC ) , in Bethesda , MD ( protocol # OSD-28 ) and at the Rocky Mountain Laboratories ( RML , protocol # 2012-004-E ) . Both facilities are accredited by the American Association for Accreditation of Laboratory Animal Care . All procedures were carried out under Ketamine anesthesia by trained personnel under the supervision of veterinary staff and all efforts were made to ameliorate the welfare and to minimize animal suffering in accordance with the “Weatherall report for the use of non-human primates” recommendations . Animals were housed in adjoining individual primate cages allowing social interactions , under controlled conditions of humidity , temperature and light ( 12-hour light/12-hour dark cycles ) . Food and water were available ad libitum . Animals were monitored twice daily ( pre- and post-challenge ) and fed commercial monkey chow , treats and fruit twice daily by trained personnel . Early endpoint criteria , as specified by the RML IACUC approved score parameters , were used to determine when animals should be humanely euthanized . Fifteen RABV and EBOV seronegative rhesus macaques were assigned to four groups to evaluate the recently developed RABV/EBOV vaccine candidates ( Figure 1 ) . Groups of four animals were used for the vaccine candidate groups while a group of three animals served as the negative control . On day 0 , group 1 ( control ) animals were immunized intramuscularly ( i . m . ) in the caudal thigh with a 5×107 FFU dose of live parent RABV vaccine , BNSP333 . Group 2 animals were immunized i . m . with a 5×107 FFU dose of the full length parent RABV vaccine expressing EBOV GP ( designated as BNSP333-GP ) . Group 3 animals were immunized i . m . with a 1×107 dose of the parent vaccine expressing EBOV GP but containing a deletion in the rabies glycoprotein gene ( designated BNSPΔG-GP ) . Group 4 animals were immunized i . m . with 250 µg of beta-propiolactone inactivated BNSP333-GP ( designated INAC-BNSP333-GP ) . Group 4 was boosted with 250 µg inactivated virus on day 28 . All macaques were bled on days 0 , 1 , 3 , 5 , 7 , 14 , 28 , 35 , and 42 before transport from the National Institutes of Health Animal Center ( Poolesville , MD ) to the National Institutes of Health , Rocky Mountain Laboratories ( Hamilton , MT ) . Since the challenge virus stock had never been utilized in rhesus macaques , we infected two of the three control animals ( NHP1 and NHP2 ) on day 56 with 1000 PFU of EBOV ( strain Mayinga ) . Based on finding that the utilized virus was virulent and caused EBOV hemorrhagic disease , we infected the remaining 13 animals on day 70 with the same challenge virus stock and dose . For each challenge experiment , physical exams and blood draws were performed on day 0 , 3 , 6 , 9 , 12 , 16 , 22 , and 28 post-challenge . Serum aliquots treated by gamma-irradiation as per approved protocol were sent to Thomas Jefferson University for analysis by enzyme-linked immunosorbent assay ( ELISA ) . The total white blood cell count , lymphocyte , platelet , reticulocyte and red blood cell counts , hemoglobin , hematocrit values , mean cell volume , mean corpuscular volume , and mean corpuscular hemoglobin concentrations were determined from EDTA blood with the HemaVet 950FS+ laser-based hematology analyzer ( Drew Scientific , Waterbury , CT ) . Serum biochemistry was analyzed using the Piccolo Xpress Chemistry Analyzer and Piccolo General Chemistry 13 Panel discs ( Abaxis , Union City , CA ) . Levels of viral RNA were determined using quantitative RT-PCR ( qRT-PCR ) as described previously [12] . For determination of virus titers in NHP blood and tissue samples , Vero E6 cells were seeded in 48-well plates the day before titration . Blood samples were thawed and serial dilutions were prepared . Tissues were homogenized in 1 ml plain DMEM and , as with the blood , serial 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 . To evaluate T- cell responses to EBOV GP , NHP PBMCs were tested using the NHP IFNγ ELISPOT Kit ( R&D Systems , Cat#EL961 ) as per the manufacturer's instructions . Briefly , microplates were filled with 200 µl per well of sterile culture media ( RPMI-1640 , 10% FBS , 1% Penstrep ) as blocking media and incubated at room temperature while NHP PBMCs and stimulating antigens were prepared for plating . Antigens were prepared in sterile culture media to achieve final concentrations as follows: GP peptide pool ( JPT ) at 10 µg/ml; Pokeweed Mitogen ( positive control ) at 1 µg/ml; Influenza NP peptide ( Mimotopes , negative control ) at 10 µg/ml . Unstimulated cells were used to normalize spot counts to background levels . Blocking media was removed , and antigen was added respectively . PBMCs were added to respective wells at 1×105 cells/well . Plates were incubated at 37°C , 5% CO2 for 48 h . Cells were then removed , and plates were washed four times with Wash Buffer ( R&D Systems ) . Plates were stained and developed according to R&D Systems protocol with Detection Antibody , Streptavidin-AP and BCIP/NBT chromogen . Plates were rinsed with deionized water and allowed to dry completely before scanning and counting using a CTL Immunospot Reader . A modified rapid fluorescent focus inhibition test ( RFFIT ) was performed to determine RABV neutralizing antibody levels in the immunized NHP sera . Three-fold serial dilutions of sera or WHO standard RABV IgG in Cellgro Complete serum free media ( Mediatech ) were incubated at 37C for 1 h with BNSP ( parent RABV of BNSP-333 that does not have the attenuating mutation at position 333 ) at a concentration to achieve an moi of 1 at 24 h post-infection in the negative control . Then , the mixture was added to one day old BSR cells ( BHK-21 derived cell line ) that had been grown in DMEM ( Mediatech ) supplemented with 10% FBS ( Atlanta Biologicals ) and 1% penicillin/streptomycin ( Mediatech ) on 96 well plates , and plates were incubated for 24 h at 34C . Plates were then fixed with 80% acetone and stained with anti-RV N ( Fujirebio ) . Plates were read for percent infected cells per well , and IUs of antibody were calculated based on the WHO standard , where 50% infection accounts for 2 IU . Neutralizing antibody titers were determined by performing focus reduction neutralization titration assays ( FRNT ) as described previously [12] . Briefly , Vero E6 cells were seeded into 96 well plates to generate a confluent monolayer on the day of infection . Serum dilutions were prepared in plain DMEM and 25 µl were incubated with 200 ffu EBOV expressing green fluorescence protein ( EBOV-GFP ) in a total volume of 50 µl . After 60 min at 37°C the media was removed from cells , the serum-virus mixture was added and samples were incubated for 60 min at 37°C . Then the mixture was removed from the cells and 100 µl of 1 . 2% carboxymethyl cellulose in MEM ( 2% FBS ) was added per well and left for 4 days at 37°C . The neutralizing antibody titer of a serum sample was considered positive at a dilution showing a>80% reduction ( FRNT 80 ) in GFP-foci compared to the control without serum . Sub-confluent T175 flasks of 293T cells ( human kidney cell line ) were transfected with a hemagglutinin ( HA ) tagged EBOV GP expression plasmid encoding amino acids 33–632 of the EBOV GP ectodomain ( EBOV GP-ΔTM ) or a truncated version that lacks amino acids 312–462 of the EBOV GP mucin-like domain ( EBOV GP-ΔMLD-ΔTM ) . Both plasmids were kindly provided by Erica O . Saphire of the Scripps Research Institute , La Jolla , CA . Supernatant was added to an equilibrated anti-HA agarose ( Pierce ) column containing a 2 . 5 mL agarose bed volume . The column was washed with 10 bed volumes of TBST ( TBS containing 0 . 05% Tween 20 ) and 2 bed volumes of TBS before adding 5 mL of 200 µg/mL HA peptide in TBS ( Pierce HA peptide ) . The peptide was added at a flow rate of 500 µL/min and incubated overnight at 4°C . Bound EBOV GP was eluted with 3 mL of 200 µg/mL HA peptide in TBS . Fractions were collected and analyzed for EBOV GP via Western blot with a nitrocellulose membrane and monoclonal anti-HA antibody ( Sigma ) prepared in 5% BSA/TBST and goat anti-mouse IgG-HRP . EBOV GP positive fractions were dialyzed with 10K MWCO dialysis cassettes ( Thermo Scientific ) to remove excess HA peptide used to elute the HA-tagged EBOV protein . Rhesus macaque sera obtained from the NIH were tested to analyze the humoral response to EBOV GP and RABV G . EBOV GP antigen for coating plates was obtained by harvesting supernatant from transfected 293T cells and purifying the secreted protein with an anti-HA agarose column as described above . 96-well plates ( Nunc , Immulon 4 HBX ) were coated overnight at 4°C with 50 ng/well purified EBOV GP or 100 ng/well purified RABV G in Na2CO3 coating buffer . Plates were washed three times with PBST ( PBS with 0 . 025% Tween 20 ) and blocked at room temperature for 1–2 hours with 5% dry non-fat milk in PBST . Serum samples were diluted 1∶50 in 0 . 5% BSA-PBST and 100 uL was added to each well in a 1∶3 serial dilution . Plates were incubated overnight at 4°C , washed three times with PBST , and incubated for 2 hours with 100 uL/well of goat anti-human IgG-HRP . Plates were washed with PBST and developed with 200 uL/well of SigmaFast o-phenylenediamine dihydrochloride ( OPD ) substrate . After incubating for 5 minutes at room temperature , the reaction was stopped with 50 uL of 3 M H2SO4 and the absorbance was read at 490 nm . IgG subclass specific ELISAs were performed for EBOV GP and RABV G with anti-human ( Abcam ) and anti-rhesus ( NIH NHP Reagent Source ) antibodies . Plates were incubated with OPD substrate for 8–13 minutes before stopping the reaction with 3 M H2SO4 . Macaque sera were measured for total IgG avidity to RABV G and EBOV GP using a sodium thiocyanate ( NaSCN ) displacement ELISA to determine the concentration of NaSCN needed to dissociate 50% of the antibody-antigen interactions . The avidity assays were set up similar to the ELISA protocol described above however the sera samples were diluted to the concentration that would yield an OD reading of 0 . 8 nm . Prior to incubation with the secondary antibody , the plates were treated with increasing concentrations of NaSCN in PBS ( 0 M , 1 M , 2 M , 3 M , 4 M , 5 M , 6 M ) for 15 minutes at room temperature . Wells receiving 0 M NaSCN were incubated with PBS . The plates were immediately washed three times with PBST ( 0 . 025% Tween in PBS ) before continuing with the ELISA protocol . All avidity assays were performed in triplicate . All data were analyzed by Prism software ( GraphPad , version 5 . 0 d ) . Statistical analysis was performed using unpaired t-test with Welch's correction to compare two groups and represented as two-tailed p-value with a confidence interval of 95% . Presented results show the mean of measurements within a group . For all statistics , the following notations are used to indicate significance between two groups: *p<0 . 05 , **p<0 . 01 , ***p<0 . 001 . | Ebola virus ( EBOV ) has been associated with outbreaks in human and nonhuman primate populations since 1976 . With a fatality rate approaching 90% , EBOV is one of the most lethal infectious diseases in humans . The increased frequency of EBOV outbreaks along with its potential to be used as a bioterrorism agent has dramatically strengthened filovirus vaccine research and development . While there are currently no approved vaccines or post exposure treatments available for human use , several vaccine candidates have shown to protect nonhuman primates from lethal EBOV challenge . Our primary focus is to develop vaccine candidates to protect humans and endangered wildlife species at risk of infection in Africa . Here , we evaluated the efficacy and immunogenicity of our dual vaccines against EBOV and rabies virus ( RABV ) in rhesus macaques . Our live replication-competent vaccine provided 100% protection following EBOV challenge while the replication-deficient and inactivated candidates provided 50% protection . Interestingly , protection is dependent on the quality of the antibodies rather than the quantity . All three RABV-based EBOV vaccines did induce antibody levels necessary for protection from RABV infection . These results encourage the further development of these novel dual vaccines directed against two of the most lethal viral diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"viral",
"vaccines",
"rna",
"viruses",
"emerging",
"infectious",
"diseases",
"immunity",
"viral",
"classification",
"virology",
"host-pathogen",
"interaction",
"biology",
"microbiology"
]
| 2013 | Antibody Quality and Protection from Lethal Ebola Virus Challenge in Nonhuman Primates Immunized with Rabies Virus Based Bivalent Vaccine |
Dengue viruses 1–4 ( DENV1-4 ) rely heavily on the host cell machinery to complete their life cycle , while at the same time evade the host response that could restrict their replication efficiency . These requirements may account for much of the broad gene-level changes to the host transcriptome upon DENV infection . However , host gene function is also regulated through transcriptional start site ( TSS ) selection and post-transcriptional modification to the RNA that give rise to multiple gene isoforms . The roles these processes play in the host response to dengue infection have not been explored . In the present study , we utilized RNA sequencing ( RNAseq ) to identify novel transcript variations in response to infection with both a pathogenic strain of DENV1 and its attenuated derivative . RNAseq provides the information necessary to distinguish the various isoforms produced from a single gene and their splice variants . Our data indicate that there is an extensive amount of previously uncharacterized TSS and post-transcriptional modifications to host RNA over a wide range of pathways and host functions in response to DENV infection . Many of the differentially expressed genes identified in this study have previously been shown to be required for flavivirus propagation and/or interact with DENV gene products . We also show here that the human transcriptome response to an infection by wild-type DENV or its attenuated derivative differs significantly . This differential response to wild-type and attenuated DENV infection suggests that alternative processing events may be part of a previously uncharacterized innate immune response to viral infection that is in large part evaded by wild-type DENV .
Dengue viruses 1–4 ( DENV1-4 ) are the world's most prevalent arthropod-borne viruses [1] . DENVs are responsible for an estimated 50–100 million cases of debilitating or life-threatening infection every year and an estimated 2 . 5 billion people in over 100 endemic countries are at risk of infection [1] , [2] . The economic impact of DENVs has been estimated to be as high , if not higher than other major global health menaces such as malaria , tuberculosis , hepatitis , bacterial meningitis and others [3]–[8] . Despite the considerable health and economic impact , there are as yet no licensed vaccines or antiviral drugs to combat DENVs and an incomplete understanding of the biology of DENV infection has hampered progress on both of these fronts . Given the limited coding capacity of their ∼11 kb RNA genome , DENVs must parasitize the host cell machinery to complete their life cycle . At the same time , these viruses must effectively evade or suppress the host responses that act to restrict their replication [9]–[11] . This interplay between host and virus and the effect it has on host gene expression has been described previously [12]–[29] . Largely uncharacterized , however , is whether the transcriptional start site ( TSS ) and post-transcriptional variations of host RNA , leading to the production of different gene isoforms , may play a role in DENV infection . Differential RNA processing is known to be a major factor underlying cellular and functional complexity [30] , [31] . In order to interrogate TSS and post-transcriptional RNA variations across the entire genome in response to DENV infection , we harnessed the power of RNA sequencing ( RNAseq ) . RNAseq is a recently developed approach to transcriptome profiling that permits a precise quantification of RNA levels and their alternatively processed variants by means of high throughput , massively parallel sequencing and subsequent mapping of the resultant short sequence fragments onto a reference genome [32] , [33] . We utilized two strains of DENV1 in our RNAseq study to identify strain-specific TSS and post-transcriptional variations in response to infection . The first strain , DENV1-16007 , was isolated from the serum of a patient in Thailand in 1964 . The second strain of DENV1 used in this study is an attenuated derivative of DENV1-16007 . This attenuated virus , DENV1-PDK13 was passaged 13 times in primary dog kidney cells and was shown to be immunogenic but minimally reactogenic in human volunteers [34] , [35] . Our RNAseq data indicate that significant differences exist between these two strains of DENV1 , not only at the transcript level but also at the level of alternative splicing . Similar trends were observed in RNAseq of an additional two low-passaged DENV1 clinical isolates . These findings suggest that subversion of the host response includes TSS and post-transcriptional modification and is part of the mechanism of virulence . These findings also suggest that variations in the viral genome can have a profound effect in modifying host response to infection .
HuH7 , C6/36 and BHK-21 cells were purchased from the American Type Culture Collection ( ATCC ) and cultured according to ATCC recommendation . DENV1 strains 16007 and PDK13 were obtained from the Division of Vector-borne Diseases , Centers for Disease Control and Prevention . Sequence analysis in our laboratory indicates that these strains match the published sequences for these viruses ( GenBank accession numbers AF180817 . 1 and AF180818 . 1 , respectively ) . These viruses were amplified three times in C6/36 cells prior to use in the current study . Additionally , the DENV1 clinical isolates EDEN3300 and SL107 , which were obtained from previously reported studies [36] , [37] and passaged in C6/36 cells for <5 times were also included . The supernatant of all virus cultures were harvested six days post infection , clarified by centrifugation at 450× g for 10 min at 4°C , filtered and concentrated by centrifugation at 30 , 000× g for 3 hrs at 4°C . Virus pellets were re-suspended in DMEM medium with 2% FBS ( Invitrogen ) and stored at −80°C until use . Infectious titer was determined by plaque assay as described previously [38] . HuH7 cells were seeded at 3×106 per flask in 25 mm flasks and incubated at 37°C for 24 hrs before infecting at a MOI of 20 with each of the DENV strains for 1 hr at 37°C/5% CO2 , with gentle rocking every 15 min . The cells were then washed thoroughly and replaced with DMEM medium supplemented with 2% FBS and incubated for 20 hrs . Immunofluorescence assay was conducted according to a previously described method [39] . Briefly , the cells from the virus culture were washed once and re-suspended with PBS , and spotted onto a Teflon coated glass slide , air dried and then immersed in 80% acetone for 10 min . The slide was rinsed with PBS and air-dried . 2 µl antibody against prM protein ( 2H2 monoclonal antibody ) was added onto each well , incubated at 37°C for 45 min in a humidified chamber , and washed twice with PBS before drying . FITC-conjugated goat anti-mouse IgG were diluted 1∶30 with 0 . 1% Evan's Blue and 2 µl was added onto each well . Slides were then incubated at 37°C for 45 min in the humidified chamber and then washed twice with PBS . Slides were dried and mounted with buffered glycerol before imaging under a fluorescent microscope . Polyadenylated mRNA was isolated from HuH7 cells by three rounds of selection with the Dynabeads mRNA Direct Kit ( Invitrogen ) and assessed by electrophoresis on the Bioanalyzer 2100 ( Agilent ) for quality evaluation . For the RNAseq sample preparation , the NEBNext mRNA Sample Prep Master Mix Set 1 was used according to the manufacturer's protocol ( NEB ) . Briefly , 0 . 5 ug mRNA was used for fragmentation and then subjected to cDNA synthesis using SuperScript III Reverse Transcriptase ( Invitrogen ) and random primers . The cDNA was further converted into double stranded cDNA and after an end repair process ( Klenow fragment , T4 polynucleotide kinase and T4 polymerase ) , was ligated to Illumina paired end ( PE ) adaptors . Size selection was performed using a 2% agarose gel , generating cDNA libraries ranging in size from 275–325 bp . Finally , the libraries were enriched using 15 cycles of PCR and purified by the QIAquick PCR purification kit ( Qiagen ) . Libraries were sequenced on an Illumina GAIIx machine at the National Cancer Center , Singapore ( Control-NCC , 16007-NCC , PDK13-NCC ) or an Illumina HiSeq 2000 machine at the Duke-NUS Genome Biology Facility , Singapore ( Control-1 , Control-2 , 16007-1 , 16007-2 , PDK13-1 , PDK13-2 ) . Resulting reads were mapped to the hg19 build of the human genome using Tophat v1 . 3 . 0 ( http://tophat . cbcb . umd . edu/index . html ) with the coverage-search , microexon-search and butterfly-search options . Differential isoform expression analysis was done using Cufflinks v1 . 3 . 0 ( http://cufflinks . cbcb . umd . edu/ ) with the multi-read-correct ( Cufflinks ) , -r and -s ( Cuffcompare; using the same annotation gtf and hg19 fasta files respectively as in Tophat ) and the mask-file ( rRNA ) , frag-bias-correct ( same hg19 fasta file used for Tophat ) and multi-read-correct options . Differential splicing analysis was done using MISO v1 . 0 ( http://genes . mit . edu/burgelab/miso/docs/ ) using the default options . Analysis of RNA sequencing quality was performed with RNAseqC v1 . 7 ( http://www . broadinstitute . org/cancer/cga/rna-seqc ) using the default options . Creation of proportional Venn diagrams was done with freeware available at www . venndiagram . tk . Hierarchical clustering of differentially regulated isoforms was done with Partek v6 . 6 ( http://www . partek . com/ ) . Pathway analysis of differentially regulated isoforms was done using Ingenuity Pathway Analysis v9 . 0 ( http://www . ingenuity . com/ ) . Total RNA derived from mock-infected and DENV-infected cells was used to synthesize cDNA using SuperScript III First Strand Synthesis System ( Invitrogen ) with random hexamers according to manufacturer's instructions . Quantitative real-time PCR was performed using LightCycler 480 Real-Time PCR System ( Roche Diagnostics GmbH , Germany ) and LightCycler 480 SYBR Green I Master ( Roche Diagnostics GmbH , Germany ) . The reaction was carried out to simultaneously amplify exon-skipped and exon-included isoforms using specific primers complementary to the exons flanking each target exon ( Table S13 ) . Percent exon exclusion levels were calculated as the percentage of the isoform excluding an alternative exon divided by the total abundance of the isoforms including and excluding the alternative exon . The statistical analysis employed to analyze biological triplicates for each condition are as previously described using the software Cufflinks and Mixture-of-Isoforms ( MISO ) [40] , [41] . Criteria used to define significance are according to the standard options in the Cufflinks and MISO programs . More detail is provided below .
To investigate the effect of DENV1-16007 ( wild-type strain ) and DENV1-PDK13 ( attenuated strain ) infection on the transcriptome of the human host , we infected human hepatoma cells ( HuH7 ) with each strain for 20 hours at a multiplicity of infection ( MOI ) of 20 ( Figure 1A ) . The MOI of 20 was chosen so that the subsequent RNAseq profiling would best reflect the infection-induced alterations to the host transcriptome and not be either masked by or derived from a large number of uninfected cells . We also measured viral RNA over the first 30 hours of infection to address the possibility that changes observed might be the result of a delayed replication cycle by one of the viruses . Although the absolute kinetics of the two viruses differ , the 20 hour time point was chosen as it represents the stage in the primary round of replication at which the genome copy numbers are most similar between the two viruses ( Figure 1B ) . Indirect immunofluorescence staining for the pre-membrane protein ( prM ) production with 2H2 monoclonal antibody also showed similar infection levels for both viruses at this time point ( Figure 1C ) . Infections with wild-type and attenuated strains were performed in three independent biological replicate experiments . Mock-infected HuH7 cells treated in the same way as the infected samples were also done in biological triplicate and served as the control for our experiment . At twenty hours post-infection , mRNA was extracted from all samples independently , and poly-A enriched cDNA libraries were constructed for 75-base , pair-end sequencing on an Illumina GAIIx ( one sample for each condition ) or Illumina HiSeq2000 machine ( two samples for each condition ) . RNA sequencing was performed independently for each of these replicate experiments . Mapping of reads ( Bowtie , Tophat ) and analysis of differential transcriptome response ( Cufflinks ) to infection was performed using the Tuxedo Suite of software [40] , [42] . Cufflinks utilizes sequence fragments mapped to the reference genome to estimate the abundance of each isoform arising from the gene . It then tests for differential expression between experimental conditions . Differential expression at the gene level is calculated and is defined as the sum of differential expression of all isoforms at a particular locus . Cufflinks also assesses differential splicing by comparing the relative abundance of isoforms using the same transcriptional start site [40] . As this definition of splicing encompasses all different types of splicing events ( see below ) [43] , systematic analysis of these events for downstream validation work is exceedingly difficult . In order to attain more detail about the types of differential splicing in our samples , we used the MISO software following mapping [41] . MISO utilizes a fixed library of previously characterized splice events to predict alternative splicing and reports the number of occurrences for each type of splicing event: skipped exon ( SE ) , mutually exclusive exons ( MXE ) , alternative 3′ splice site ( A3SS ) , alternative 5′ splice site ( A5SS ) , alternative first exon ( AFE ) , retained intron ( RI ) , tandem untranslated regions ( Tandem UTR ) [43] . MISO also utilizes a different algorithm than Cufflinks to predict differential splicing between samples which , when compared with the results from Cufflinks , provides an additional level of stringency in selection of candidates for downstream analysis . To assess the quality of our libraries and sequencing performance and to ensure that any differences observed between our samples was due to the biology and not bias in sequencing , we used the open source program RNAseqC to examine our data [44] . Results indicate that despite using two different Illumina machines to generate the sequences , the individual samples are highly comparable to each other across all the metrics interrogated ( Dataset S1 ) . Over 18 , 000 changes to the host transcriptome were observed in response to infection by the wild-type strain and >41 , 000 were observed in response to infection with the attenuated strain ( Table 1 ) . Differential isoform regulation is the largest category of response due to infection by both strains . Interestingly , there are over two-fold more differentially regulated isoforms following infection with the attenuated than with the wild-type strain . Similarly , infection with the attenuated strain also resulted in three-fold more differentially regulated genes than infection with the wild-type ( Figure 2 ) . In order to determine whether this large differential response to infection was specific to these two strains or whether the ‘quieter’ response to the parental DENV1-16007 strain was typical of wild-type DENV1s , we repeated our experiment in HuH7 cells with two low passaged clinical isolates ( EDEN3300 and SL107 ) and compared them to our uninfected control . RNAseq analysis for these isolates indicates even fewer transcriptomic changes ( Table 1 ) , suggesting that attenuated virus triggers more host cell response than wild-type viruses . This observation is consistent with what has been reported for yellow fever virus and its attenuated derivative , YF17D [45] . Qualitative analysis of the RNAseq data also provides insights into the pathways that are essential to both strains or unique to one strain . Ingenuity Pathway Analysis ( IPA ) of isoforms indicates that the commonly regulated isoforms are enriched in pathways associated with viral infection and modulation of protein translation . Examples include EIF2 signaling , prolactin signaling , acute phase response signaling , regulation of EIF4 and p7056K and glucocorticoid receptor signaling . Differential expression of pathways associated with cellular growth and proliferation such as mTOR signaling and aryl hydrocarbon receptor signaling are also enriched following infection with both strains ( Figure 3A ) . Conversely , the pathways that are differentially regulated between the wild-type and attenuated strains include the innate immune response and cell cycle control . The top differentially regulated host pathways associated only with wild-type strain infection are involved with immunomodulation and cell cycle arrest , such as PPAR/RXR activation pathway , G2/M DNA damage checkpoint , ATM signaling and the PDGF signaling pathway ( Figure 3C ) . Conversely , infection with the attenuated strain triggered pathways associated with inflammation , induction of apoptosis and stress such as the TNFR1 pathway , TWEAK signaling , NRF2-mediated oxidative stress response , IGF -1 signaling and ERK5 signaling ( Figure 3E ) . IPA also identified molecular and cellular functions associated with both or each of the strains ( Figures 3B , 3D and 3F ) . Taken collectively , the molecular and cellular functions in response to wild-type virus infection are associated with cell signaling and metabolism while those to attenuated virus are associated with transcriptional activation , cell cycle modification and post-translational modification . Next , we interrogated the data for alternative splicing within the isoforms sharing the same transcriptional start site . Interestingly , 80% and 74% of all differential splicing following infection with the wild-type strain and attenuated strain , respectively , are common to both viruses ( Figure 2 ) . This degree of commonality in the splicing response to infection by each strain is significantly different than what was observed for differential isoform and gene regulation in response to infection . This novel finding suggests that mechanisms responsible for the specific regulation of host splicing may be critical for DENV1 propagation and thus remained relatively unchanged during the attenuation process . The candidate list from our Cufflinks analysis of alternatively spliced transcripts for both strains of virus was then cross-referenced against the list of skipped exon ( SE ) events generated by MISO analysis . This cross comparison resulted in 79 total SE events . To assess the accuracy of our alternative splicing predictions , we performed qPCR on each of the predicted SE events . Of the 79 events tested by qPCR , 32 ( 40% ) were differentially expressed in the wild-type strain , the attenuated strain or both in comparison to an uninfected control ( Figure 4 ) . The splicing patterns were similar for both viruses across all time points measured , indicating that our observations are not artifacts of temporal sampling bias ( Figure 4 ) . Furthermore , the genes involved in these splicing events belong to many of the same pathways shown in Figure 3 suggesting that the virus is , at least in part , exerting its influence on these pathways through alternative splicing ( Table S1 ) . If this were true , the SE events should share mechanisms of regulation by utilizing common RNA motifs within and surrounding the identified SE's . Indeed , using the software RegRNA ( http://regrna . mbc . nctu . edu . tw ) [46] and setting an arbitrary boundary of 250 nucleotides upstream and downstream of the 3′ and 5′ splice-site of the SEs , respectively , we identified 74 predicted RNA regulatory motifs and elements which could be bound by 17 RNA binding proteins . By using GeneCards ( http://www . genecards . org ) to convert these putative motif binding proteins into HUGO nomenclature , we observed that nearly two thirds ( 11 of 17 ) of genes encoding these motif-binding proteins are themselves differentially expressed following infection by one or both of the DENV1 strains ( Table S2 ) . These results suggest that DENV1 may be regulating alternative splicing of host mRNA by a hitherto unknown mechanism . Many differentially processed transcripts found in our study have also been identified in genome-wide RNAi screens for flavivirus host factors , differentially processed in microarray studies of DENV-infected cells and/or shown to interact with DENV gene products [10]–[29] , [47]–[51] ( Jamison and Garcia-Blanco , unpublished data ) . To gain additional functional insights into the differentially regulated transcripts identified here , we compiled a cross-platform integrative analysis of our data with other available genomic data on host factors in DENV and other flaviviral infection ( Tables S3 , S4 , S5 , S6 , S7 , S8 , S9 , S10 , S11 , S12 , S13 ) . Briefly , we examined 11 canonical pathways that were enriched in our study and/or had been previously implicated in the DENV life cycle: apoptosis , autophagy , clathrin-mediated endocytosis , interferon signaling , lipid metabolism , oxidative phosphorylation , regulation of stress granules and P-bodies , splicing-related RNA post-transcriptional modification , ubiquitination , endoplasmic reticulum stress , virus recognition and interferon induction . The proportion of host factors for DENV or other flaviviruses identified through functional genomic studies that were differentially regulated following infection with the wild-type strain or the attenuated strain is indicated in Table 2 . Differential isoform regulation of flaviviral host factors ranged from 15 . 5% to 60 . 6% and is uniformly higher for the attenuated strain than the pathogenic wild-type strain . Alternative splicing ranged from 7 . 3% to 24 . 2% and the rates of these events in the wild-type strain and its attenuated derivative were comparable ( Table 2 ) . The intersection of specific host factors in these eleven canonical pathways that are either required for flaviviral propagation or involved in direct interaction with DENV proteins , which are differentially regulated are graphically depicted in Figure 5 .
Investigation of host gene expression to date has relied primarily on microarray technology [12]–[29] . This technology is insensitive to the regulation of genes through alternative RNA processing . Thus , detection of differential expression down to the level of alternative isoforms has not been examined and the subtleties of TSS and post-transcriptional modifications that can dramatically alter the function of the derived proteins have been largely ignored . These processes could be a mechanism by which DENV attains specific isoforms of required host factors while suppressing those that act to restrict its replication [27] , [52] , [53] . Indeed , the need to understand the host response beyond simple gene expression is underscored by the observation that DENV can modify the splicing pattern of an endogenous gene , XBP1 , to its advantage [53] . Our findings indicate that there is an extensive amount of previously uncharacterized gene isoforms and alternative processing of host transcripts over a wide range of pathways and host functions in response to DENV infection . Interestingly , the DENV1-16007 and DENV1-PDK13 viruses only differ from each other by 14 nucleotide and 8 amino acids [54] , yet the host transcriptional response to these viruses is pronounced . This suggests that infection with different strains of DENV can result in significantly different disease phenotypes despite few nucleotide differences . We have attempted , in this study , to provide a comprehensive guide to the transcriptomic changes with DENV infection . By analyzing our RNAseq data using two different programs , Cufflinks and MISO [40] , [41] , maximal information on RNA transcript regulation , could be gleaned . Specific splice events could hence be identified for subsequent mechanistic studies that clarify their role in the host response . In particular , the integrative analysis of this study with existing functional genomics data reveals previously undocumented expression and post-transcriptional regulation of required host factors that should serve as a road map for future mechanistic investigations . A caveat , however , is that our work only profiled the transcriptome at the end of a single round of DENV replication . As hinted by Figure 4 , both quantitative and qualitative differences may exist in the transcriptome at different stages of the virus life cycle . Furthermore , the possibility exists that bystander uninfected cells may exert some influence on the observed transcriptional changes although we have attempted to minimize this by using a high MOI in our experiments . Future studies may need to take these possibilities into account . The large difference in the number of alternative splicing events identified by Cufflinks and MISO also underscores the fledgling nature of RNAseq . While the former identifies all possible splicing events from the data de novo , the latter relies on a pre-defined library to map alternatively spliced transcripts . Additional studies are needed to clarify whether Cufflinks over-estimated the number of splice variants , or there are authentic variants absent from the library used by MISO . Regardless , for transcriptome analysis of host response to infection , RNAseq is superior to microarray in terms of the breadth of information derived . Our results also indicate that the human transcriptome response to an infection by wild-type DENV or its attenuated derivative differs significantly ( Table 1 , Figure 2 ) . These differences suggest that alternative processing events may be part of a previously uncharacterized innate immune response to DENV1 infection that is in large part evaded by wild-type strains . This second hypothesis is supported by the greater than two-fold increase in the number of differentially regulated transcripts when infected with the attenuated strain of DENV1 as compared to the parental wild type strain of DENV1 , many of which belong to pathways associated with inflammation , induction of apoptosis and stress response ( Table 1 , Figure 3 ) . This inability to escape the innate immune response achieved by the wild-type virus may explain the lack of reactogenicity . It may also explain why antibody titer engendered by vaccination is not as high as those observed following natural infection unless supplemented with an adjuvant that stimulates this innate immune response . This observation also suggests a mechanism of pathogenicity where DENV regulates host transcriptome changes by interacting with a group of RNA binding proteins to control multiple splicing events . The development of a live attenuated tetravalent vaccine for DENV1-4 has bedeviled researchers for the past 60 years . The less than optimal efficacy of the leading dengue vaccine candidate makes an improved understanding of the molecular basis of a good vaccine all the more critical [55] . The differential host transcriptome response to infection with DENV1-16007 and DENV1-PDK13 provides an insight into the characteristics of an attenuated virus which , is likely a complex phenotype [45] . A molecular understanding of the basis of attenuation could lead to a quantitative approach to balancing reactogenicity and immunogenicity , which presently remains a hit-or-miss finding made only after lengthy clinical trials . In conclusion , we provide here a detailed view of the host cell transcriptome response to infection with wild-type DENV-1 and its attenuated derivative that could be useful for future studies on the genetic determinants of viral virulence and attenuation . | Dengue is the most common insect-borne viral disease globally . The continued absence of an effective therapy stems from an incomplete understanding of disease pathogenesis , of which the host response to infection is thought to play a central role . While previous studies have described the changes in total gene expression with dengue virus infection , they have not been able to provide any information on the subtle variations of the host RNA . These variations lead to the production of gene isoforms that can have a profound effect on gene function . In the current study , we have used the newly developed technique of RNA sequencing to more accurately interrogate the variations in the host RNA after infection with a wild-type dengue virus or its attenuated derivative . Findings from this study show that there is an extensive amount of previously uncharacterized variation in host RNA response to dengue infection . The response to infection with the wild-type dengue also differs significantly from infection with the vaccine strain . This suggests that variations in the host RNA comprise a part of the host response to viral infection that is in large part evaded by wild-type dengue viruses . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"genome",
"expression",
"analysis",
"infectious",
"diseases",
"dengue",
"fever",
"neglected",
"tropical",
"diseases",
"biology",
"genomics",
"genetics",
"and",
"genomics"
]
| 2013 | Host Cell Transcriptome Profile during Wild-Type and Attenuated Dengue Virus Infection |
HIV is adept at avoiding naturally generated T cell responses; therefore , there is a need to develop HIV-specific T cells with greater potency for use in HIV cure strategies . Starting with a CD4-based chimeric antigen receptor ( CAR ) that was previously used without toxicity in clinical trials , we optimized the vector backbone , promoter , HIV targeting moiety , and transmembrane and signaling domains to determine which components augmented the ability of T cells to control HIV replication . This re-engineered CAR was at least 50-fold more potent in vitro at controlling HIV replication than the original CD4 CAR , or a TCR-based approach , and substantially better than broadly neutralizing antibody-based CARs . A humanized mouse model of HIV infection demonstrated that T cells expressing optimized CARs were superior at expanding in response to antigen , protecting CD4 T cells from infection , and reducing viral loads compared to T cells expressing the original , clinical trial CAR . Moreover , in a humanized mouse model of HIV treatment , CD4 CAR T cells containing the 4-1BB costimulatory domain controlled HIV spread after ART removal better than analogous CAR T cells containing the CD28 costimulatory domain . Together , these data indicate that potent HIV-specific T cells can be generated using improved CAR design and that CAR T cells could be important components of an HIV cure strategy .
It is well established that T cells play an important role in controlling HIV-1 ( HIV ) replication , and that HIV-infected individuals develop robust HIV-specific T cell responses [1 , 2] . However , HIV evades the endogenous T cell-mediated immune response by altering key residues required for T cell recognition and downregulating class I major histocompatibility complexes ( MHC ) [3 , 4] . Additionally , HIV-mediated depletion of HIV-specific CD4 T helper cells and the chronic persistence of the virus functionally impair HIV-specific CD8 T cells so that , in most individuals , T cells are unable to control HIV replication [5] . Although antiretroviral therapy ( ART ) can suppress HIV replication by many orders of magnitude , it fails to eliminate the virus , forcing HIV-infected individuals to be treated by ART regimens for the rest of their lives . ART also reduces the number of HIV-specific T cells present , due to a massive reduction in HIV antigen , and the T cells that remain often have functional defects [6 , 7] . Thus , at the time of ART removal when the number of HIV producing cells is minimal , the resident HIV-specific T cell immune response is ill-equipped to control the re-emerging infection and uniformly fails , with viral loads returning to patient-specific set point within weeks after ART removal [8] . As a result , we and others have postulated that instead of relying on the endogenous immune response to control HIV replication as part of an HIV cure strategy , the introduction of a potent engineered immune response designed to overcome HIV’s escape mechanisms will be required to provide durable control HIV in the absence of ART [9–11] . Previous adoptive cell therapy ( ACT ) trials have demonstrated that simply reinfusing expanded patient T cells does not result in durable control over HIV replication [9] . Thus , both the quality and quantity of infused HIV-specific T cells must be enhanced for a sustained therapeutic benefit . Attempts to manufacture better T cells for ACT have recently been made through antiviral transgenes , coreceptor editing , and redirecting T cells with HIV-specific T cell receptors ( TCRs ) or chimeric antigen receptors ( CARs ) [9 , 12] . CARs consist of an extracellular antigen binding domain fused to intracellular T cell activation domains [13] . These synthetic receptors can redirect T cells to recognize viral proteins independent of antigen processing , TCR , and MHC . CARs targeting CD19 have revolutionized the treatment of leukemia and lymphomas through their ability to persist and maintain durable anti-tumor effects in vivo [14 , 15] . Of note , the first CAR to enter human trials redirected T cells to target the GP120 region of the HIV Envelope ( Env ) glycoprotein [16–18] . This was achieved by fusing the extracellular and transmembrane domains of CD4 , the cellular receptor for HIV , to the CD3-zeta cytoplasmic region ( CD4-zeta ) . HIV-specific cytotoxicity was established in vitro , and safety was demonstrated in vivo , with transient reductions in HIV RNA , DNA , and quantitative HIV outgrowth assays [16 , 17 , 19–21] . A long-term follow up study determined the half-life of the CAR-modified cells to be over 16 years , with CAR expression up to 10 years post infusion and no serious adverse events [22] . However , the CD4-zeta CAR did not lead to sustained reductions in viral loads or the viral reservoir , and the clinical data were not sufficiently promising to warrant additional development . Subsequent to the HIV CAR clinical trials , a number of advances have been made in CAR design for optimal antitumor responses . Xenograft models and clinical trials have established that costimulation augments function , proliferation , and survival in vivo [23–25] . In addition , CAR structural and signaling domains have been found to greatly impact T cell function and susceptibility to exhaustion [26 , 27] . For example , it was recently demonstrated that CAR costimulatory domains influence T cell metabolic and phenotypic profiles , with 4-1BB promoting a central memory phenotype and CD28 promoting an effector memory phenotype [28] . While there has recently been a renewed interest in utilizing CARs to control HIV , including efforts to increase CAR T cell function and survival , a systematic optimization of CAR T cell cytotoxicity was lacking [29 , 30] . We sought to apply the lessons learned from engineering CARs for hematologic malignancies to optimize the CD4 CAR for superior control over HIV replication . We demonstrate here that changing the CAR expression vector , promoter , transmembrane , and costimulatory domains improved control over HIV in vitro by over 50-fold . In vivo , humanized mice engrafted with T cells expressing these optimized vectors had significantly higher CD4 T cell counts , greater CAR+ CD8 T cell proliferation after HIV infection , and 90% less HIV RNA , compared to mice that received T cells transduced with the clinical trial CD4-zeta CAR . An HIV treatment model demonstrated superior control over HIV replication by 4-1BB containing CARs , compared to similar CARs containing the CD28 costimulatory domain . These data provide a compelling reason to revisit human clinical trials with CD4 CARs that have been optimized for control over HIV in vivo for use in HIV cure studies .
Preclinical studies testing the original CD4-zeta CAR showed that T cells expressing this construct had equivalent antiviral activity to naturally generated HIV-specific T cells [21] . We hypothesized that the reason these clinical trials failed to demonstrate durable clinical responses was that the T cells used in these trials were no more potent than the endogenous HIV-specific T cell response . Therefore , we sought to optimize the CD4-zeta CAR based on lessons learned from cancer-specific CARs , in order to augment control over HIV replication [31] . To do this , we optimized each component of the CAR in a step-by-step manner . We first addressed to what extent the vector backbone contributes to the ability of CAR+ CD8 T cells to control HIV . The original CD4-zeta CAR was expressed by a murine retroviral vector ( MMLV-based ) . Since MMLV-based vectors target promoter regions and lentiviral vectors ( HIV-based ) integrate preferentially into open reading frames , we reasoned that lentiviral vectors would result in higher expression than MMLV-based vectors [32] . We generated the MMLV clinical trial construct and an analogous lentiviral vector , both with the PGK promoter , and transduced primary human CD8 T cells . Lentiviral transduction consistently resulted in a ~10-fold higher median fluorescence intensity ( MFI ) of CAR expression compared to MMLV retrovirus ( Fig 1A–1D ) . To determine whether the higher transgene expression was the result of more vector integrations per cell , we measured the integrated vector copy number in the different T cell populations using a previously established assay [22] . We found that the MMLV-based vector had almost twice the number of integrations compared to the HIV-based vector ( Fig 1E ) , indicating that the higher transgene expression of the HIV-based vector is due to its intrinsic properties . A co-culture assay was used to compare the different CARs in terms of their ability to control HIV replication ( Fig 1F ) . In this assay , HIV-infected CD4 T cells were cultured with either nontransduced ( NTD ) CD8 T cells or CAR transduced CD8 T cells , and the ability of the effector CD8 T cells to control HIV spread was measured over 7–14 days . To distinguish between HIV spread throughout the CD4 T cells and infection of the CD4 CAR+ CD8 T cells , we gated separately on the CD8 negative and the CD8 positive T cells , and then analyzed intracellular p24 ( Gag ) in CD4 T cells and CD8 T cells , respectively ( see gating strategy in S1 Fig ) . High levels of HIV replication were observed in CD4 T cells co-cultured with NTD CD8s ( Fig 1G ) . In contrast , CD8 T cells transduced with either a MMLV-based or HIV-based CAR vector were able to control HIV replication at a 1:1 E:T ratio , similar to what was observed in previous studies of this strategy [19–21] . However , upon diluting the CAR transduced CD8 T cells to lower E:T ratios , T cells transduced with the HIV-based vector were superior at controlling HIV replication over time ( Fig 1G–1J ) . Ultimately , neither population of transduced CD4 CAR CD8 T cells could control HIV spread at a 1:50 E:T ratio . In contrast to recent reports that CD4 CAR transduced CD8 T cells are susceptible to infection by cell-free virus [29 , 30] , we were only able detect intracellular Gag in CAR+ CD8 T cells after diluting to low E:T ratios with HIV-infected CD4 T cells ( Fig 1H , S2 Fig ) . Of note , there was a small proportion of nontransduced CD8 T cells that stained positively for Gag , regardless of the E:T ratio . This was likely the result of the ability of CD8 T cells to transiently express CD4 after activation [33–36] . This data highlights the complex relationship between CD4 CAR expression and susceptibility to HIV infection . At high E:T ratios the CAR+ CD8 T cells are able to fully suppress HIV replication , and thus they and the co-cultured CD4 T cells are protected from HIV infection . However , at low E:T ratios , the CAR+ CD8 T cells are no longer able to suppress HIV replication , and the virus is able to spread throughout both populations of cells ( Fig 1H ) . Transgene expression in T cells wanes when driven by the PGK promoter as T cells rest down . In contrast , the EF1α promoter induces higher expression that is better sustained as T cells return to quiescence [23] . We hypothesized that the EF1α promoter might be beneficial in our system , since greater CAR MFI expression correlated with better control over HIV ( Fig 1 ) . Under the EF1α promoter , CAR expression MFI increased ~10-fold compared to the PGK promoter ( Fig 2A and 2B ) . Next , we substituted the CD8α transmembrane ( TM ) domain in place of the CD4 TM domain to promote CAR dimerization , remove CD4 TM motifs targeted by HIV Vpu for downregulation , decrease homology to the HIV cellular receptor , and ultimately augment cytotoxicity [37–40] . Greater control over HIV replication was achieved by both modifications individually , and a combination of the two modifications led to complete control over HIV replication down to a 1:50 E:T ratio ( Fig 2C and 2E ) . Substitution of the CD8α TM domain decreased infection of CAR CD8 T cells regardless of the promoter used at the 1:25 and 1:50 E:T ratios ( Fig 2D ) . We observed similar results when examining the culture supernatants for p24 Gag ( S3 Fig ) . However , as seen in Fig 1 and S2 Fig , the CAR+ CD8 T cells could be diluted to the point where they no longer controlled HIV infection and succumbed to infection themselves ( Fig 2D ) . To ensure this was not an artifact of gating on a few CD8 T cells , we performed a larger scale experiment where at least 1x104 CD8 T cells were collected per condition and the infection pattern was the same ( S4A and S4B Fig ) . HIV rapidly downregulates CD4 expression in an HIV Nef dependent manner , and we wanted to determine whether the CD4 CAR construct was susceptible to downregulation by HIV infection . Therefore , we compared CD4 expression in HIV-infected primary human CD4 T cells and CD4 CAR transduced CD8 T cells ( S4C Fig ) . We observed that CD4 CAR expression was maintained at a high level even in presence of robust HIV infection . This finding is consistent with previous findings that demonstrated a dileucine motif within the CD4 cytoplasmic tail is required for HIV Nef mediated downregulation [41] and since CD4 CAR does not have this motif , it is immune to HIV Nef downregulation . Thus , CD4 CAR T cells may still function even after being infected with HIV . Thus , altering the viral vector , promoter , and transmembrane domains afforded a 50-fold increase in potency over the clinical trial MMLV-based retrovirus , resulting in complete control over HIV replication at a 1:50 E:T ratio in vitro ( Fig 2E and 2F ) . Elite controllers are rare individuals who control HIV replication in the absence of ART . Certain HLA alleles , such as HLA-B57 , are overrepresented in these cohorts , suggesting that T cell responses play a key role in controlling their virus [42] . Therefore , we wished to determine whether T cells expressing a re-engineered CD4 CAR ( with the EF1α promoter and CD8α TM ) could control HIV replication better than T cells expressing a HLA-B57 restricted TCR that is associated with better control over HIV replication . To do this , we generated an analogous lentiviral vector that expressed the TCRα and TCRβ chain ( obtained from B57-KF11 specific T cell clone generated from an elite controller generously provided by Xu Yu and Bruce Walker ) linked by T2A sequence which when expressed in T cells conferred specificity for B57-KF11 epitope ( HIV p24 Gag epitope KAFSPEVIPMF ) . To confirm these cells recognized Gag-expressing cells , we mixed them with target CD4 T cells from HLA-B57+ individuals that were transfected with Gag RNA , or Pol RNA as a negative control , and detected a robust Gag-specific cytokine response ( S5B Fig ) . Next , we compared the ability of KF11 TCR versus CD4 CAR transduced CD8 T cells to limit HIV spread in HLA-B57+ CD4 T cells . While KF11 TCR-transduced CD8 T cells reduced HIV replication down to a 1:25 E:T ratio , complete control over HIV replication was never achieved ( Fig 3 ) . In contrast , the re-engineered CD4 CAR controlled HIV almost completely down to a 1:100 E:T ratio . Efforts to improve the efficacy of B57-KF11 transduced T cells could be achieved by finding higher avidity TCRs [43] or by approaches that force B57-KF11 TCR pairing . Nonetheless , these data suggest that the synthetic CD4 CAR approach is more potent than the natural TCR based approach and that the CD4 CAR approach will likely be more effective as a cellular therapy tool compared to T cells transduced with a patient-derived , natural TCR . The most commonly used CAR ectodomains are antibody-derived single chain variable fragments ( scFvs ) [13] . Over the past several years , a number of HIV-specific antibodies have been described that bind or neutralize HIV with high affinity and/or target a wide breadth of viruses [44] . We tested whether the use of scFvs to redirect T cells to HIV was superior to the use of CD4 . In addition , we wanted to test whether CD8 T cells expressing scFv-based CARs were less susceptible to HIV infection than CD4-based CARs . A panel of scFvs derived from the VRC01 , 3BNC60 , PG9 , PGT128 , or PGDM1400 parental antibodies were generated due to their neutralization breadth and/or potency against HIV and cloned into the most effective CAR design identified in Fig 2 , with the EF1α promoter and the CD8α TM domain [45 , 46] . To determine that each scFv CAR had folded properly , could interact with Env , and promote CD8 T cell lysis , we measured specific lysis of chromium labeled K562 target cells expressing the Env protein from the HIV YU2 strain . All CARs were capable of lysing Env-expressing targets to a similar degree , indicating that these CARs recognized HIV Env ( Fig 4A ) . Many of the scFv CARs consistently produced higher levels of intracellular cytokines in response to Env-expressing targets when compared to the CD4 CAR ( S6A Fig ) . However , when co-cultured with HIV-infected CD4 T cells , the CD4 CAR controlled HIV better than all of the scFv-based CARs ( Fig 4B and 4D and S6B Fig ) . Interestingly , the PGT128 CAR repeatedly controlled HIV better than the other scFvs tested , despite being less broad and less potent in neutralization studies than PGDM1400 [46] . Surprisingly , at low E:T ratios we detected high levels of intracellular Gag in the scFv CD8 T cells , similar to CD4 CAR+ T cells diluted to a 1:200 E:T ( Fig 4C ) . This was not a byproduct of lentiviral transduction or generic CAR expression , as this was not seen for GFP-transduced or CD19 CAR-transduced cells ( S6C Fig ) and thus appears to depend on HIV binding ability , possibly concentrating the virus near the CAR-transduced cell membrane . Based on the in vitro superiority of the CD4 CAR against this limited scFv subset , and the inherent difficulty of HIV escaping from CD4 binding , we chose to pursue development of CD4-based CARs for in vivo testing . T cells require costimulatory signals for proliferation , effector function , and long-term survival [47] . Costimulatory domains , such as CD28 and 4-1BB , have been incorporated into recent CAR designs for durable CAR T cell responses in vivo [24] . In chronic HIV infection , T cell dysfunction and exhaustion have been well documented , and decreased CD28 and 4-1BB signaling impair cytolytic and effector function [7 , 48 , 49] . Therefore , we generated a panel of CD4 CARs that incorporated a variety of costimulatory domains in conjunction with the CD3-zeta domain , including CD28 , 4-1BB , CD28+4-1BB , OX40 , ICOS , or CD27 and tested their ability to control HIV infection in vitro . CD8 T cells expressing CARs that contained 4-1BB , CD27 , or ICOS costimulation domains did not control HIV as effectively as T cells expressing CARs that expressed the other costimulatory domains , suggesting that these costimulatory pathways interfere with control over HIV replication ( Fig 5 ) . Regardless of the costimulatory domain , the control seen with these CD4 CARs was superior to the control seen with an HIV-specific TCR ( Fig 3 ) or with scFv based CARs ( Fig 4 ) . While CD28 promoted better control over HIV in vitro compared to 4-1BB , discrepancies between the in vitro and in vivo activity of cancer-specific CARs containing 4-1BB had been reported [26 , 28] . Thus , we were curious if this held true for HIV-specific CARs , and decided to further characterize the safety profile and in vivo efficacy of CD4 CARs expressing either CD28 or 4-1BB . Preclinical data demonstrated that T cells expressing the clinical trial CD4-zeta CAR did not kill Raji cells , which express high levels of MHC class II , the low affinity ligand of CD4 [20] . However , we were concerned that our optimized , highly expressed CD4 CAR might recognize MHC class II expressing cells . To test this , we measured CD4 CAR transduced CD8 T cell responses against K562 cells stably expressing high levels of the HLA-DR*0401 allele ( S7 Fig ) . CD8 T cells were transduced with optimized CD4 CARs containing CD3-zeta alone or with the 4-1BB and CD28 costimulatory domains and cultured with unmodified K562 target cells , HLA-DR*0401+ K562 cells , or HIV YU2 Env+ K562 cells as a positive control . CD4 CAR transduced CD8 T cells produced IL-2 , CD107a , IFN-γ , and MIP-1β in response to HIV Env+ targets but not in response to HLA-DR+ or parental K562s , with the most robust production by CD28-containing CAR T cells ( Fig 6A ) . A small MIP-1β signal was observed for all CARs mixed with parental or HLA-DR expressing targets that was not observed with NTD controls , likely due to some constitutive signaling observed in CAR T cells [50] . It has been shown that autocrine production of beta chemokines by CMV-specific T cells decreases CCR5 expression and protects these cells from HIV infection , so this low-level MIP-1β production may help protect the CAR+ CD8 T cells in vivo [51] . Importantly , no difference in cytokine or CD107a production was detected by CAR T cells cultured with parental versus MHC class II expressing cells , suggesting that our optimized CD4 CARs do not facilitate off-target recognition of MHC class II . Coculturing the optimized CD28-containing CAR with HLA-DR+ K562 cells over an extended period further confirmed the lack of off-target responses . We mixed HLA-DR+ K562 cells 1:1 with HLA-A2+ K562 cells , which served as a negative control , and did not see a change in the ratio of the two K562 populations over time ( Fig 6B and 6C ) , suggesting that the re-engineered CAR will exhibit a similar safety profile in humans as the original CD4 CAR . CD19-specific CARs containing the CD28 signaling domain had superior in vitro activity than those containing the 4-1BB signaling domain , but the 4-1BB containing CARs proved superior in humanized mouse models and in patients [15 , 23 , 52] . We wished to determine whether the same was true for HIV-targeting CARs . In addition , we wanted to determine if our optimized CD4 CARs could control HIV better in vivo then the original CD4-zeta CAR that was tested in the clinic . To do this , we utilized a NSG humanized T cell ( NSG hu-T cell ) mouse model . In this model , detectable T cell engraftment ( >10 cells per μl of blood ) takes 2–3 weeks , and over the next 2–3 months T cell engraftment slowly rises until the mice become sick due to xenograft-mediated GVHD [53 , 54] . HIV infection prevents CD4 T cell expansion and , paradoxically , makes the animals healthier . Thus , evidence of GVHD and high levels of CD4 T cell engraftment are strong evidence that anti-viral agents , such as CAR T cells in this case , are effective . Four groups of mice were compared with different CD8 effector cell populations: nontransduced ( NTD ) , transduced with the optimized lentiviral vector containing either 4-1BB ( BBz ) or CD28 ( 28z ) costimulatory domains , or transduced with the clinical trial MMLV-based vector ( MMLV-CD4z ) . Pre-infection , baseline CD4 T cell counts did not differ significantly between the NTD , BBz , or 28z groups , and were significantly higher for the MMLV-CD4z treated mice ( Fig 7A ) . After HIV infection , we observed that mice infused with T cells expressing the BBz or 28z construct had a 17-fold and 177-fold expansion of the number of human CD4 T cells , respectively ( Fig 7B ) . In contrast , endpoint CD4 counts were depleted in NTD or MMLV-CD4z mice ( Fig 7B ) . Examination of the number of CAR+ CD8 T cells in the different mouse cohorts revealed 389-fold , 587-fold , and 2-fold expansions in the BBz , 28z , and MMLV-CD4z T cells , respectively ( Fig 7C and 7D ) , suggesting there is a correlation between the ability of CAR T cells to expand and the ability to protect CD4 T cells from HIV-mediated destruction . We also examined viral loads . Since HIV replication is highly dependent on the number of CD4 T cells present in this model , we normalize viral load to the number of CD4 T cells present at a given timepoint to fairly compare different treatment groups . Seven days following HIV infection , BBz CAR T cells exhibited the greatest control over virus replication , with many mice showing undetectable virus loads , whereas plasma from NTD , 28z , and MMLV-CD4z treated animals contained approximately 1 normalized copy of HIV RNA per μl ( Fig 7E ) . Eighteen days post infection , the median copy number of HIV RNA was reduced by more than 10-fold in both BBz and 28z treatment groups , compared to the mice that were treated with NTD T cells ( Fig 7F ) . However , MMLV-CD4z treated mice had similar HIV RNA loads as NTD treated mice . Thus , the optimized CARs are superior at protecting CD4 T cells , promoting CD8 T cell expansion , and controlling HIV replication in vivo , with BBz CARs superior in preventing the early spread of HIV . After establishing that optimized CARs can function in humanized mice to control HIV replication , we next wanted to model CAR treatment of pre-established HIV infections and further examine whether 4-1BB or CD28 costimulation promoted better in vivo control . To mimic how CARs would be applied in a clinical trial , we injected CAR T cells into NSG mice with a previously established pool of HIV-infected T cells in the presence of ART and monitored virus rebound after ART was stopped ( S9A Fig ) . After three days of HIV infection in the presence of ART , the peripheral blood CD4 T cell counts were similar for all groups ( Fig 8A ) . However , at 18 days post ART removal CD4 T cell depletion was apparent in NTD and 28z CAR treated mice , with significantly higher CD4 T cell counts in the BBz treatment group ( Fig 8B ) . By the endpoint bleed , the 28z CARs demonstrated increased protection of CD4 T cells , and only the NTD mice had significantly lower CD4 T cell counts ( Fig 8C ) . In contrast , mock infected mice maintained similar CD4 T cell counts in all treatment groups at all timepoints ( S10A–S10C Fig ) . Interestingly , endpoint CD4 T cell counts were similar in this experiment for both BBz and 28z CAR treatment groups ( Fig 8B and 8C ) , as opposed to Fig 7 where CD4 counts were significantly higher in 28z treated mice . Ten days post ART removal , the CAR treated mice had higher peripheral blood CD8 T cell counts compared to NTD mice ( Fig 8D ) . This effect was HIV-specific , as all mock treated mice had similar CD8 T cell counts ( S10D Fig ) . However , by 18 days post ART removal the CD8 T cell counts were significantly higher in mice that received BBz CARs compared to NTD or 28z CAR-treated mice ( Fig 8E and 8F ) , and this was also seen in mock treated mice ( S10E and S10F Fig ) , consistent with the notion that 4-1BB signaling promotes T cell persistence in the absence of antigen [23 , 26 , 55] . Prior to CD8 T cell injection , while the animals were on ART , most mice had undetectable viremia and 3/20 had very low levels of plasma HIV RNA ( <1 copy per ul , S1 Table ) . Ten days post ART removal , all NTD mice had detectable plasma RNA , whereas all CAR treated mice had very low or undetectable HIV , as measured by plasma HIV RNA ( Fig 8G ) . Similar patterns of control were detected via measuring plasma HIV p24 Gag protein using an ultrasentive assay ( S11 Fig ) . However , after 18 days post ART removal , 28z treated mice experienced an increase in HIV replication and had similar plasma levels of HIV as NTD mice , whereas the BBz mice maintained significantly better control ( Fig 8H and 8I and S11B Fig ) . Together , these data suggest that a CD4 CAR containing the 4-1BB zeta signaling domain will be most effective in HIV cure strategies because 1 ) its ability to act rapidly to prevent HIV spread ( Fig 7E and 7F ) , 2 ) its ability to durably prevent viral rebound ( Fig 8H and 8I ) , and 3 ) its ability to promote T cell survival in the absence of antigen ( S10E and S10F Fig ) .
T cell control over virus replication is enabled through potent effector mechanisms that ensure rapid killing and prevent dissemination of progeny viruses [56] . However , HIV employs multiple strategies to evade T cell recognition and control . For instance , the HIV Nef protein modulates expression of MHC class I , CD28 , and other proteins involved in immune recognition to evade cytotoxic T lymphocytes ( CTLs ) [3 , 57] . Additionally , the phenomenal capacity of HIV to modify its MHC class I restricted peptide antigens promotes escape from CTL responses [4] . Moreover , due to chronic HIV persistence , CTLs become exhausted and progressively lose their effector functions [7] . For these reasons , there is a strong rationale to develop HIV-specific T cells with enhanced , supraphysiologic ability to control HIV replication for therapeutic studies aimed to establish long-term control , or a “functional cure , ” in the absence of antiretroviral treatment . We hypothesized that we could re-engineer the original CD4 CAR that was tested in the clinic and determined to be safe and long lived , but lacked potent antiviral activity , to develop T cells that were far more effective in controlling HIV infection [16–18] . We found that switching from a MMLV-based gammaretroviral vector to an HIV-based lentiviral vector resulted in much higher CAR surface expression , and expression was further augmented upon substituting the EF1α promoter for than the PGK promoter , consistent with previous cancer-targeting CAR studies [23] . Higher CD4 CAR expression in primary human CD8 T cells correlated with improved control over HIV replication . However , this was not a perfect correlation , as incorporating the CD8α TM domain rather than CD4 TM domain resulted in lower CAR expression but improved control over HIV replication . We favor two non-mutually exclusive explanations of this finding: the CD8α TM domain facilitates less HIV fusion than the CD4 TM domain , making cells less susceptible to CD4 CAR-mediated infection; and the CD8α TM domain promotes dimerization , which may potentiate signaling [23 , 37 , 39] . Indeed , improved control over HIV replication by CD4 CARs inversely correlated with the susceptibility of CAR+ CD8 T cells to become infected , suggesting that HIV infection limits CAR effector activity . We found that , despite similar levels of specific lysis and stronger cytokine production in response to Env+ K562 cells , scFv-based CARs could not control HIV replication as well as CD4-based CARs in vitro , suggesting that CD4 may recognize HIV Env expressed on the cell surface faster than the scFvs we examined . Although it is certainly possible that additional and/or better optimized scFv CARs could have superior activity than those described here , we favor the use of T cells expressing the CD4 CAR due to its extensive clinical safety profile , lack of immunogenicity , and the dependence HIV shows for using CD4 as an entry factor [22] . While escape from antibody targeting is common in HIV , the reliance on CD4-mediated entry suggests that escape from CD4 binding will impose significant , if not lethal , fitness costs . As safety and efficacy of this re-engineered CAR are demonstrated in HIV-infected individuals , combinatorial approaches with scFv based CARs may further augment control of HIV replication in the absence of ART . The improved control exhibited by the re-engineered CD4 CAR is impressive , with at least a 50-fold augmentation in control over HIV replication ( S13 Fig ) . Many studies have used similar methods to study the ability of previous CD4 CAR designs and HIV-specific T cells to limit HIV replication in vitro and an E:T ratio of 1:1 was generally required to obtain complete control , in line with what we observed with these constructs [19–21] . The superiority of CARs compared to a TCR-based approach may be due to antigen-binding affinity , target cell-binding avidity , T cell activation kinetics , or bypassing the detrimental effects HIV Nef [3] . Overall , the potent control achieved by the re-engineered CARs provides optimism for clinical utility and achieving a functional cure . We opted to use a NSG hu-T cell mouse model in which human T cells isolated from human donors would be manufactured in an analogous manner as a clinical trial . While there are certainly drawbacks to this model , such as GVHD and the inability to replace CD4 T cells once they are depleted by HIV infection , this model has been successfully used in preclinical , FDA mandated biotoxicity and efficacy studies [54 , 58] and has mirrored the outcome of several clinical trials exploring gene therapy approaches to treat HIV infection [53 , 54] . Our in vivo results demonstrated that the re-engineered CD4 CARs had potent antiviral activity , vastly superior to the original CD4-zeta CAR construct . Interestingly , we saw superior control by the 4-1BB containing CARs both early in infection within the HIV prevention model and more durable control at late timepoints within the treatment model , compared to mice treated with CD28 containing CARs . In contrast to the mouse data , our in vitro studies show that CD28 costimulation promoted higher cytokine production and better control over HIV replication , relative to a CD4 CAR containing 4-1BB costimulation . One way to reconcile the difference between our in vitro and in vivo finding is to consider CAR T cell expansion kinetics . We showed in vitro that if the CAR+ CD8 T cells are in sufficient numbers , they can prevent the spread of infection in all cells; however , if they fall below a critical level , then both the CD4 T cells as well as the CAR+ CD8 T cells are infected and HIV begins to spread . In our HIV treatment model , effective early control results in an expansion of more CD4 T cell targets . If the CAR+ CD8 T cells fail to expand in proportion to the CD4 T cells or are depleted by HIV , then they may fall below the critical threshold required to control HIV replication . Differences in the relative expansions of BBz and 28z CARs in vivo , which may partially be due to better antigen-independent expansion of BBz , may underlie the differences in HIV control observed between Figs 5 and 8 . In support of this hypothesis , we observed less 28z CD8 T cell expansion relative to BBz and this correlated to less viral control ( Fig 8 ) . Moreover , by the end of the HIV-prevention model , shown in Fig 7 , similar control by either CAR resulted when the 28z CARs expanded to a similar degree as the BBz CARs . In any case , one would predict that using CCR5 ZFNs [59] , C34 based fusion inhibitors [58] , or other methods to protect CD4 CAR expressing cells from HIV infection would further potentiate the therapeutic potential of CD4 CAR T cells . In fact , a recent paper using scFv based CARs demonstrated this by inserting the CAR into the CCR5 locus [60] . In addition , the ability to persist in the absence of antigen may be important for a functional cure , in which durable T cell control may need to last for decades . Together , our data suggests that CD4 CARs containing the 4-1BB costimulation domain may be the most effective way to deliver T cell control of HIV replication . Unfortunately , there is no animal model that faithfully mirrors HIV infection in humans , and issues of trafficking , immune privilege , and durability are difficult to fully model in humanized mice . Nonetheless , these data provide the rationale to re-visit the clinical utility of CD4 CAR in HIV-infected individuals and provide optimism for CAR T cells to achieving durable control over HIV in the in the absence of ART .
pRT43 . 2 GFP , the backbone of the original clinical trial vector , was obtained courtesy of Dr . Maribeth Eiden [19 , 61] A restriction site linker was inserted into the PstI and SalI sites , removing the CMV promoter . The PGK promoter CD4-zeta sequence was amplified from pRRL . PGK . F3 ( a gift of Dr . Tom Dull ) with oligos 5’ GTATCGATCACGAGACTAGC and 5’TTAAACCGGTGTCTGGCCTTTGAGTGGTGA and inserted into XhoI and AgeI sites in the linker within pRT43 . 2 . pTRPE CD4 zeta was created by amplifying the CD4 extracellular domain was amplified from pRRL . PGK . F3 with primers: 5' TTAATGGGATCCATGAACCGGGGAGTCCCTTT and 5' AAGGACTTCCGGATGGCTGCACCGGGGTGGACCATG-3' and inserted into the BamHI and BspE1 sites in the pTRPE backbone containing the CD8α extracellular hinge and transmembrane domains and the 4-1BB and CD3 zeta intracellular costimulatory domains [62] . pTRPE lentiviral vectors containing the CD8α hinge-CD8αTM-CD3ζ or the CD8α hinge-CD28TM-CD28-CD3ζ ICD were used as template to PCR amplify the hinge-TM-and ICD region into the BspE1 and Sal1 sites with primers:5’ GGGACACTCCGGAACCACGACGCCAGCGCCGCG and 5’ GGGACACGTCGACTTAGCGAGGGGGCA . A lentiviral vector that expressed a B57 restricted TCR capable of recognizing HIV p24Gag epitope KAFSPEVIPMF ( pTRPE B57-KF11 ) was generated by synthesizing the TCRα and TCRβ gene sequence ( IDT , the TCR sequences were a generous gift of Xu Yu and Bruce Walker [63] ) . The TCRα and TCRβ gene sequence was separated by the T2A for expression of both TCR genes as previously described [64] . VRC01 , 3BNC60 , PGT128 , and PGDM1400 scFv CARs were generated from the published parental antibody sequences , with a light-linker-heavy chain configuration [46 , 65–68] . The linker sequence is: GGSSRSSSSGGGGSGGGG . Amino acid sequences were codon-optimized ( Geneart ) and synthesized as double-stranded DNA fragments ( IDT or Geneart ) , flanked with suitable restriction sites and cloned into pTRPE plasmids with the BamHI and BspE1 sites . The PG9 scFv was obtained as a generous gift from Dr . Phil Johnson and cloned into the pTRPE plasmid with the BamHI and BspE1 sites . The amino acid sequences are found in S18 Fig . To generate lentiviral particles , expression vectors encoding VSV glycoprotein , HIV Gag and Pol , and Rev ( pTRP pVSV-G , pTRP g/p . RRE pTRP . REV ) were synthesized by DNA 2 . 0 and transfected onto HEK293T cells with pTRPE transfer vectors using the Lipofectamine 2000 transfection reagent ( Invitrogen , Life Technologies ) as previously described [69] . Transfected HEK293T supernatant was collected at 24- and 48-hour timepoints , filtered through 0 . 45 um nylon syringe filters , and concentrated by ultracentrifugation at 18 hours at 8 , 500RPM at 4°C . Medium was aspirated and pellet was resuspended in 1 . 2ml total volume and stored at -80°C . Murine retrovirus: 107 293T cells were plated and after 18 hours co-transfected with 20 ug pNGVL3-g/p , 20 ug pMSCV-RD114 , and 40 ug pMMTV CD4 zeta transfer vector also using the Lipofectamine 2000 transfection reagent ( Invitrogen , Life Technologies ) . After 24 , 48 , and 72 hours , supernatants were harvested , filtered through 0 . 45 um nylon syringe filters , and frozen at -80C . T cells were purified from normal donors by negative selection using the RosetteSep Human CD4+ or CD8+ T Cell Enrichment Cocktails according to the manufacturer’s protocols ( StemCell Technologies ) . T cells were cultured at 1x106 per mL in “complete RPMI 1640:” RPMI 1640 ( Life Technologies ) supplemented ( ThermoFisher Scientific ) wih 10% fetal calf serum ( Seradigm ) , 1% Penn Strep ( Life Technologies ) , 2 mM GlutaMax ( Life Technologies ) , and 25 mM HEPES buffer ( Life Technologies ) . T cells were stimulated with anti-CD3/CD28 coated Dynabeads ( Life Technologies ) at a 3:1 bead to cell ratio and 100–300 IU/mL of recombinant human interleukin-2 for 5 days prior to bead removal . 1 day after stimulation , 200ul of lentivirus supernatant was added to 0 . 5x106 cells so that between 40–70% of the T cells would be transduced . MMLV vector transduction was performed on days 3 and 5 , with 1ml virus supernatant added to a Retronectin ( Takara ) -coated 24 well plate and spinoculated according to the manufacturer’s instructions . Medium was doubled on day 3 and changed completely on day 5 , and then added every other day throughout cell culture , or as necessary based on cell counts . Two days after removing the anti-CD3/CD28 beads , CD4 T cells were infected with the CCR5-tropic HIV strain Bal , and 24 hours later were co-cultured at varying effector to target ( E:T ) ratios with CAR CD8 T cells . Bal viral stocks ( 280ng/ml p24 ) was prepared by harvesting the cell-free supernatant from anti CD3/CD28 activated CD4 T cells and freezing in aliquots . Activated CD4 T cells were infected by adding approximately 1ml of supernatant per 20 million cells 2–3 days after removing beads . The following day CD4 and CD8 T cells were co-cultured at varying E:T ratios and HIV spread was monitored by intracellular p24 Gag with the KC57 anti-Gag-RD1 antibody ( Beckman Coulter ) and the Invitrogen Fix and Perm buffers , according the manufacturers’ instructions , gating on a population of uninfected cells . To ensure that the same numbers of CAR+ CD8 T cells were being compared , we diluted out populations with higher transduction efficiencies by adding in nontransduced T cells until all CAR+ CD8 T cell populations matched the population with the lowest CAR or TCR transduction efficiency . CD4 CAR surface expression was monitored with mouse anti human CD8-FITC and anti-human CD4 APC antibodies ( BD biosciences ) . The scFv CARs were detected with biotinylated F ( ab' ) 2 goat anti-human IgG ( Jackson ) and Streptavidin-PE ( BD biosciences ) . HLA-DR was detected on Raji B cells and K562 cells originally obtained from the American Type Culture Collection ( ATCC ) using a mouse anti-human HLA-DR PE antibody ( BD biosciences ) . Cells were visualized on a LSR II flow cytometer ( BD Biosciences ) and analyzed using Flowjo software ( Tree Star ) as previously described [70] . B57-KF11 TCR transduction efficiency was detected with an antibody to the TCR Vβ17 chain , subtracting the background Vβ17 signal from the NTD T cells ( S5A Fig ) . In vitro killing of K562 cell derived targets was tested with a 51Cr-release assay . 5x105 target cells were loaded with 50 mCi of Na251CrO4 ( Perkin Elmer ) for 90–120 minutes , washed twice and resuspended in phenol red-free medium with 5% FBS . NTD , CD4 CAR , or scFv CAR transduced T cells ( two weeks after initial activation ) were co-incubated with loaded YU2 Env+ K562 target cells for 4 hours at various E:T ratios , and chromium release into the supernatant was measured with a MicroBeta2 plate counter ( Perkin Elmer ) . Intracellular cytokine production was measured after co-culturing 5x105 NTD , CD4 CAR , or scFv CAR transduced CD8 T cells at a 1:1 E:T ratio with the various target K562 cell populations for 6 hours . Cytokine production was detected as previously described [71] using rat anti human IL-2 APC ( BD biosciences ) , mouse anti human MIP-1β PerCP Cy5 . 5 ( BD biosciences ) , mouse anti human IFN-γ FITC ( BD biosciences ) , and mouse anti human CD107a PE ( BD biosciences ) , along with Invitrogen Fix and Perm buffers . Genomic DNA was isolated from transduced CD8 T cells with the iPrep Purification Instrument ( Thermo fisher scientific ) and qPCR analysis was performed using ABI Taqman technology , with a modified version of the previously described assay designed to detect the integrated CD4-zeta sequence in genomic DNA ( gDNA ) [22] . To determine copy number per unit DNA , a standard curve was generated consisting of 5 to 106 plasmid copies spiked into 200 ng nontransduced control gDNA . The plasmid copy number in the standard curve was verified using digital qPCR with the same CD4-z primer/probe set , and performed on a QuantStudio 3D digital PCR instrument ( Life Technologies ) . Each data-point was evaluated in triplicate with a positive Ct value and % CV less than 0 . 95% for all quantifiable values . To control for the quantity of interrogated DNA , a parallel amplification reaction was performed using 10 ng gDNA and a primer/probe set specific for a non-transcribed genomic sequence upstream of the CDKN1A ( p21 ) gene as previously described [72] . These amplification reactions generated a correction factor to adjust for calculated versus actual DNA input . Copies of transgene per cell were calculated according to the formula: [Average copies of transgene ( from qPCR ) x gDNA input Correction Factor/Input gDNA ( ng ) ]x 0 . 0063 ng gDNA/cell . 6 week old NSG ( NOD-scid IL2Rgnull ) mice were obtained from The Jackson Laboratory ( JAX ) and at 7 weeks treated with 30mg/kg Busulfan mixed 1:1 with PBS . 24 hours later mice were injected via tail vein with 10x106 human lymphocytes in 100ul 0 . 5% human serum albumin in PBS , comprised of 8 million CD4 T cells and 2 million CD8 T cells ( NTD , BBz , 28z , or MMLV-CD4z transduced with a 50% transduction efficiency ) . Three weeks later mice were tail vein injected with 15ng HIV Bal mixed 1:1 with PBS . Peripheral blood was obtained by retro-orbital bleeding , and human CD4 and CAR+ CD8 lymphocyte counts were enumerated using BD lysis buffer and BD TruCount tubes as previously described [73] , staining with mouse anti human CD45 PerCp Cy5 . 5 ( BD Biosciences ) , mouse anti human CD4 BV421 ( Biolegend ) , and mouse anti human CD8α BV711 ( Biolegend ) . 5 week old NSG ( NOD-scid IL2Rgnull ) mice were obtained from The Jackson Laboratory ( JAX ) and at 6 weeks were injected with 5 million CD8-depleted human PBMCs , and 12 days later injected with 1 million HIV Bal-infected ( or mock-infected ) autologous CD4 T cells that had been in vitro infected with HIV Bal and cultured with ART for 2 days prior to freezing . The same day as HIV infection , mice began receiving 200mg/kg daily intraperitoneal injections of the reverse transcriptase inhibitor nucleotide analog tenofovir disoproxil fumarate ( TDF ) for 4 days . On Day 16 , 5 million CD8 T cells were injected ( NTD , BBz , or 28z ) . Peripheral blood was obtained by retro-orbital bleeding , and human CD4 and CAR+ CD8 lymphocyte counts were enumerated using BD lysis buffer and BD TruCount tubes as previously described [73] , staining with mouse anti human CD45 PerCp Cy5 . 5 ( BD Biosciences ) , mouse anti human CD4 BV421 ( Biolegend ) , and mouse anti human CD8α BV711 ( Biolegend ) . RNA was extracted from 10–30μl of plasma using methods as described [74] and reconstituted in a final volume of 15ul . Prior to extraction , a uniform quantity of Replication Competent Avian Sarcoma ( RCAS ) virus spiked into each plasma sample and amplified separately to verify virus/RNA recovery and absence of PCR inhibition [75] . RNA was reverse transcribed using random hexamers and quantified by Q-PCR using the LightCycler 480 Probes Master ( Roche; Indianapolis , IN ) on an ABI 7500FAST real-time thermocycler using an in vitro transcribed RNA standard . For each sample , the Q-RT-PCR reaction was run in duplicate on 5ul RNA; no-reverse transcriptase reaction and RCAS amplification were run on one well per sample using 2 . 5ul RNA . The HIV-1 primer/probe targets the pol gene and detects all group M clades as described in [74] , and RCAS amplification used primer/probe as described in [75] . HIV-1 quantification was normalized to equivalent volumes of starting plasma . Where indicated , viral load was expressed as number of RNA copies per CD4 T cell . Culture supernatant was harvested after 7 days of co-culture and diluted 1:10 , 000 and analyzed using the commercially available p24 ELISA assay kit ( Perkin-Elmer ) . Assay protein standards ranged from 9 . 4pg/ml to 150pg/ml . Plasma was collected by centrifugation of the whole blood and diluted according to a protocol supplied by Bonnie Howell ( Merck & Co , Inc . ) . The HIV p24 Gag protein was measured using the p24 single molecule array using the Simoa HD-1 Analyzer ( Quanterix ) following the manufacturer’s instructions . Each sample was measured in duplicate and concentration calculated based on a standard curve . The average concentration of two replicates for each sample was reported . The accurate detection range was 0 . 008pg/ml to 39 . 5pg/ml . In vitro HIV replication control significance was detected using a 1-way ANOVA test , stratifying based on the E:T ratio ( p values: ns >0 . 05 , *<0 . 05 , **<0 . 01 , ***<0 . 0001 ) , using the 30:1 E:T ratio for Fig 4A . All E:T ratios are presented in the figures as a single graph due to space limits . For the mouse models , non-parametric distributions were determined and Kruskall Wallis analysis was performed and , if overall comparison showed significant differences , then Mann Whitney Test was performed for pairwise comparisons ( as samples were not powered for post-hoc analysis of multiple comparisons ) and significance results are reported on each figure ( p values: ns >0 . 05 , *<0 . 05 , **<0 . 01 , ***<0 . 0001 ) . Purified CD4 and CD8 T lymphocytes were obtained by University of Pennsylvania Human Immunology Core/CFAR Immunology Core from de-identified healthy donors . All humanized mouse experiments were approved by the University of Pennsylvania’s Institutional Animal Care and Use Committee ( Protocol 804563 ) and were carried out in accordance with recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . | Conventional T cells rarely provide durable control over HIV replication . Chimeric antigen receptors ( CARs ) alter how T cells recognize infected T cells , bypassing many of HIV’s immune escape mechanisms . Therefore , we hypothesized that T cells engineered to express these receptors would be more potent ( supraphysiologic ) and would be able to control HIV infection at much lower effector to target ratios than conventional T cells . Starting with a CAR that entered clinical trials but did not reliably reduce measures of HIV in patients , we systematically optimized the expression , structure , and activation components to increase its functionality . CD8 T cells expressing this re-engineered CAR were extremely potent at suppressing HIV in vitro , exhibiting greater than 50-fold more potency in the ability to prevent viral spread than the original CAR construct or an HIV-specific TCR approach . Importantly , in a humanized mouse model of HIV infection , these optimized CAR T cells could protect CD4 T cells from HIV-mediated depletion and could significantly delay viral rebound after the cessation of ART . These data provide optimism that a re-engineered HIV-specific CAR may be able to provide durable control of HIV replication in patients . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"blood",
"cells",
"t",
"helper",
"cells",
"hiv",
"infections",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"animal",
"models",
"model",
"organisms",
"rna",
"viruses",
"experimental",
"organism",
"systems",
"cytotoxic",
"t",
"cells",
"infectious",
"disease",
"control",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"medical",
"microbiology",
"hiv",
"t",
"cells",
"microbial",
"pathogens",
"mouse",
"models",
"viral",
"replication",
"cell",
"biology",
"virology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"viral",
"diseases",
"lentivirus",
"organisms"
]
| 2017 | Supraphysiologic control over HIV-1 replication mediated by CD8 T cells expressing a re-engineered CD4-based chimeric antigen receptor |
Viruses are obligate intracellular parasites and therefore their replication completely depends on host cell factors . In case of the hepatitis C virus ( HCV ) , a positive-strand RNA virus that in the majority of infections establishes persistence , cyclophilins are considered to play an important role in RNA replication . Subsequent to the observation that cyclosporines , known to sequester cyclophilins by direct binding , profoundly block HCV replication in cultured human hepatoma cells , conflicting results were obtained as to the particular cyclophilin ( Cyp ) required for viral RNA replication and the underlying possible mode of action . By using a set of cell lines with stable knock-down of CypA or CypB , we demonstrate in the present work that replication of subgenomic HCV replicons of different genotypes is reduced by CypA depletion up to 1 , 000-fold whereas knock-down of CypB had no effect . Inhibition of replication was rescued by over-expression of wild type CypA , but not by a mutant lacking isomerase activity . Replication of JFH1-derived full length genomes was even more sensitive to CypA depletion as compared to subgenomic replicons and virus production was completely blocked . These results argue that CypA may target an additional viral factor outside of the minimal replicase contributing to RNA amplification and assembly , presumably nonstructural protein 2 . By selecting for resistance against the cyclosporine analogue DEBIO-025 that targets CypA in a dose-dependent manner , we identified two mutations ( V2440A and V2440L ) close to the cleavage site between nonstructural protein 5A and the RNA-dependent RNA polymerase in nonstructural protein 5B that slow down cleavage kinetics at this site and reduce CypA dependence of viral replication . Further amino acid substitutions at the same cleavage site accelerating processing increase CypA dependence . Our results thus identify an unexpected correlation between HCV polyprotein processing and CypA dependence of HCV replication .
The hepatitis C virus ( HCV ) is a hepatotropic virus that has a high propensity to establish persistence . At present , more than 170 million people suffer from chronic hepatitis C [1] . Current therapy of this disease is based on the combination of pegylated interferon-alpha and ribavirin . However , sustained viral response rates are not satisfying and side-effects associated with this therapy are high . Thus , the development of alternative and more effective strategies to counteract HCV infection are of great importance . One promising drug showing potent anti HCV function is cyclosporine A ( CsA ) . The cyclic undecapeptide CsA is a secondary metabolite of the fungus Tolypocladium inflatum and was originally discovered as powerful immunosuppressive drug [2] . The immunosuppressive properties of CsA are due to its ability to block the phosphatase Calcineurin in activated T cells [3] by binding with high affinity to cyclophilins ( CyPs ) [4] . Because of the non-immunosuppressive properties combined with profound antiviral activity , CsA derivatives such as DEBIO-025 [5] , NIM811 [6] , and SCY-635 are more likely to be used as anti HCV agents . Unlike CsA , these molecules binds to CyPs but do not display calcineurin inhibition . As a member of the Flaviviridae family , HCV has a positive strand RNA genome encoding a single polyprotein that is cleaved by cellular and viral proteases into 10 different proteins ( reviewed in [7] , [8] ) . The structural proteins core , envelope protein 1 ( E1 ) and E2 reside in the N-terminal region of the polyprotein and they are the major constituents of the virus particle . Virus assembly and release requires p7 , a presumed viroporin [9]–[11] and nonstructural protein 2 ( NS2 ) that contains a complex N-terminal trans-membrane domain and a C-terminal protease domain responsible for cleavage between NS2 and NS3 [12] , [13] . The latter is composed of two domains , an N-terminal protease domain which is activated via the NS4A cofactor and a C-terminal helicase domain . NS4B most likely plays a major role in the induction of membrane alterations that are required for the assembly of viral replication complexes . NS5A is an RNA binding protein required for replication and virus assembly [14]–[16] and NS5B is the RNA-dependent RNA polymerase ( RdRp ) . Cleavage of the polyprotein in the NS3 to NS5B region is mediated by the NS3/4A protease complex . Processing at the NS3-4A site is a rapid intramolecular reaction . Subsequent cleavages take place intermolecularly in the following preferred , but not obligatory order: NS5A-B , NS4A-B , NS4B-5A [17] , [18] . Replication of HCV occurs in the cytoplasm of infected cells in distinct virus-induced compartments designated the membranous web [19] . It is a complex membrane network composed of an accumulation of membranous vesicles of various sizes and interspersed lipid droplets which are the presumed sites of HCV assembly [16] , [20]–[22] . It is thought that core and NS5A are key players in mediating transfer of viral proteins and RNA from viral replication complexes to lipid droplets to trigger assembly [16] , [21] , [22] . Our knowledge about cellular proteins required for HCV RNA replication and virus assembly is scarce , but recent studies suggest that cyclophilins ( Cyp ) play an important role . Cyps are molecular chaperones catalyzing the cis-trans isomerization of proline residues and hence are called peptidyl-prolyl cis-trans-isomerases ( PPIases ) . Up to now 16 Cyp members have been identified with 7 major members found in humans ( CypA-E , Cyp40 , CypNK; [23] ) . Cyps share a common domain , but other than their PPIase activity differ in subcellular distribution and function . Conflicting data exist as to which Cyp is required for HCV replication . Watashi and coworkers reported that CypB is important for viral replication and they observed a direct binding of CypB to NS5B which results in enhanced RNA binding and thus increased RNA polymerase activity [24] . Others reported that CypA or CypA , B and C are required for HCV replication [25] , [26] . For both CypA [26] and CypB [24] , [27] , [28] direct interactions with NS5B were reported arguing for a direct involvement of cyclophilins in HCV RNA replication . However , the underlying mechanism is not known . We report here that an enzymatically active CypA is essential for HCV replication and describe an interrelation between CypA requirement for HCV replication and processing kinetics at the NS5A-B site .
Contrary opinions about which type of Cyp is essential for HCV prevail in current literature [24] , [26] . To clarify these contradictions we established a panel of cell lines derived from the highly permissive Huh-7 subclones Huh7 . 5 and Huh7-Lunet by retroviral transduction of shRNAs targeting the 3′ non-coding region of CypA or CypB ( Fig . 1 ) . Cell pools transduced with the retroviral vector expressing an unrelated shRNA were generated as a control . Transduced shRNAs lead to a profound and stable reduction of the expression of CypA or CypB in the respective cell pools as detected by Western blot ( Fig . 1A ) and immunofluorescence analysis ( Fig . 1B–C ) . To determine the impact of reduced CypA or CypB expression on HCV RNA replication , cells were transfected with a subgenomic JFH1 luciferase reporter replicon RNA ( sgNS3/JFH1-Luc; Fig . 2A , upper panel ) . Cells were lysed 4 , 24 , 48 and 72 h after transfection and replication was scored by measuring luciferase activity in cell lysates . Luciferase activity was normalized to the 4 h value , reflecting transfection efficiency . Replication was not affected by CypB knock-down and was comparable between naïve and vector control-transduced Huh7-Lunet or Huh7 . 5 cells ( Fig . 2A , left and right panel , respectively ) . However , when we transfected the same replicon into cells with the stable CypA knock-down luciferase activity was reduced more than 100-fold at the early time point ( 24 h post transfection ) in case of Huh7-Lunet cells and about 50-fold in case of Huh7 . 5 cells . Only at the latest time point ( 72 h post transfection ) luciferase activities were comparable in all transfected cell lines . This kinetic is due to the fact that replication of JFH1 is limited by the host cell [29] . For this reason RNA replication phenotypes are often only detectable at early time points after transfection before host cell factor ( s ) ( other than CypA ) become limiting and thus restrict HCV replication in naïve cells . Therefore , HCV replication in cells with the CypA knock-down can ‘catch up’ until it reaches wild type level [30] . By using the luciferase values ( which because of the very short half life of luciferase are ideally suited to determine replication kinetics ) we calculated the apparent doubling time of HCV RNA during the exponential phase in the various cell lines . We found that apparent doubling time in naïve and control cells were in the range of 3 . 7 h whereas apparent doubling time was 11 . 4 h in Huh7-Lunet cells with stable CypA knock-down . In case of Huh7 . 5-derived cell lines apparent doubling times were 3 . 6 h and 7 . 4 h for control and CypA knock-down cells , respectively . Thus results obtained with both cell lines were remarkably similar arguing against a cell line-specific phenotype . Given the controversial discussion on genotypic differences with respect to Cyp-dependence of HCV replication , we also determined the impact of CypA and CypB knock-down on replicons derived from the genotype 1b isolate Con1 . As shown in Fig . 2B , replication of a highly adapted Con1-derived replicon ( Con/ET; [31] ) was completely blocked in Lunet cells with CypA knock-down as deduced from the comparable drop of luciferase expression in cells transfected with this replicon or the replication defective NS5B active site mutant ( Con/GND ) . In contrast , replication of Con1/ET was completely unaffected by knock-down of CypB expression . In case of Huh7 . 5 derived cell lines that support Con1 RNA replication to a much lower extent , we also observed a complete inhibition of replication in CypA knock-down cells , but in some experiments a slight inhibition in cells with a CypB knock-down was found as well ( Fig . 2B , right panel ) . The latter was however , at the limit of statistical significance ( Mann-Whitney-Test ) . Finally , we studied whether CypA dependence of HCV replication is specific for this virus or applies to other flaviviruses by analyzing replication of a Dengue virus 2 ( New Guinea C isolate ) replicon in the same cell lines . We found that RNA replication of Dengue virus was not affected by CypA or CypB knock-down in Huh7-Lunet cells ( Fig . S1 , left panel ) whereas in Huh7 . 5 cells with a CypA knock-down a slight reduction of replication was found ( right panel ) . However , also in this case this difference was at the limit of statistical significance . In summary , we concluded that CypA is an essential host cell factor required for RNA replication of HCV but not Dengue virus , whereas CypB appears to play no role for HCV replication . To study the impact of these knock-downs on replication of full length HCV genomes we used a bicistronic reporter genome ( Jc1-Luc; Fig . 3A , upper panel ) . It is derived from the highly assembly competent chimera Jc1 composed of the genotype 2a isolate J6 and JFH1 that are fused at a distinct site in the N-terminal trans-membrane domain of NS2 [32] ( Fig . 3A ) . For this and all subsequent studies phenotypic analyses were performed primarily in Huh7-Lunet cells because of the more potent knock-down of CypA and CypB expression as compared to Huh7 . 5-derived cell lines ( Fig . 1 ) . A very potent inhibition of Jc1-Luc replication was found in CypA , but not CypB knock-down cells ( Fig . 3A , lower left panel ) . In fact , the inhibition was much stronger compared to the one observed with the subgenome and replication did not recover even at later time points . Infectivity was not detected in supernatants of CypA knock-down cells ( Fig . 3A , right panel ) , but not significantly reduced in supernatants of CypB knock-down cells . The profound inhibition of Jc1-Luc replication in CypA knock-down cells might be due to the less robust replication of this bicistronic genome compared with the sgNS3/JFH1-Luc replicon [33] . Alternatively , CypA dependence of JFH1 full length HCV genomes and subgenomes may differ . We addressed these possibilities by transfection of CypA knock-down cells with a full length genome lacking heterologous sequences . Owing to highest assembly efficiency we used the Jc1 chimera ( Fig . 3B ) . Viral replication was scored by determination of intracellular core protein amounts whereas virus production was determined by measurement of core released into the supernatant of transfected cells as well as infectivity titers in culture supernatant ( TCID50; see materials and methods ) . Replication of the Jc1 genome was potently reduced in CypA knock-down cells , but unaffected by CypB knock-down . In addition to impaired replication , core release was also reduced as deduced from the relative core protein amounts in cell lysate and culture supernatant . In agreement with this profound block of replication and impaired assembly no infectivity was detected in supernatants of CypA knock-down cells even though virus titers in control cells were in the range of 105 TCID50/ml and higher . The congruent results obtained with the genomic reporter replicon and the Jc1 genome suggest that CypA targets an additional viral factor outside of the minimal replicase ( NS3 to NS5B ) . We assumed that NS2 would be the most likely candidate for three reasons . First , cleavage at the NS2-3 site by NS2 is essential for RNA replication [34]; second , the structural proteins core , E1 and E2 as well as p7 are dispensable for RNA replication [35]; third , NS2 is required for HCV assembly [12] , [13] . To put this assumption to the test we performed comparative transient replication assays by using subgenomic JFH1 replicons encoding either an NS2 to NS5B or NS3 to NS5B replicase ( sgNS2/JFH1-Luc and sgNS3/JFH1-Luc , respectively ) ( Fig . 3C ) . Replication of the sgNS2 replicon was almost completely blocked in Huh7-Lunet cells with a CypA knock-down and only slightly above background as determined with the replication defective NS5B active site mutant JFH1/ΔGDD . In contrast , replication kinetic of the sgNS2 RNA was well comparable between Huh7-Lunet cells with a CypB knock-down and control cells that had been transduced with a non-targeting luciferase shRNA . We note a lower replication of the sgNS2 replicon in CypB knock-down cells at 24 h post transfection as compared to the sgNS3 replicon , but this reduction was also observed with control cells ( Fig . 3C ) . At later time points , replication of sgNS2 and sgNS3 replicons was comparable . In summary these results provide genetic evidence that CypA also targets a factor outside of the replicase , most likely NS2 . Although the results described so far clearly support an important role of CypA for HCV replication , we could not exclude that the observed phenotype was due , at least in part , to off-target effects . For this reason , we transduced the Huh7-Lunet CypA knock-down cell lines with a CypA expression construct that was not targeted by the shRNA ( ‘rescue’ cell line in Fig . 4 ) . In addition , we generated CypA knock-down cell lines expressing a CypA mutant ( H126Q ) that retains less than 1% of wild type isomerase activity [36] , [37] . As shown in Fig . 4A and B , CypA expression levels achieved upon transduction of the shRNA-resistant CypA genes was very high ( Fig . 4A ) . In fact , by using quantitative immunofluorescence we found that expression level of CypA in wild type ‘rescue’ cell lines was about two-fold higher as compared to sh-control cells ( 234+/−4 . 7 counts per cell based on 8-bit gray value histograms , generated with the ImageJ software package , versus 125+/−25 . 3 counts per cell , respectively ) see materials and methods ) . Knock-down of CypA had no effect on expression levels of CypB ( Fig . 4B , lane 7 ) and vice versa ( data not shown ) . Upon transfection of CypA wt ‘rescue’ cells with sgNS3/JFH1-Luc , RNA replication was fully restored ( Fig . 4C ) thus excluding off-target effects of the shRNA that was used for silencing of CypA expression . However , no rescue was obtained in cells stably transduced with the H126Q mutant ( Fig . 4C ) , arguing that isomerase activity of CypA is essential for HCV RNA replication . To gain further insight into the role of CypA for replication of HCV , we wanted to take advantage of the fact that cyclosporine ( CsA ) binds to and sequesters CypA [4] and causes a profound inhibition of HCV replication [25] , [38] . Moreover , we have recently shown that the cyclosporine analogue DEBIO-025 also inhibits replication of HCV , but the cellular target of this compound was not clear [5] . To clarify this question we first compared the antiviral activity of CsA and DEBIO-025 side-by-side in a JFH1 replicon system . Cells transfected with the sgNS3/JFH1-Luc replicon that expresses the firefly luciferase ( Fig . 2A , upper panel ) were treated 4 h post transfection for 72 h with various concentrations of CsA or DEBIO-025 . Cells were then lysed and luciferase activity was determined . The JFH1 subgenome was potently inhibited both by CsA and DEBIO-025 , with the latter having a higher antiviral activity ( about 3–4 fold lower IC50 and IC90 ( Fig . 5A ) ) . We next determined whether CypA is targeted by DEBIO-025 using transient replication assays with the same subgenomic JFH1 RNA . Transfected cells were treated with various concentrations of DEBIO-025 and after 72 h luciferase activity contained in cell lysates was determined . The results depicted in Fig . 5B show that HCV replication was much more sensitive to DEBIO-025 treatment in CypA knock-down cells whereas IC50 and IC90 values determined in CypB knock-down cells were comparable to naïve and control shRNA-transduced cells . Moreover , when dose-response assays were performed with CypA wild type ‘rescue’ cell lines , DEBIO-025 sensitivity was reduced even to beyond the level determined for naïve cells ( Fig . 5C ) . This enhanced resistance is probably due to the elevated expression levels of CypA in the rescue cell line as compared to naïve cells . Most importantly , DEBIO-025 sensitivity was not affected by transduction of the gene containing the CypA isomerase mutation ( H126Q ) suggesting that an enzymatically active CypA is required for HCV RNA replication or that this CypA protein binds DEBIO-025 with lower efficiency . In summary , our results show that CypA is the target for the antiviral activity of DEBIO-025 and that isomerase activity of CypA appears to be required for HCV replication . Given the finding that DEBIO-025 can be used as a convenient tool for the pharmacological inhibition of CypA and to study its role in HCV RNA replication , we sought to identify the target of CypA in the HCV replicase by selection for DEBIO-025 resistance . Due to the high potency of DEBIO-025 and the small window between IC50 and IC90 ( 0 . 2 µg/ml and 0 . 5 µg/ml; Fig . 5A ) , attempts to select for resistance by using virus passage or cells persistently infected with Jc1 were not successful . Concentrations in the range of the IC90 or above lead to rapid elimination of HCV from cell cultures and treatment with a dose corresponding to the IC50 did not sufficiently affect replication . We therefore utilized a selectable JFH1-derived subgenome that carried a stable in-frame insertion of the red fluorescent protein ( RFP ) in domain 3 of NS5A [39] , which allowed visualization of replicating HCV via RFP autofluorescence in live cells . Huh7-Lunet cells were transfected with this sgNS3/JFH1-neo/RFP replicon and cultured in double-selection medium containing 500 µg/ml of G418 and 0 or 0 . 5 or 1 µg/ml of DEBIO-025 . After 18 cell passages , we had selected for a DEBIO-025 resistant cell pool with a calculated IC50 of about 1 µg/ml corresponding to an about 5-fold lower sensitivity as compared to non-selected cells ( see Fig . 5A ) . To identify mutations that confer increased DEBIO-025 resistance and thus presumably reduced CypA dependence , replicons present in two independent RNA preparations of the cell pool were prepared from DEBIO-025 selected cells . From each RNA preparation the complete HCV coding region was amplified in two overlapping fragments ( Fig . 6A ) and 3 or 4 molecular clones of each fragment were subjected to nucleotide sequence analysis , respectively . As summarized in Table 1 and Fig . 6A , 4 mutations were identified residing in the NS3 helicase domain ( Y1421F ) or at various positions in NS5A . Two of these were mapped to domain 2 of NS5A ( D2229G , L2266F ) and one to the very C-terminus of domain 3 ( V2440A ) . Quite surprisingly , we have recently identified position 2440 , which is close to the NS5A-B cleavage site , as a major determinant of HCV assembly [40] . However , in case of this virus titer enhancing mutation ( TEM ) the substituting amino acid residue was leucine instead of the alanine found in the DEBIO-025 resistance selection . Multiple sequence alignments covering HCV isolates of all genotypes as deposited in the European HCV database ( euHCVdb ) [41] revealed that all positions identified in DEBIO-025 selected replicons are polymorphic ( Table 1 ) . The mutations residing in the NS3 helicase and at the C-terminus of domain 3 of NS5A appear to be natural variants , because the substituting residues ( phenylalanine in case of the helicase and alanine in case of NS5A ) are present in patient isolates ( Table 1 ) . Other substituting residues are naturally found in case of the two mutations in domain 2 of NS5A . The impact of these 4 selected mutations on DEBIO-025 sensitivity was determined by transferring each mutation individually into a wild type JFH1 luciferase replicon ( sgNS3/JFH1-Luc; Fig . 2A ) and determining replication fitness ( Fig . 6C ) as well as DEBIO-025 sensitivity ( Fig . 6B ) . In this and all subsequent analyses we included the TEM V2440L , because it resides at the very same position like the putative resistance mutation V2440A . The results in Fig . 6B demonstrate that only the mutations affecting residue 2440 in NS5A reduced DEBIO-025 sensitivity to a significant extent . In fact , the IC50 of replicons containing the V2440A or the V2440L substitution was increased about 5- to 10-fold ( Fig . 6B ) whereas the mutations in domain 2 of NS5A had either minor or no effect . Replication fitness of only the V2440A mutant was reduced ( about 10–15-fold as compared to the wild type ) whereas all other mutants were unaffected ( Fig . 6C ) . Since V2440L is a TEM , we next assessed the impact of this and the V2440A substitution on virus production . The two mutations were introduced into the JFH1 wild type genome and virus titers released from cells at various time points after transfection were determined ( Fig . 6D ) . In agreement with our earlier report [40] the V2440L substitution enhanced virus production without affecting RNA replication , whereas the V2440A substitution reduced virus production to an extent corresponding to the lower replication . These results suggest that only V2440L is a TEM . The data described so far show that CypA is an important cellular determinant of HCV RNA replication and they suggest that DEBIO-025 blocks HCV replication by interfering with CypA . Having identified V2440L/A as DEBIO-025 resistance conferring mutations we assumed that they might render HCV replicons less dependent on CypA . To corroborate this assumption we performed transient replication assays with JFH1 wild type , JFH1 V2440L and JFH1 V2440A sgNS3-replicons in CypA knock down cells ( Fig . 7A and B , respectively ) . In line with our assumption we found that replication of the V2440L mutant was less impaired in CypA depleted cells . The analogous result was found when we scored replication by immunofluorescence of transfected cells ( not shown ) . In case of the V2440A mutation , due to reduced replication fitness only a slight but statistically significant difference ( p<0 . 002 , α = 0 . 05; for values 72 h and 96 h post transfection , respectively ) compared to the wt replicon was apparent ( Fig . 7B ) . These data suggest that the mutations close to the C-terminus of NS5A conferring DEBIO-025 resistance render HCV replication less dependent on CypA . We have recently shown that the V2440L substitution leads to delayed cleavage at the NS5A-B site [40] . Having identified another substitution at the same site that also confers DEBIO-025 resistance and lower dependence on CypA ( V2440A ) , we wondered whether this mutation affects cleavage kinetics as well . We therefore performed pulse-chase experiments using a T7-based expression system . Huh7-Lunet cells stably expressing the T7 RNA polymerase were transfected with NS3 to NS5B polyprotein expression constructs corresponding to the wild type or containing the V2440L or the V2440A mutation . Proteins were radiolabeled metabolically with [35S] methionine/cysteine for 90 minutes and treated with non-radioactive medium for one or two hours . Cells were lysed and NS5A- and NS5B-containing proteins were isolated by immunoprecipitation with mono-specific antibodies ( Fig . 8A and B , respectively; a quantification of the autoradiogram shown in panel A is given in panel C ) . We found that both mutations slowed down cleavage kinetics at the NS5A-B site , with the V2440A mutation causing a much stronger delay of cleavage than the leucine substitution ( Fig . 8A , B , lanes 5–7 ) . Best visible was the accumulation of an uncleaved NS5AB precursor protein as well as a precursor with an apparent molecular weight of about 175 kDa corresponding most likely to uncleaved NS4B5AB [17] . Only small amounts of these precursors were found in case of the wild type reflecting faster cleavage at the NS5A-B site . These results suggest a link between processing kinetic at this site and CypA dependence of HCV replication . To substantiate this observation , we generated additional mutants affecting the NS5A-B cleavage site . Selection of these mutations was guided by a multiple alignment of amino acid residues at the P7-P1 positions of the NS5A-B cleavage site in HCV isolates deposited in the European HCV database ( euHCVdb ) [41] ( Fig . 9A ) . Interestingly , amino acids at positions P7-P4 of the JFH1 isolate are rather unique amongst all the other genotypes including genotype 2 ( Fig . 9A , upper half ) . In case of genotype 2a the P5 and P4 residues are serine and valine , respectively . Genotype 2b isolates have in addition a P7 glutamic acid residue and a P3 isoleucine residue . Based on this alignment we constructed two cleavage site mutants corresponding to the NS5A-B site of genotype 2a and ( with the exception of the P3 isoleucine ) to genotype 2b ( Fig . 9A , lower half ) . In addition these mutations were combined with the P3 leucine substitution ( V2440L ) to determine its contribution on DEBIO-025 resistance , CypA dependence and cleavage kinetics in the context of the 2a and 2b cleavage site . As shown in Fig . 8 , cleavage between NS5A and NS5B was enhanced with mutants 2a-5AB and 2b-5AB . A comparison of the protein pattern obtained after immunoprecipitation with the NS5A-specific antibody revealed that the NS5AB precursor was well detectable right after labeling in case of the wild type ( Fig . 8A , lane 2 ) and this protein remained detectable throughout the chase period , albeit at low amounts ( lane 3 , 4 ) . In case of the 2a and the 2b mutants the amounts of this precursor were lower after the 1 h labeling period and it was not detectable at later time points ( lane 11–16 ) . However , upon insertion of the V2440L substitution into each of these mutants , processing was again delayed compared to the wild type and cleavage kinetic was similar to the one observed with the V2440L single mutant ( compare lanes 17–22 with 8–10 in Fig . 8A–B and quantification in Fig . 8C ) . Upon transfection of Huh7-Lunet cells with replicons containing these mutations , we found that save for the 2b-5AB mutant , which was slightly impaired , replication of all the other mutants was well comparable to wild type ( Fig . 9B ) . To our great surprise mutants 2a-5AB and 2b-5AB showed an about 10-fold increased DEBIO-025 sensitivity ( Fig . 9C ) and replicated less efficiently in stable CypA knock-down cells ( Fig . 9D ) . However , the analogous mutants containing in addition the V2440L substitution were again less sensitive to DEBIO-025 ( Fig . 9C ) and replicated more efficiently in CypA knock-down cells ( Fig . 9D ) . These results reveal a correlation between cleavage kinetics at the NS5A-B site , DEBIO-025 resistance and CypA dependence ( Table 2 ) . Delay of cleavage kinetics correlates with low DEBIO-025 sensitivity and reduced CypA dependence whereas ‘hyper-processing’ correlates with increased DEBIO-025 sensitivity and CypA dependence . Given the link between the V2440L mutation and enhanced HCV assembly , we tested all cleavage site mutations for their effect on virus production . As shown in Fig . 9E all cleavage site mutations increased virus titers and kinetics . With the exception of the 2b-5AB V2440L mutant , peak titers were reached already 24 h post transfection with all mutants and titers were elevated 100- to 1 . 000-fold . Especially for the 2a-5AB mutant , already 7 h after transfection infectious virus was detected in the culture supernatant ( compare 2a-5AB also with the very efficient Jc1 in Fig . 9E ) . A significant , but less pronounced enhancement of kinetics and overall production of virus was achieved with mutant 2b-5AB . When the V2440L substitution was introduced into these mutants , kinetics were reduced , best visible with the 7 h values ( Fig . 9E ) . In summary , our results reveal a correlation between cleavage kinetics at the NS5A-B site , DEBIO-025 resistance and CypA dependence . The data also suggest that cleavage kinetics at the NS5A-B site and virus assembly may be linked .
Exploitation of host cell machineries to achieve efficient replication is a strategy shared by all viruses . In case of HCV , Cyps have gained increasing interest because of the possibility to interfere pharmacologically with their activity thus providing a new approach for antiviral therapy [42] . In the present study we demonstrate that CypA but not CypB is the major player for HCV replication . This conclusion is supported by the profound impairment of replication by stable CypA , but not CypB knock-down and the rescue of viral replication by expression of CypA in these cells . Rescue was not achieved by expression of the active site mutant H126Q . This mutation resides in the hydrophobic pocket of CypA and causes a reduction of PPIase activity to about 1% of the wild type [36] , [37] . Although this result argues for a role of PPIase activity in HCV replication , it should be kept in mind that CypA mutants that are impaired in isomerase activity have reduced substrate binding [37] , [43] . Thus , PPIase activity and substrate binding cannot be separated genetically and therefore , the exact role of isomerase activity in HCV replication remains to be determined . Cyps are a diverse group of proteins that share a PPIase activity and a 109 amino acids long Cyp-like domain [23] . There are 7 major Cyps in humans that differ in their N- and C-terminal sequences , which determine the intracellular localization of a given Cyp . For instance , CypA ( as well as Cyp40 and CypNK ) are cytosolic whereas CypB and CypC possess N-terminal signal sequences , which target the proteins to the ER-resident secretory machinery . CypD has a signal sequence recruiting the protein to mitochondria and CypE is localized in the nucleus . Although the role of Cyps in HCV replication has been established in several reports , conflicting results exist as to which individual Cyp is required . Two studies identified CypB , but not CypA , as the key player in HCV replication [24] , [25] . One study describes that CypA , B and C are required [44] and the study by Yang and co-workers suggests that CypA , but not CypB and C is critical for HCV replication [26] . The results presented in this report are in full support of the work by Yang and colleagues . We also identify CypA as the key player in HCV replication and show that this host factor is the target of the cyclosporine analogue DEBIO-025 . The reasons for the discrepant results is not known , but may in part be due to the use of different cell systems , HCV isolates or experimental conditions such as utilized siRNAs . However , we note that in our study HCV replication was impaired upon CypA knock-down both in Huh7-Lunet and Huh7 . 5 cells; two widely used Huh7-derived cell clones that are highly permissive for HCV replication . We also note that the impairment of HCV replication was found both with the JFH1 and the Con1 isolate similar to what Yang and colleagues described [26] . In support of these findings , a very recent study also describes CypA as the main actor in HCV replication [45] . Considering the subcellular localization , the cytosolic CypA is more likely to interact with the HCV replicase that also resides on the cytosolic side of the ER membrane , in contrast to the ER luminal CypB . In addition , CypB and C are expressed 10–150-fold lower in hepatoma cells as compared to CyPA thus increasing the likelihood of the latter to interact with the viral replicase [26] . In an attempt to study the mode-of-action of CypA , we wanted to exploit the possibility to interfere with this protein pharmacologically by using DEBIO-025 , a compound that shows promise in clinical trials for the treatment of chronic hepatitis C [42] . The advantage of this compound is its higher anti-HCV activity as compared to CsA and reduced immunosuppressive activity , the latter being due to reduced calcineurin affinity [5] , [46] . As shown here , dose-response assays with cell lines expressing various levels of CypA confirm that DEBIO-025 indeed targets this host cell factor: HCV replication can be sensitized to the antiviral activity of DEBIO-025 by stable knock-down of CypA whereas over-expression of CypA leads to partial DEBIO-025 resistance . Although these results confirm CypA as a DEBIO-025 target , it is not known whether other Cyps are also affected by this compound , because interference with these proteins ( e . g . CypB ) has no effect on HCV replication , which was used as a read-out in our assays . By using replicons to select for DEBIO-025 resistance we identified Y1421F ( Y391F in NS3 ) that resides in the NS3 helicase domain , two mutations in domain 2 of NS5A ( D2229G/D253G; L2266F/L290F ) and one mutation at the P3 cleavage site position of the NS5A-B junction ( V2440A/V464A ) . Among these mutations , only V2440A ( V464A in NS5A ) reduced DEBIO-025 sensitivity and rendered HCV replication less dependent on CypA . We had recently identified a mutation at the very same site that enhances virus assembly about 100-fold without affecting RNA replication ( V2440L/V464L ) [40] . As shown here this mutation also confers DEBIO-025 resistance , renders HCV replication less dependent on CypA and slows down processing at the NS5A-B site . Results obtained with additional mutations at the NS5A-B site revealed a correlation between processing kinetics and CypA dependence ( Table 2 ) : mutants with slower cleavage kinetic require lower level of CypA and are less sensitive to DEBIO-025 treatment compared to wild type whereas mutants with accelerated cleavage kinetic are more dependent on CypA and more sensitive to DEBIO-025 treatment . So far we can only speculate about the underlying mechanism , but for any model the following observations should be considered: One possible explanation for our observations is that the CypA binding site in the C-terminal region of NS5B might be accessible only in an extended ( non-folded ) conformation right after protein synthesis . Proper folding of NS5B may require liberation of the N-terminus from the polyprotein ( i . e . cleavage at the NS5A-B site ) . NS5B folded in the absence of CypA may adopt a conformation that is enzymatically inactive or that is not competent to be incorporated into the replicase complex [48] . In this model binding of CypA to NS5B would be required to induce a conformation necessary for replicase formation or activity . Binding however , would be possible only during a short time period , i . e . prior to or shortly after release of full length NS5B from the polyprotein . In case of a delayed processing ( V2440L/A ) the time during which the CypA binding site is accessible is extended and therefore lower amounts of CypA would suffice to bind to NS5B . Nevertheless , also under these circumstances CypA is required for HCV replication thus explaining why these replicons are still susceptible to DEBIO-025 , albeit to a lesser extent . In case of hyper-processing the time for CypA binding would be much shorter and therefore the chance for interaction with NS5B would be lower . This can however be compensated by higher CypA amounts . This model would also explain why the mutations modulating NS5A-B processing do not have an effect on RNA replication in naïve cells . Under normal conditions , CypA levels are not limiting and even in case of the hyper-processing mutants CypA amounts are sufficient to bind to NS5B . Only when CypA levels are reduced either by knock-down of CypA expression or by DEBIO-025 treatment the chance that CypA can bind to NS5B during the much shorter time period , compared to wild type , becomes apparent . In contrast , delaying processing will extend the time for CypA binding and therefore even lower CypA amounts would be sufficient for binding to NS5B . However , a too slow processing at the NS5A-B site appears to impair replicase activity as deduced from the reduced replication of the V2440A mutant . The CsA resistance mutations residing in domain 2 of NS5A that have been described in a recent study [27] may affect NS5B activity indirectly by altering binding of NS5A to NS5B . It has been shown that NS5A binds to NS5B and this interaction appears to modulate RdRp activity [49] . Alternatively , these mutations may affect directly a function of NS5A required for RNA replication [50] or polyprotein cleavage kinetics , but these possibilities need to be addressed . Finally , the impact of altered cleavage kinetics on HCV assembly might be an epiphenomenon caused by the mutations in domain 3 of NS5A itself that is a major determinant of virus assembly [15] , [16] . Another surprising observation we made is the higher sensitivity of replication of JFH1 full length genomes towards CypA depletion and the impairment of virus production . This observation suggests that CypA may have an additional target outside of the minimal replicase . The most likely candidate is NS2 , because it contributes indirectly to RNA replication and is required for assembly [12] , [13] whereas all other proteins residing in the N-terminal domain of the polyprotein ( core , E1 , E2 , p7 ) are completely dispensable for RNA replication . In fact , the stronger inhibition observed with the JFH1 sgNS2-replicon containing NS2 and the impairment of HCV assembly supports this assumption . Since NS2 contributes to RNA replication indirectly by cleaving off itself from NS3 , a possible explanation is that CypA supports proper folding of NS2 . However , in addition to a viral factor CypA may also be required for host cell factor ( s ) contributing to virus production . In fact , preliminary results suggest that CypA may be required for integrity of lipid droplets ( A . Kaul , C . Berger , and R . Bartenschlager , unpublished observation ) , which play a key role in HCV assembly [20]–[22] . Further studies will be required to identify CypA-dependent viral and host cell factors required for HCV RNA replication , assembly and eventually egress .
All cell lines were grown in Dulbecco's modified minimal essential medium ( DMEM; Life Technologies , Germany ) supplemented with 2 mM L-glutamine , nonessential amino acids , 100 U/ml of penicillin , 100 µg/ml of streptomycin , and 10% fetal calf serum . The experiments were performed either in Huh7-Lunet cells supporting high level RNA replication or Huh7 . 5 cells that are highly infectable [51] , [52] . For selection of DEBIO-025 resistant JFH1 replicons Huh7-Lunet sgNS3/JFH1-neo/RFP replicon cells [39] were double-selected for 18 cell passages with 500 µg/ml of G418 and 0 or 0 . 5 or 1 µg/ml of DEBIO-025 . For transient dose response experiments subgenomic Luc-JFH1 replicons were transfected into Huh7-Lunet cells and 4 h post electroporation cells were treated with various concentrations of Cyclosporine A ( CsA ) or DEBIO-025 , respectively . Seventy two hours later the impact on replication was determined by luciferase assay . All nucleotide and amino acid numbers refer to the JFH1 genome ( GenBank accession no . AB047639 ) or the Con1 genome ( GenBank accession number AJ238799 ) . The chimera Jc1-Luc was described recently [33] . For replication analyses the subgenomic reporter replicon pFKI389Luc/NS3-3′_dg_JFH ( abbreviated as sgNS3/JFH1-Luc ) and the replication deficient mutant carrying a deletion of the NS5B active site ( sg/luc/JFH1/ΔGDD ) were used [39] . The basic Con1 construct pFK-rep PI-luc/ET has been described somewhere else [31] . To generate the subgenomic JFH1 replicon construct pFKI389Luc/NS2-3′_dg_ ( designated sgNS2/JFH1-Luc in this report ) we performed two separate PCRs: the first one with sense primer S/EMCV190 ( 5′-AATGCAAGGTCTGTTGAATGT-3′ ) and antisense primer A/EMCV-NS2 ( 5′-CGTGCACAGGTGCGTCATACATGGTATCATCGTGTTTTTCA-3′ ) and pFKI389Luc/NS3-3′_dg_[39] as template; the second one with primers A/JFH1/2894 ( 5′-TGACGGCCCACGCGATGCCAT-3′ ) and S/EMCV-NS2 ( 5′-TGAAAAACACGATGATACCATGTATGACGCACCTGTGCACG-3′ ) and pFKI389Luc/Core-3′_dg_JFH [32] as template . Fragments were combined by overlap-PCR using S/EMCV190 and A/JFH1/2894 and the resulting DNA fragment was inserted into pFKI389Luc/Core-3′/DelE1E2_dg_JFH [32] . The final construct pFKI389Luc/NS2-3′_dg_JFH contains ( 5′ to 3′ ) the T7 promoter sequence fused to nucleotides 1 to 389 of the JFH1 consensus sequence , the firefly luciferase gene , the encephalomyocarditis virus ( EMCV ) IRES , the NS2 to 5B coding sequence , the 3′ NTR of JFH1 , the hepatitis delta virus genomic ribozyme ( dg ) and the T7 terminator sequence . Amino acid substitutions were introduced by PCR-based site-directed mutagenesis and amplified DNA fragments were analyzed by automated nucleotide sequencing using an ABI 310 sequencer ( Applied Biosystems , Darmstadt , Germany ) . For CypA and CypB knock-down microRNA-based shRNA lentiviral vectors were produced by co-transfecting 293T cells with transfer vectors encoding the puromycin resistance gene and a shRNA targeting CypA , CypB , or a control shRNA targeting luciferase , the HIV-1 packaging plasmid psPAX2 , and a VSVg expression plasmid ( pMD2 . G ) using Lipofectamine 2000 ( Invitrogen ) . The shRNA targeting sequences were: luciferase , 5′-tacaaacgctctcatcgacaag-3′ ( not present in the luciferase gene of the reporter replicon and reporter virus ) , CypA , 5′-ctggattgcagagttaagttta-3′; and CypB , 5′-gccgggtgatctttggtctctt-3′ . Transfection medium was changed the next day and viral supernatant was harvested 48 hrs after transfection , clarified by centrifugation ( 5 min at 200×g ) , and filtered through a 0 . 45 µm syringe filter ( Sarstedt ) . Huh7-Lunet and Huh7 . 5 cells were transduced on two consecutive days and placed into selection medium containing 5 µg/ml puromycin ( Sigma ) 72 h after transduction . For selected Huh7-Lunet cell pools , knock-down of CypA and CypB was stable for at least 20 passages and cell growth as well as viability was not affected ( long-term passage of Huh7 . 5 cell pools was not performed ) . For stable over-expression of CypA in CypA knock-down cells we used a lentiviral vector system . 293T cells ( 5×105 ) were seeded in each well of a 6-well cell culture plate in complete DMEM . About 24 h later , cells were transfected with lentiviral plasmids by using Lipofectamine 2000 or Lipofectamine LTX/Plus Reagent ( Life Technologies ) according to the instructions of the manufacturer . To generate the CypA rescue cell lines , 2 . 5 µg of the transfer vector pWPI-CypAwt or pWPI-CypA/H126Q , was transfected together with 2 µg of the packaging vector ( pCMV ) and 0 . 6 µg of the VSV envelope vector ( pMD . G ) into 293T cells by lipofection as described above . After 16–24 h medium ( 2 ml ) was replaced by fresh one and an additional 24 h later culture supernatant was filtered through 0 . 45 µM pore size polyvinylidene difluoride ( PVDF ) syringe filter ( Carl Roth GmbH , Karlsruhe Germany ) and used for transduction . Huh7-Lunet/CypA knock-down cells were seeded at a density of 1–2×104 cells per well of a 12-well plate and inoculated with 0 . 5–1 ml of filtered supernatant . Twenty four hours later , inoculation was repeated . Two to three days after transduction medium was replaced by fresh cell culture medium containing puromycin and/or blasticidin S . Drug concentrations were increased steadily during cell passages up to 10 µg/ml puromycin in case of Huh7-Lunet or 20 µg/ml in case of Huh7 . 5 and 20 µg/ml blasticidin S ( Life Technologies ) . In vitro transcription and electroporation of HCV RNAs was performed as described previously [40] . Transfected cells were immediately diluted into complete DMEM and seeded as required for the given assay . Total RNA was isolated from a confluent 6 cm diameter dish of Huh7-Lunet cells containing the sgJFH1-neo/RFP replicon by using the Nucleo Spin RNAII Kit ( Macherey-Nagel , Düren , Germany ) as recommended by the manufacturer . One µg total RNA and 50 pmol of primer A9482 ( 5′-GGA ACA GTT AGC TAT GGA GTG TAC C-3′ ) were used for cDNA synthesis by using the Expand-RT system ( Roche , Mannheim , Germany ) as recommended by the manufacturer . Two to four microliters of the reaction mixture were used to amplify the complete open reading frame in two overlapping fragments with the Expand Long Template PCR kit ( Roche ) according to the instructions of the manufacturer . To amplify the 5′ half of the replicon , PCR was performed with primers S59-EcoRI ( 5′-TGT CTT CAC GCA GAA AGC GCC TAG-3′ ) and A4614 ( 5′-CTG AGC TGG TAT TAT GGA GAC GTC C-3′ ) and the PCR product was inserted into sgJFH1-Luc after restriction with EcoRI and SpeI . The 3′ half of the HCV genome was amplified with primers S3813 ( 5′-GGA CAA GCG GGG AGC ATT GCT CTC-3′ ) and A9466-MluI ( 5′-AGC TAT GGA GTG TAC CTA GTG TGT GCC-3′ ) and after restriction with SpeI and MluI , the fragment was inserted into pFK-I389neo/NS3-3′/Con [35] . Sequence analysis was performed with a set of primers covering the complete replicon sequence . Huh7-Lunet and Huh7 . 5 derived cell lines were seeded onto glass cover slips in 24-well plates at a density of 2–4×104 cells per well . Three days after seeding cells were fixed by treatment with methanol at −20°C for 10 min . , washed 3 times with PBS , incubated for 30 min in 5% normal goat serum or bovine serum albumin ( diluted in PBS ) and incubated for 1 h at room temperature ( RT ) with one of the following primary antibodies: antiCyPA rabbit polyclonal antiserum ( BIOMOL International , LP ) diluted 1∶400 , or antiCypB rabbit polyclonal antiserum ( Affinity Bio Reagents ) diluted 1∶800 , or antiNS5A mouse monoclonal antibody ( 9E10 ) [53] . After an extensive wash with PBS , cells were treated with Alexa Fluor 488 or 546 conjugated antibodies , targeting rabbit or mouse IgG domains ( dilution 1∶1 , 000 in normal goat serum ) for 1 h at RT in complete darkness . Unbound secondary antibodies were removed by washing three times with PBS and once with water . DNA was stained with 4′ , 6′-diamidino-2-phenylindole ( DAPI; Molecular Probes , Karlsruhe , Germany ) for 1 min at RT . Finally samples were mounted on slides with FluormountG ( Southern Biotechnology Associates , Birmingham , USA ) and analyzed by using fluorescence microscopy . For quantitation of immunofluorescence signals raw pictures were imported into the image J software package and pixel intensities based on 8-bit gray value were determined using the histogram function of the program ( National Institutes of Health , Bethesda , MD , USA ) . Quantification of luciferase reporter activity was used to determine transient HCV RNA replication as described previously [33] . In brief , transfected Huh7-Lunet cells were resuspended in 41 ml complete DMEM and 1 . 5 ml of the suspension was seeded in each well of a 12-well plate . For each time point , duplicates of wells were harvested . For dose response experiments , 4 h post electroporation transfected cells were treated in duplicate with different concentrations of cyclosporine A ( CsA ) or DEBIO-025 and analyzed 72 h later by luciferase assay . Cells were washed once with PBS , 350 µl of lysis buffer ( 0 . 1% Triton X-100 , 25 mM glycylglycine , 15 mM MgSO4 , 4 mM EGTA and 1 mM DTT , pH 7 . 8 ) was added and freeze-thaw lysates were prepared . For each well , two times 100 µl lysate was mixed with 360 µl assay buffer ( 25 mM glycylglycine , 15 mM MgSO4 , 4 mM EGTA , 1 mM DTT , 2 mM ATP and 15 mM K2PO4 , pH 7 . 8 ) and , after addition of 200 µl of a luciferin solution ( 200 µM luciferin , 25 mM glycylglycine , pH 8 . 0 ) , measured for 20 s in a luminometer ( Lumat LB9507; Berthold , Freiburg , Germany ) . Kinetic of replication was determined by normalizing the relative light units ( RLU ) of the different time points to the respective 4 h value . For dose response experiments the RLU obtained with cyclosporine A ( CsA ) or DEBIO-025 treated cells were normalized to the corresponding values obtained with untreated cells . A total of 2 . 5×105 Huh7-Lunet/T7 cells [54] were seeded in each well of a 6-well cell culture plate in complete DMEM . About 24 h later , cells were transfected with 2 . 5 µg per well pTMNS3-3′JFH1 wild type or analogous constructs containing mutations specified in the results section or empty vector ( pTM1-2 ) [55] . Transfection was performed by using Lipofectamine LTX/Plus Reagent ( Life Technologies ) according to the instructions of the manufacturer . After 4 h , cells were washed once with methionine/cysteine-free medium and starved for 1 h . For radiolabeling cells were incubated for 90 min in 1 ml methionine/cysteine-free medium , supplemented with 2 mM glutamine , 10 mM Hepes ( pH 7 . 5 ) , and 150 µCi/ml of Express Protein labeling mix ( Perkin Elmer , Boston ) . Cells were lysed either directly or washed with PBS and incubated in complete DMEM for 1 or 2 h . Cell lysates were prepared by using NPB ( 50 mM Tris-Cl [pH 7 . 5] , 150 mM NaCl , 1% Nonidet P-40 , 1% sodium deoxycholate , 0 . 1% SDS , 1/10 , 000 vol aprotinin ( 1 U/ml ) , 1/1 , 000 vol leupeptin ( 4 mg/ml ) and 1/100 vol phenyl-methyl-sulfonyl-fluoride ( 100 mM ) ) and cleared by centrifugation at 13 , 000 g for 15 min at 4°C . Cleared lysates were used for immunoprecipitation using either the NS5A-specific monoclonal antibody 9E10 [53] or a polyclonal NS5B-specific antibody . Immunocomplexes were dissolved in 70 µl 2× sample buffer ( 400 mM Tris pH 8 . 8 , 10 mM EDTA , 0 . 2% bromophenolblue , 20% sucrose , 3% SDS and 2% ß-mercaptoethanol ) , separated in a 10% polyacrylamide-SDS gel and analyzed by autoradiography . HCV-specific bands were quantified by phosphoimaging using the QuantityOne software ( BioRad , Munich , Germany ) . Huh7-Lunet cells of a confluent 6-well cell culture plate were washed once with ice-cold PBS and lysed with NP40 buffer buffer ( 50 mM Tris-HCl pH 8 , 150 mM NaCl , 1% NP40 ) . After 30 min incubation at 4°C , cell debris was pelleted by centrifugation for 30 min at 10 . 000 g and at 4°C . Four µg of cleared supernatant was diluted in 2× sample buffer , heated 5 min at 95°C and loaded onto a 12 . 5% polyacrylamide-SDS gel . After electrophoresis proteins were transferred to a PVDF membrane ( PerkinElmer Life Sciences ) for 1 h with an electric current of 1 mA/cm2 . Membrane was blocked in PBS supplemented with 0 . 5% Tween ( PBS-T ) and 5% dried milk ( PBS-TM ) for at least 1 h prior to 1 h incubation with primary antibody diluted 1∶1 , 000 in PBS-TM . Membrane was washed 3 times with PBS-T and incubated for 1 h with horseradish-peroxidase conjugated anti rabbit secondary antibody diluted 1∶10 , 000 in PBS-TM . Bound antibodies were detected after 3 times washing with the ECL Plus Western Blotting Detection System ( GE Healthcare Europe , Freiburg , Germany ) . HCV core protein in transfected cells or cell culture supernatants was quantified using the Ortho® trak-C™ ELISA kit ( Ortho Clinical Diagnostics , Neckargemünd , Germany ) . Lysates of transfected Huh7-Lunet cells were prepared by addition of 1 ml per 6 cm diameter culture dish of PBS containing 1% Triton X-100 , 1/10 , 000 vol aprotinin ( 1 U/ml ) , 1/1 , 000 vol leupeptin ( 4 mg/ml ) and 1/100 vol phenyl-methyl-sulfonyl-fluoride ( 100 mM ) . Lysates were cleared by centrifugation ( 18 , 000×g; 5 min ) and samples were diluted 1∶10 or higher and processed for ELISA according to the manufacturer's protocol . Culture supernatants were filtered through 0 . 45 µm pore-size filters and used directly for core ELISA . Colorimetric measurements were performed using a Sunrise colorimeter ( Tecan Trading AG , Switzerland ) . Kinetic of replication was determined by normalizing the intracellular core amount of the different time points to the respective 4 h value . To determine the efficiency of core protein release , the percentage of extracellular core to total core protein ( the sum of intra- and extracellular core protein ) was calculated . Virus titers were determined as described elsewhere with slight modifications [52] . Huh7 . 5 target cells were seeded at a concentration of 1 . 1×104 cells per well of a 96-well plate in a total volume of 200 µl complete DMEM . Twenty four hours later , serial dilutions of virus containing supernatant were added with 6 wells per dilution . Three days later , cells were washed with PBS and fixed for 20 min with ice-cold methanol at -20°C . After three washes with PBS NS5A was detected with a 1∶2 , 000 dilution of antibody 9E10 in PBS for 1 h at room temperature . Alternatively NS3 was detected with a 1∶100 dilution of antibody 2E3 ( kindly provided by H . Tang , Florida State University , USA ) in PBS for 1 h at RT . Cells were washed again three times with PBS and bound primary antibodies were detected by incubation with peroxidase – conjugated or Alexa Fluor 546 – conjugated anti mouse antibody ( Sigma – Aldrich ) , respectively , diluted 1∶1 , 000 in PBS . After 1 h incubation at room temperature cells were washed three times with PBS and in case of peroxidase – conjugated antibody the Vector NovaRED substrate kit ( Linaris Biologische Produkte GmbH , Wertheim , Germany ) was used for detection . Virus titres ( 50% tissue culture infective dose per ml ( TCID50/ml ) ) were calculated based on the method of Spearman and Kärber [56] , [57] . | Owing to limited genetic information , viruses have to exploit host cells to achieve efficient production of virus progeny . Host cell factors and pathways therefore play an important role for virus replication and thus represent a possible target for antiviral therapy . In case of the hepatitis C virus ( HCV ) , an RNA virus infecting liver cells and causing chronic liver disease , host cell cyclophilins were shown to play an important role in replication . Pharmacological inhibition of cyclophilins , which are catalysts of protein folding , causes profound inhibition of HCV replication , but neither the underlying mechanism by which cyclophilins contribute to viral replication , nor the exact nature of the cyclophilin are known . In this study we demonstrate that HCV replication and presumably also virus particle assembly requires cyclophilin A ( CypA ) , which can be blocked by the cyclosporine analogue DEBIO-025 . We identify mutations affecting proteolytic cleavage of the viral polyprotein that render HCV replication less dependent on CypA and thus cause DEBIO-025 resistance . Studies with additional mutants reveal a correlation between polyprotein cleavage kinetics and CypA dependence . Our results support a model by which CypA activates the viral replicase in a manner that depends on the kinetics with which the viral polyprotein is cleaved . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"virology/virulence",
"factors",
"and",
"mechanisms",
"infectious",
"diseases",
"virology/viral",
"replication",
"and",
"gene",
"regulation",
"gastroenterology",
"and",
"hepatology/hepatology",
"biochemistry/protein",
"folding",
"cell",
"biology",
"infectious",
"diseases/gastrointestinal",
"infections",
"virology/new",
"therapies,",
"including",
"antivirals",
"and",
"immunotherapy",
"virology",
"virology/antivirals,",
"including",
"modes",
"of",
"action",
"and",
"resistance",
"infectious",
"diseases/viral",
"infections",
"virology/mechanisms",
"of",
"resistance",
"and",
"susceptibility,",
"including",
"host",
"genetics",
"biochemistry/drug",
"discovery",
"infectious",
"diseases/antimicrobials",
"and",
"drug",
"resistance"
]
| 2009 | Essential Role of Cyclophilin A for Hepatitis C Virus Replication and Virus Production and Possible Link to Polyprotein Cleavage Kinetics |
As organisms adaptively evolve to a new environment , selection results in the improvement of certain traits , bringing about an increase in fitness . Trade-offs may result from this process if function in other traits is reduced in alternative environments either by the adaptive mutations themselves or by the accumulation of neutral mutations elsewhere in the genome . Though the cost of adaptation has long been a fundamental premise in evolutionary biology , the existence of and molecular basis for trade-offs in alternative environments are not well-established . Here , we show that yeast evolved under aerobic glucose limitation show surprisingly few trade-offs when cultured in other carbon-limited environments , under either aerobic or anaerobic conditions . However , while adaptive clones consistently outperform their common ancestor under carbon limiting conditions , in some cases they perform less well than their ancestor in aerobic , carbon-rich environments , indicating that trade-offs can appear when resources are non-limiting . To more deeply understand how adaptation to one condition affects performance in others , we determined steady-state transcript abundance of adaptive clones grown under diverse conditions and performed whole-genome sequencing to identify mutations that distinguish them from one another and from their common ancestor . We identified mutations in genes involved in glucose sensing , signaling , and transport , which , when considered in the context of the expression data , help explain their adaptation to carbon poor environments . However , different sets of mutations in each independently evolved clone indicate that multiple mutational paths lead to the adaptive phenotype . We conclude that yeasts that evolve high fitness under one resource-limiting condition also become more fit under other resource-limiting conditions , but may pay a fitness cost when those same resources are abundant .
R . A . Fisher's fundamental theorem of natural selection relates the rate of adaptation by populations of organisms to their genetic variance in fitness at a given time [1] . Understanding the mechanistic basis for this variance , and the distribution of a population's fitness variance under alternative modes of selection , have been goals of evolutionary biologists since the modern synthesis . Experimental laboratory evolution using metazoans such as Drosophila [2] , and microorganisms such as bacteria [3] , algae [4] , or yeast [5] has provided the most direct route to these goals , providing deep insight into the forces that guide the adaptive process under different modes of selection [6] . In foundational work , Paquin and Adams monitored the evolution of laboratory strains of the budding yeast Saccharomyces cerevisiae [7] during growth under aerobic glucose limitation in continuous culture [8] , [9] . By monitoring population genetic dynamics over the course of these experiments and characterizing the fitness phenotypes of individual evolved clones , they arrived at two key insights concerning the mechanism of adaptive evolution in clonal populations . First , adaptive shifts , inferred from scoring fluctuations in the frequency of neutral markers , occurred more often in evolving diploids than in otherwise isogenic haploids [8] . Second , in a subset of evolutions , relative fitness of successive adaptive clones was non-transitive , that is , although any particular clone was more fit than its immediate predecessor it was not necessarily fitter than the ancestral strain used to found the population [9] . As for the specific mechanisms underlying changes in fitness , common phenotypes among adaptive clones included increased glucose transport capacity and characteristic cell morphology changes that increased surface area to volume ratios , as might be expected for cells adapted to better scavenging low concentrations of limiting growth substrate [10] . The clones derived from Paquin and Adams' original experimental evolutions have shown an enduring usefulness over the past 25 years for addressing fundamental questions concerning the nature of adaptive evolution . 15 years after the original experiments , Brown et al . , discovered that at least one genetic mechanism underlying enhanced glucose transport was tandem duplication of adjacent genes encoding the high-affinity glucose transporters Hxt6 and Hxt7 [11]; this genomic rearrangement has subsequently been observed in other independent glucose-limited evolution experiments [12] . Ferea et al . [13] probed more deeply into physiological changes that result from prolonged glucose-limited selection using one of these strains and two others evolved under identical conditions from the same ancestor . In the first experiment to use gene-expression microarrays in evolutionary biology [3] , they showed an overall shift in these clones from fermentation to respiration , in what they termed an “enhanced classical Pasteur effect” that allowed for more efficient metabolism of the available glucose [13]; subsequently , Dunham et al . used these same clones to discover genomic rearrangements that occur during adaptation [14] . One fundamental question that these strains have not yet been used to address is: “Does evolution of increased fitness under one type of selection cause decreased fitness under another ? ” - in other words , are there fitness trade-offs ? That trade-offs occur and constrain organismal evolution is foundational to much evolutionary theory , theory that extends into ecology where it has guided analyses of how communities are structured in relation to resource availability [15]–[17] and which factors constrain life history evolution [2] , [18] . The question of how niche breadth evolves has been addressed both theoretically and experimentally [19]–[22] . It is widely held that adaptation to a homogenous environment should favor a narrowing of niche-breadth , whereas adaptation to a heterogeneous environment should favor evolution of a broad niche and maintenance of population genetic variation [19] , [23] . One theory for why niche breadth might be narrowed in an environment where selection is uniform and constant is based on the possibilities that either adaptive mutations or neutral mutations that accumulate under one selection pressure are deleterious under others – possibilities known , respectively , as antagonistic pleiotropy or mutation accumulation . However , these trade-offs can be hard to demonstrate directly and mechanistically [24]–[26] , in part because they must be tested in relation to the ancestral state , which may not always be known or accessible . Correlated responses of fitness to selection are conventionally measured in terms of how well an organism performs in an environment different from the one in which it evolved [24] , [27] , [28] . However , the observation of correlated responses does not by itself prove the existence of trade-offs . What is required are experimental data showing that in alternate environment ( s ) fitness is reduced relative to the ancestor [29] . Experimental microbial evolution has shown that trade-offs do occur , but not inevitably , following selection . Clones from populations of E . coli serially diluted for 20 , 000 generations in minimal medium containing glucose as the sole carbon source exhibit reduced fitness on a variety of alternative carbon sources [30]–[32] . Narrowed niche-breadth does not appear to be specific to evolution on a particular nutrient , as populations of E . coli experimentally adapted to low temperature can show trade-offs at high temperature [33]–[35] . Experimental evolution using the facultatively photosynthetic algae Chlamydomonas has also revealed trade-offs: strains evolved in the presence of light often grow more poorly than the original ancestor in the dark , and vice versa [27] , [29] . In a final example , among E . coli populations evolved in a continuous , mixed-sugar chemostat environment ( lactulose and methyl-galactoside ) , clones evolve most often through either amplification of the lac operon or mutations in the mgl operon . In only one out of thirteen chemostats did a clone evolve having mutations in both of these operons [36] . Taken together these experimental studies suggest that fitness trade-offs ( which could be due to either antagonistic pleiotropy [AP] or mutation accumulation [MA] ) , while not inevitable , can play an important role in determining an organism's niche breadth . In addition to these studies , other recent work has delved deeply into the molecular genetic basis for adaptation using gene-expression analysis , targeted gene sequencing , array comparative genomic hybridization [aCGH] , and/or whole genome sequencing [12] , [37]–[41] . Notwithstanding such breakthroughs , few attempts have been made to comprehensively integrate whole-genome sequence data with estimates of physiological performance and fitness; this activity is essential to achieving the goals of understanding the mechanistic basis for population genetic variance , as well as for the distribution of fitness variance under alternative modes of selection . Here we present just such an integrated set of fitness , physiological , and whole-genome sequence data that we use to test whether evolutionary adaptation to one type of carbon limitation diminishes organismal performance under other types of carbon limitation ( or non-limitation ) . Specifically , we asked whether the well-studied Paquin and Adams [7]–[9] and Ferea et al . [13] yeasts that were evolutionarily adapted to aerobic glucose limitation fared better , no differently , or worse than their common ancestor when cultured in two other carbon-limited environments: anaerobic glucose limitation in chemostats or aerobic acetate limitation in chemostats . Additionally , we assayed these strains' fitness under non-limiting glucose in serial batch culture , and in glucose-rich , nitrogen-limited chemostats . Remarkably , we discovered that evolved strains were consistently more fit than their common ancestor under every condition where carbon was limiting , but that this advantage disappeared when carbon was abundant , indicating the existence of a trade-off . To understand how this might be so , we measured for each strain in each environment indicators of physiological performance including yield and global gene-expression profiles . Then , to discover the genetic mechanisms that underlie these phenotypes and to further unravel the evolutionary history of these well-studied clones , we sequenced the genomes of all five adaptive clones and their common ancestor .
Paquin & Adams [8] , [9] and Ferea et al . [13] isolated end-clones from independent evolution experiments originating from a diploid strain of S288c ( CP1AB ) that was grown under continuous aerobic glucose limitation [7] . To determine whether five of these clones from independent lineages ( hereafter referred to as E1 through E5 , see Materials and Methods ) maintained their fitness advantage relative to the ancestor in “novel” carbon-source environments , selection coefficients were calculated by competing each clone and their ancestor against a common reference strain ( see Materials and Methods ) in three environments: aerobic glucose limitation ( the “direct” fitness response , i . e . , to the original selection ) , anaerobic glucose limitation and aerobic acetate limitation ( the latter two measure “correlated” responses ) . The two alternative carbon-limiting environments were chosen to test their effects on the “enhanced classical Pasteur effect” observed in these clones by Ferea et al . [13] . Specifically , they provide complementary environments to test the fitness consequences of adaptively switching from respiro-fermentative metabolism to respiration alone . In one case , only fermentation is possible ( anaerobic glucose limitation ) , while in the other , only respiration is possible ( aerobic acetate limitation ) . Competitions were carried out for approximately 20 generations , which was short enough to ensure no further adaptive genetic changes would appreciably affect the outcome . As expected , each evolved clone had a significantly higher relative fitness than the ancestor in the aerobic glucose-limited environment in which all of the original evolutions were performed ( Figure 1 ( “Aerobic” ) and Table S1 ) . In the alternative carbon-limited environments , i . e . , “Anaerobic” and “Acetate” , each adaptive clone also exhibited significantly higher fitness relative to the ancestral diploid CP1AB , ( in a 2-tailed t-test ) ( Figure 1 and Table S1 ) . These data indicate that the adaptation to aerobic glucose limitation in each of the clones is not accompanied by a reduction in fitness compared to the ancestral state in either of two alternative environments . They also suggest that selection has improved these clones' ability to scavenge the limiting nutrient and has also enhanced respiratory efficiency . In the above analyses , we compared evolved clones' direct and correlated responses to selection within each environment . To compare the responses to selection between environments , we calculated grand means of all five clones' relative fitnesses within each environment and tested these means for significance differences between each of the three environments ( 2-tailed t-test ) . We hypothesized that some of the adaptive mutations due to the original aerobic regime would be deleterious or neutral under anaerobic growth , specifically those that resulted in the “enhanced classical Pasteur effect . ” This hypothesis predicts that these same mutations would produce a fitness advantage in aerobic acetate-limited growth ( provided they continue to enhance respirative growth ) . The data in Table S2 show significant differences in overall mean fitness between all three environments . As predicted by our hypothesis , mean relative fitness in the aerobic glucose-limited environment is higher than in anaerobic glucose limitation ( competition coefficients 0 . 280 vs . 0 . 143 respectively , p<0 . 05 in two-tailed t-test ) ; strikingly , the relative fitness under acetate-limitation is higher still than under aerobic glucose limitation ( competition coefficients 0 . 437 vs . 0 . 280 respectively , p<0 . 05 in two-tailed t-test ) . These data are consistent with the notion that while the mutations conferring the adaptive advantages for E1 through E5 have an overall net positive effect on fitness when compared to the original ancestor in each environment , some of these mutations might be deleterious or neutral in the anaerobic environment . More specifically , these results indicate that increased glucose transport is still advantageous under anaerobic glucose limitation , but that enhanced respiration provides no benefit in the absence of oxygen . However , enhanced respiration provides obvious benefits under acetate limitation , where oxygen is available . While these trends are important , they are not universally true for all clones . For example , one clone ( E4 , Figure 1 ) has the highest relative fitness in the anaerobic environment , yet it has one of the weakest fitness advantages compared to the other four clones in aerobic glucose limitation and an intermediate fitness under acetate limitation . There are also two cases ( E1 and E5 ) in which the correlated responses in the acetate-limited environment are less than the direct fitness response under aerobic glucose limitation . Taken together , these two observations are consistent with a hypothesis that the more adapted to one environment a particular clone might be , the higher the chance that there will be a fitness cost in a different environment , and vice-versa . This variation in fitness also suggests that multiple genetic paths that have answered the selection are represented among these independent clones . To determine whether unique physiological traits are associated with the fitnesses we observed , three independent , single colonies of each evolved clone and the common ancestor were grown to steady state in the chemostat and three different parameters of culture growth—culture density ( optical density [OD] at 600 nm ) , cell number ( cells mL−1 ) , and biomass ( g 100 mL−1 ) —were measured ( Table S3 , Figure 2 ) . The data in Figure 2 are represented as fold-change relative to the ancestor; please see Table S3 for raw values and statistics . For the aerobic glucose-limitation growth condition , we observed , for all five evolved clones , the same physiological changes reported in previous work [10] , [11] , [13] , namely a two to four-fold increase with respect to the ancestor in optical density , cell number , and biomass ( Table S3 , Figure 2A ) . The observed increases in all measured cell growth parameters among the evolved clones have been postulated to arise from both an increase in glucose transport and from an adaptive switch to increased rates of respiration , resulting in a more energetically efficient use of the available glucose [13] . In alternative carbon-limited environments , evolved clones also demonstrate increased cell yield , relative to their common ancestor , although these differences are much more pronounced under aerobic acetate limitation than under anaerobic glucose limitation . In fact , the differences in magnitude we observe under acetate limitation are comparable to those we observed under aerobic glucose limitation , with values ranging from an almost 5-fold increase in one case ( cells mL−1 for E2 ) , down to a roughly 2-fold increase in biomass for E5 ( Figure 2C , Table S3 ) . In the case of anaerobic glucose limitation , however , the increases are more modest , ranging from a maximum 2 . 5-fold increase ( cells mL−1 for E1 over CP1AB ) to no change or perhaps even a slight reduction ( cells mL−1 for E5 and OD600 for E1 ) for these traits ( Figure 2B , Table S3 ) . Another general observation is that under both acetate limitation and aerobic glucose limitation , there seems to be general concordance between the three physiological traits measured , i . e . , changes relative to the ancestor in biomass , cell number , and culture density change in the same direction and with similar magnitudes when considering any individual clone . This is to be contrasted with anaerobic glucose limitation , where there is no concordance between any of the three physiological parameters; in fact , the data are consistent with steady state OD and cells mL−1 being anti-correlated . Finally , we observe a clear relationship between relative fitness increases and the magnitude of the growth parameter increases relative to the ancestor for these three physiological parameters , both in the aerobic glucose-limited environment ( Figure 2A ) and to a lesser extent in the acetate-limited environment ( Figure 2C ) . However , none of the three parameters seem to be correlated to relative fitness in the anaerobic environment , suggesting that mechanisms independent of enhanced respiration are contributing to higher relative fitness under anaerobiosis in these clones compared to the ancestor ( Figure 2B ) . The “enhanced classical Pasteur effect” described by Ferea et al . [13] was inferred from their gene-expression microarray data . We , too , have used microarrays to determine how the transcriptome responds to alternative carbon-limited environments when cell populations are at steady state . RNA was isolated from the same cultures that were used to estimate physiological parameters ( using two of the three biological replicates ) , and transcript abundance was measured on Agilent yeast catalog arrays , relative to a pooled reference that contained equimolar amounts of each sample . The values ( Log2 ( sample/reference ) ) for the biological replicates were averaged for the subsequent analyses . Because Ferea et al . [13] performed gene expression microarray analysis with only three of these evolved clones ( E1 through E3 ) we wished to determine if the “enhanced classical Pasteur effect” also occurred in the other two clones ( E4 and E5 ) under aerobic glucose limitation , as well as whether the effect manifested in the two alternative environments . We therefore examined our microarray data alongside the 88 “enhanced classical Pasteur effect” genes shown in Figure 1 of Ferea et al . ( Figure 3 , Dataset S1 ) . Under aerobic glucose limitation our data ( Figure 3 ) largely recapitulate the “enhanced classical Pasteur effect” described by Ferea et al . [13] , as well as evolved transcriptional changes reported under aerobic glucose limitation by Jansen , et al . [42] . In all adaptive clones the expression levels of genes involved in glucose oxidation increased while levels of glycolytic genes decreased , relative to the common ancestor . Note that the previously uncharacterized E4 and E5 clones appear to share many of the changes observed in E1 , E2 , and E3 in the aerobic glucose-limited environment , although not exclusively; in fact , E5 appears to be the most divergent of the five evolved clones . Some deviations from the original experiments are seen , particularly in a number of the glycolytic genes that do not appear as highly repressed in our experiments as in the original work ( ADH1 , ADH2 , ENO1 , ENO2 , PGK1 , PDC1 ) . Under anaerobic glucose-limiting conditions , glucose-oxidation pathways are repressed ( or simply not induced ) as would be expected in the absence of oxygen . Contrary to our initial expectations , these same pathways are not highly expressed relative to the ancestor under aerobic acetate limitation . Thus , at least for this set of genes , it appears that evolved clones do not have mutations in one or more global regulators of pathways that result in constitutive up-regulation . Interestingly , under aerobic and anaerobic glucose limitation ( but not acetate limitation ) we observed up-regulation of the hexokinase gene HXK2 . HXK2 encodes a bifunctional enzyme whose cytosolic form irreversibly commits glucose to metabolism by phosphorylating it [43]; also , the nuclear form of Hxk2 is required for Mig1-dependent glucose repression of multiple genes , including HXK1 and itself [44] . Although this enzyme is thought to be a key element of the high-glucose sensing pathway [45] , in chemostats fed with 0 . 08% glucose HXK2 expression was increased 4- to 16-fold in all five evolved clones compared to the parent , a result we provisionally attribute to a 2-fold decrease in expression of the Mig1-Hxk2 regulator SNF1 observed under this condition . To uncover general expression patterns underlying the direct and correlated responses to selection in these strains , we performed a two-class , unpaired Significance Analysis of Microarrays [SAM] [46] , comparing all of the data for the evolved strains in all three environments to all of the data for the ancestor . This procedure identified 160 genes whose expression values significantly differed between the evolved and ancestral strains ( FDR<5% ) ( Figure 4 , Dataset S2 ) . Because the Ras and TOR pathways provide obvious candidates for a general adaptive response that could lead to improved growth of evolved clones relative to their ancestor in all three environments [12] , [47]–[50] , we also considered how these 160 genes behaved in three publicly-available datasets: one that assayed gene-expression in response to induction of RAS1 [51]; and two that measured gene-expression when cells were treated with rapamycin , a drug that inhibits the TOR pathway [52] , [53] . What is visually striking about this list of genes is the degree of correlation ( or anti-correlation in the case of TOR ) with the up-regulation of RAS1 , and the down regulation of the TOR pathway . GO::TermFinder [54] analysis supports this visual observation , as the up-regulated genes are enriched for functions including ribosome biogenesis ( GOID 42254 , Bonferroni corrected p-value = 4 . 67e-5 ) , while down-regulated genes are enriched for response to oxidative stress ( GOID 6979 , Bonferroni corrected p-value = 3 . 3e-6 ) and [small molecule/vacuolar/protein] catabolic process ( GOID 9056 , Bonferroni corrected p-value = 1 . 71e-06 ) among others . These three functions all have regulatory ties to both TOR and Ras/cAMP signaling [55]–[57] and support the hypothesis that mutations that modulate signaling through the Ras/cAMP and/or TOR pathways are likely to provide a mechanism for the evolution of a broad niche that encompasses multiple carbon-limited environments . The genomes of CP1AB and the five evolved clones E1–E5 have been previously interrogated in a number of different ways including southern blot [11] , gene-expression microarrays [13] , array comparative genomic hybridization [14] , and most recently by whole-genome tiling arrays [39] . While some of the genomic events that have occurred as a result of adaptation are known – notably , HXT6/7 amplifications [11] , [14] , rearrangements near the CIT1 locus [14] , and mutations in the AEP3 gene [39] – a comprehensive resequencing of these strains has not been performed . We therefore performed high-throughput whole-genome sequencing ( Table S4 ) and then determined single-nucleotide polymorphisms [SNPs] , small insertions and deletions [indels] , and larger-scale genomic copy number variations [CNVs] in each evolved strain relative to the common ancestor ( see Materials and Methods ) . We validated each substitution or indel in both the evolved and the ancestral strains by Sanger sequencing of the locus , using the same genomic DNA sample that was used for whole-genome sequencing , isolated from a single colony of the indicated strain ( for primers , see Table S5 ) . Table 1 , Table 2 , Table 3 , Table 4 , and Table 5 show the results of this analysis , which identified 28 single-nucleotide polymorphisms relative to the ancestor in E1 ( evolved for 460 generations ) , 17 in E2 ( 250 generations ) , 11 in E3 ( 250 generations ) , 9 in E4 ( 301 generations ) , and 10 in E5 ( 264 generations ) , as well as two short indels , one each in E1 and E2 . One general observation is that the strain that underwent the most number of generations of selection ( E1 ) contained the most polymorphisms relative to the other evolved strains . Another general observation is that these strains have accumulated polymorphisms at a faster rate than haploid populations evolved under almost identical conditions; a haploid adaptive clone isolated from one of the populations from Kao & Sherlock [12] after 440 generations had only accumulated 5 SNPs , 1 transposon insertion , and the HXT6/7 amplification [58] . This observation supports Paquin and Adams' original conclusion that diploids accumulate adaptive mutations more rapidly than haploids [8] . Based on the fact that these independent diploid yeast colonies that were sequenced were single adaptive clones that each represent one lineage throughout the entirety of the evolution , we used simple coalescent theory to estimate the number of mutations we would expect by chance . Theory predicts that the number of neutral mutations we would expect to see in any given clone is simply μ * L * t where μ = mutation rate ( per base per generation ) , L = genome size ( bases ) , and t = time ( generations ) . Using even the most generous estimate of mutation rate ( 6 . 44e-10 per bp per generation estimated by Lang & Murray [59] at the CAN1 locus ) – we would only expect to see a small number of mutations per strain . For E1 , we only expect 7–8 neutral mutations , and for E2–E5 we only expect 3–5 mutations . Under the assumption that this is a Poisson process , seeing the observed number of mutations is significant for each clone ( p<0 . 01 for E1 , E2 , E3 , and E4 , and p<0 . 05 for E4 , Table S6 ) . These data support the hypothesis that a significant fraction of the mutations that we identified are adaptive . Finally , the vast majority ( 63 out of 69 ) of these polymorphisms are heterozygous , as might be expected in evolving diploid populations . To gain further insights about the nature of these mutations as a group , we characterized them with respect to whether they lie in coding regions and if so , whether missense or nonsense amino-acid substitutions are created . Again under the assumption that the mutational events are distributed across the genome in a Poisson fashion , we can determine whether both the distribution of mutations in coding regions , and the frequency with which mutations within a coding region result in an amino acid change , deviate from our expectations . The probability of a mutation occurring in a coding region ( including stop codons ) is ∼0 . 721 and , using clone E1 as an example , the expected number of coding mutations out of 28 observed mutations is between 20 and 21 . Given these estimates , our null hypothesis under a Poisson distribution is that we will not observe greater than 20–21 coding mutations out of 28 total mutations . Because 21 of these 28 mutations actually occur in coding regions , we cannot reject the null hypothesis , and thus we do not see more mutations in coding regions than we would expect by chance ( p = . 085 ) ( Table S7 ) . Similarly for clone E1 , we know that 18 of our 21 coding mutations result in an amino acid change . Using the known probability of a coding sequence mutation effecting an amino acid change ( see Materials and Methods ) , our expectation is that ∼0 . 787 ( or 16–17 out of 21 ) coding mutations will be non-synonymous . Under a Poisson distribution , we again cannot reject the null hypothesis that the number of non-synonymous mutations observed in E1 ( 18/21 ) is greater than the expected number ( 16–17/21 ) with p = 0 . 088 ( Table S7 ) . These data combined with our previous observation suggest that despite a large fraction of these mutations probably being adaptive , the gene-dense nature of the genome ( ∼72% coding ) and a large probability that a mutation occurring in a coding region will result in an amino acid change ( ∼79% ) does not allow us to predict that any given mutation being non-synonymous means it will necessarily be adaptive . In light of these data , but with the caveat that segregating out dozens of mutations and individually testing their fitness effects is a large undertaking , we can still draw some interesting conclusions about the biological implications of these mutations from the actual genes that are affected . We observed that the gene AEP3 , which encodes a mitochondrial integral membrane protein that stabilizes mRNA of the ATP synthase complex [60] , is affected by polymorphisms in three of the evolved strains – E1 , E2 , and E3 . This strongly argues for these mutations being adaptive , particularly because each of the mutations creates a different amino acid substitution . These mutations had previously been observed in E2 and E3 [39] , and each created a novel growth phenotype on acetate at 37°C in haploid progeny of E2 and E3 [14] . These mutations , now also confirmed in E1 , presumably confer an adaptive phenotype during growth in limiting glucose and in limiting acetate despite being heterozygous . A number of other genes that have been mutated are clearly connected to adaptation in the original evolution condition , notably those involved in glucose transport and its regulation ( MIG2 , RGT1 ) , as well as glucose and nutrient signaling ( IRA2 , CYR1 , AVO1 , TOR1 , ARP7 ) . In particular , we discovered mutations in the Ras/cAMP signaling pathway that have been previously identified in glucose-limited evolutions [12] , strongly suggesting adaptive roles for IRA2 , one of the Ras-GTPase-activating proteins [61] , and CYR1 , which encodes the yeast adenylate cyclase [62] . Signaling through the TOR pathway has also been implicated in physiological adaptation to limiting glucose [63] , again suggesting adaptive phenotypes for mutations in TOR1 and AVO1 , a member of the TORC2 complex [64] . These mutations in the Ras/cAMP and TOR pathways have clear implications for the changes that we observed in the transcriptomes of these evolved clones . In E1 we also observe a mutation in MUK1 , which has no known function , but which has also been the target of selection in independently-evolved , haploid , glucose-limited populations of yeast; the mutation appears not to be adaptive on its own ( [12] , [58] ) suggesting the presence of epistasis between mutations in these strains . Taken together , our data reveal likely genetic bases for adaptation to glucose limitation in diploid yeasts , including changes in pathways affecting glucose/nutrient signaling , regulation of glucose transport , and enhancement of aerobic respiration , as well as other intriguing mutations whose roles in adaptation remain to be elucidated . To confirm the DNA copy number changes and other larger scale genome rearrangements discovered by Dunham , et al . [14] , we applied a depth-of-sequence-coverage approach [37] to identify areas of increased or decreased coverage relative to the ancestor , CP1AB ( see Materials and Methods ) . Figure S1 shows the genome mean-centered log2 ratio of coverage ( evolved/ancestor ) in E1 through E5 . Our data recapitulate those of [14] , specifically the ChrIV ( HXT6/7 ) amplification and ChrXIV rearrangement in E1 ( Figure S1 , E1 and E1 , HXT6/7 ) , the ChrVII amplification and ChrXV deletion in E4 ( Figure S1 , E4 ) , and the ChrIV amplification and ChrXIV deletion in E5 ( Figure S1 , E5 ) . While the biological significance of all of these structural variants remains to be elucidated , the specific rearrangements in both E1 and E5 near the important TCA cycle gene CIT1 on ChrXIV , as well as the specific amplification of the HXT6/7 chimera in E1 and of the right arm of ChrIV in E5 ( which includes the HXT6 and HXT7 loci ) have clear implications for adaptation to carbon-limited growth . Glucose at high concentrations is toxic to cells [65] , and glycolytic intermediates can produce reactive carbonyl species that damage DNA and proteins [66] . Not surprisingly , yeast tightly regulates glucose flux into glycolysis by coordinating expression of low- , medium- and high-affinity hexose transport genes in response to changing concentrations of extracellular glucose [67] . Given these observations and our own observations of the increased copy number of HXT6/7 and mutations in RGT1 and MIG2 discovered by genome sequencing , an obvious candidate condition in which to test for the presence of trade-offs is glucose-rich medium , as enhanced glucose transport may no longer be advantageous when glucose is abundant , and may even be costly . To test this hypothesis , we grew evolved and ancestral clones under glucose non-limiting conditions in batch culture and found that nearly all showed diminished maximum specific growth rate ( μmax ) , relative to their common ancestor ( Figure 5 ) . To determine whether evolved strains' diminished growth rate on glucose translated into fitness differences when this resource was abundant we competed the strains and their common ancestor against the same reference strain as before , under two continuous conditions: nitrogen-limited , glucose-sufficient chemostat , and glucose-sufficient serial batch cultures ( Figure 6 and Table S8 ) . Under both conditions , the fitness advantages observed under carbon limitation disappeared . In serial dilution , evolved strains performed no , or only very slightly , better than their ancestor ( while statistically significant , the effect sizes are only ∼1% ) , and in NH4+ limited , carbon-sufficient chemostats , evolved strains were invariably out-competed by their ancestor . Thus , these evolved yeasts are specifically adapted to growth on carbon as a limiting resource , and these adaptations are either of no benefit or actually detrimental when that resource is abundant . Because amplification of high-affinity hexose transporters appears to be in negative epistasis with adaptive mutations in MTH1 [58] , we further tested for the possibility that high HXT6/7 copy number is disadvantageous under glucose-sufficient conditions . We founded 5 independent populations with CP1AB and 10 independent populations with the E1 clone ( which contains this amplification in addition to other putative adaptive mutations ) , experimentally evolved these for >100 generations by serial transfer in 2% YEP dextrose medium and then tested for changes in HXT6/7 copy number by quantitative PCR . We discovered that in at least one instance , copy number decreased ( Figure S2 ) , indicating that this condition can favor reduction or loss of the amplification . Longer-term experiments in rich media will be required to determine whether lower-copy number variants are consistently selected . To determine whether the HXT6/7 amplification alone decreases the growth rate relative to a strain with wild-type HXT6/7 copy number , we characterized spores derived from a diploid strain that was heterozygous for this amplification and carried no other adaptive mutations ( spores courtesy D . Kvitek ) . We observed that the HXT6/7 amplification resulted in decreased growth rate relative to sister spores containing the “wild-type” HXT6/7 locus ( Figure S3 ) . These data support the hypothesis that the HXT6/7 amplification is deleterious during growth in excess glucose , and hence is an example of antagonistic pleiotropy .
Our choice of assay regimes was motivated by a desire to understand the generality , relative magnitude and mechanistic bases of adaptations for acquiring [11] and processing [13] limiting glucose under prolonged selection . All evolved strains showed significant improvement in fitness under the selective regime and assay regimes where carbon was limiting . Microarray analysis of cells grown under the selective regime showed that , relative to their ancestor , evolved yeasts had diminished expression of genes in fermentative metabolism and increased expression of genes in oxidative metabolism . These results essentially recapitulate earlier findings [13] , even though our analyses were performed on a different platform and included strains from the Paquin and Adams experiments that had not been previously investigated . Interestingly , beginning with a different yeast ancestor ( CEN . PK 113-7D ) and using a slower dilution rate ( 0 . 1 h−1 ) , Jansen , et al . [42] also saw diminished fermentative capacity in yeast evolved under prolonged glucose limitation , evidenced at both the transcriptional and enzymatic levels . Thus , adaptive evolution of an “enhanced classical Pasteur effect” under this selective regime appears to be a general result . Remarkably , although the clones we investigated evolved in independent populations under aerobic glucose limitation , they performed better than their common ancestor in other carbon limiting environments under both anaerobic and aerobic conditions . The adaptive clones' superior performance is manifest in cell yield and fitness in both the selective and assay regimes . The relative magnitudes of the physiological values associated with fitness differences can be easily explained in terms of the energetics of aerobic vs . anaerobic catabolism [68] and the phenotypes most likely to bring about a competitive advantage under resource limitation: either or both enhanced capacity to scavenge limiting resource which increases fitness without significant gains in yield , or increased efficiency of limiting resource utilization resulting in higher cell yield and higher fitness [17] , [69] . Specifically , we suggest that the heritable changes we identified that improve glucose uptake capacity in aerobic conditions [10] result in the modest increased yield and fitness under anaerobic glucose limitation; under aerobic acetate limitation , we see heritable changes that improve aerobic capacity , evidenced by more pronounced changes in these parameters . The relative magnitude of yield and fitness differences in these two assay regimes reflects the scope for selection: only modest gains are possible in high-affinity glucose transport , whereas much more substantial gains are possible by shifting to and then improving upon the machinery of oxidative metabolism , whose ATP yield is many-fold greater than fermentation [68] . Thus , cells have greater scope for adaptive change by simply enhancing the classical Pasteur effect . That said , a striking result of our data is that adaptation to glucose limitation has not only resulted , as expected , in increased glucose transport and diminished catabolite repression , but also to more efficient machinery for carrying out oxidation , even of the non-fermentable , non-repressing substrate , acetate . Our phenotypic data indicate that five independently evolved clones have converged on growth phenotypes that give them a competitive advantage in selective ( and carbon scarce ) assay regimes alike . We would not necessarily have predicted this from previous studies [11] , [13] , [14] , as they provided few clues as to the possible costs of adaptive change . Remarkably , the similar phenotypes we observed arise from different sets of mutations in each clone , although certain genes and pathways seem more likely to be targeted by selection than others . For example , while increased glucose transport is clearly an adaptive phenotype , it appears to have been accomplished by different mutations in two of these clones . Clone E1 contains mutations in two genes that regulate glucose transporter gene-expression , MIG2 and RGT1 , as well as a tandem duplication of the genes encoding the hexose transporters Hxt6 and Hxt7 . Interestingly , the mutations in MIG2 and RGT1 are synonymous , indicating potential functional roles for mutations that do not effect an amino-acid change . By contrast , clone E5 contains an amplification of the entire right arm of chromosome IV , containing the HXT6 and HXT7 loci . Our microarray results , viewed through the lens of our whole genome sequencing data , suggest other adaptive mechanisms additional to the changes in HXT6/HXT7 copy number and CIT1 regulation noted previously [11] , [14] . These gene-expression data in particular lend support to the hypothesis that the glucose/nutrient signaling pathways of these strains are affected in such a way as to promote cell division even in a nutrient-poor environment . Two clones have mutations that likely affect signaling through the Ras/cAMP pathway . E1 is heterozygous for a mutation in IRA2 , a gene that encodes a Ras-GAP that functions to decrease intracellular cAMP levels . Interestingly , E2 contains a mutation in the same pathway in the gene that encodes yeast adenylate cyclase itself , CYR1 . These mutations would be particularly interesting to characterize , as our gene expression data and data from [12] would predict that the cyr1 mutation would be a gain-of-function mutation that increases intracellular cAMP levels , whereas the ira2 mutation should be a loss-of-function mutation resulting in constitutive Ras signaling and similar increases in cAMP levels . Gene-expression data also suggest increased signaling through the TOR pathway across all evolved clones , relative to their common ancestor , an observation that is again supported by the presence of novel mutations in this pathway . We found that E2 has a mutation in AVO1 , a component of the TORC2 complex , while E5 has mutations in TOR1 itself as well as in MDS3 , a putative component of the TOR signaling cascade . Again , our prediction is that these mutations should be gain-of-function that increase signaling through this key regulatory pathway . Finally , our fitness and physiological data point to increased function in oxidative metabolism as an alternate mode to answer the challenge of limiting glucose , while simultaneously creating a fitness advantage when grown in acetate limitation . We uncovered three independent mutations in AEP3 , a gene that encodes a mitochondrial protein important for ATP synthase function . Strikingly , the mutations in E1 , E2 , and E3 all affect the same codon but effect independent amino acid substitutions , and E2 contains 2 additional nucleotide changes that change another amino acid in close proximity to the other mutated codon . We observe additional mutations in E1 that likely affect mitochondrial function , including BNA4 ( involved in biosynthesis of nicotinic acid ) , NGR1 ( over-expression of which impairs mitochondrial function ) , and TOM70 ( which is a translocase of the outer mitochondrial membrane ) . It will be illuminating to follow-up these observations with a characterization of their individual or epistatic fitness contributions . These data also provide a hypothesis for observed variation in correlated responses to selection . The whole genome sequence of clone E3 provides no mechanistic basis for enhanced glucose transport , but does have a mutation in the integral mitochondrial protein , AEP3 . Significantly , we observed that relative to the ancestor and to other independently evolved clones , E3 exhibited highest fitness under acetate limitation and less of a selective advantage under glucose limitation . A final observation regarding clones E3 is that it appears to have the most distinct gene expression pattern compared to the other five clones under acetate limitation ( Figure 3 and Figure 4 ) , possibly suggesting roles for altered transcription leading to higher relative fitness under acetate-limited growth . One possible contributor to the observed gene expression differences of E3 under acetate limitation is the mutation in GPA1 , an upstream G-protein that leads to activation of a transcription factor , Ste12p , that plays a role in both pheromone response and regulation of invasive growth . Indeed , many of the genes that show altered transcription in Figure 3 and Figure 4 are known Ste12p targets , but again , more work will be required to determine if the mutation in GPA1 is responsible for the observed gene expression phenotype . Our sequence data provide a rich resource to begin answering other fundamental questions about the nature of yeast's evolutionary adaptation to a limiting resource: What are the fitness and biochemical effects of each new mutation ? Which mutations are adaptive , and which are neutral or mildly deleterious and merely hitchhiking ? How pervasive is epistasis between new mutations ? And , because most novel alleles are heterozygous , which , if any , are over-dominant ? Finally , because we have seen haploids adapt to limiting glucose by similar mechanisms , albeit more slowly , ( see genotypes in Kao & Sherlock [12] ) , we may ask: are mutational differences seen in diploids due to ploidy or due to our sampling not having comprehensively obtained all possible genotypes that can respond to this selection ? This work has also addressed unanswered questions posed by Ferea et al . [13] concerning the genetic basis of the “enhanced classical Pasteur effect . ” While the specific causal mutations of these gene-expression changes remain to be determined , our data lead us to two conclusions . The first is that these changes are not constitutive: mutations that cause increased expression of glucose-oxidation pathways specifically under aerobic glucose limitation can still be repressed in the absence of oxygen , when their expression is inappropriate . The second is that there appear to be multiple adaptive paths to the same phenotype , in opposition to one of the original hypotheses that there are few [13] . Additional experiments will be required to isolate individual mutations and determine how each , alone and in combination with others , impacts differential regulation of glycolysis and the TCA cycle under the selective and assay regimes . Our work brings new evidence to bear on the longstanding question of how trade-offs influence adaptive evolution . Constant , homogenous environments are widely believed to favor evolution of narrow niches in contrast with heterogeneous environments , which are believed to favor evolution of broad niches [19] , [70] , [71] . Corollary to this belief is that narrow niches arise from trade-offs due to antagonistic pleiotropy , and/or differences in the rates at which beneficial and/or deleterious mutations accumulate in these different selection regimes [28] , [72] . Here we find that clones evolved under constant glucose limitation are , as expected , more fit than their common ancestor in the selective regime , but also more fit in two assay regimes: anaerobic glucose limitation and aerobic acetate limitation . In retrospect , given the changes we have discovered in strains' physiology , gene expression and genome sequence , the apparent lack of trade-offs under the assay regimes we chose is perhaps not so surprising . An increased capacity to scavenge glucose should produce a fitness advantage in any environment in which glucose is meager; thus the direct genetic evidence we see for this in at least two strains , E1 and E5 ( HXT6/7 ) , likely outweighs any cost of uselessly increasing glucose-oxidation ability under anaerobic glucose-limitation . Similarly , although increased glucose transport is unlikely to be adaptive under acetate limitation , any cost imposed thereby is likely offset by an increased capacity to oxidize carbon . More generally , we can use a term defined by Bell & Reboud [29] to describe selection in aerobic glucose limitation as synclinal – meaning that the direct and correlated fitness responses to this selective regime were positive with respect to the ancestor in all five evolved clones . These conclusions , however , might only apply to carbon ( or even particular kinds of carbon ) limitation , and in further work it would be appropriate to test these evolved clones under a much more diverse set of environments to determine the breadth of their niche . These types of experiments will be crucial to discerning whether trade-offs exist under other assay regimes and , if so , how mutation accumulation and antagonistic pleiotropy combine to produce them . The Paquin and Adams [7]–[9] and Ferea et al . [13] yeasts evolved under limiting glucose have been used over the last quarter century to address fundamental questions relating to the dynamics and mechanisms of adaptive evolution . Our work continues in that vein , providing evidence to support the conclusion that evolution under one resource limiting condition leads to generalists that are more fit than their ancestor under other resource limiting conditions , but less fit when the original limiting resource is abundant . Additionally , we have sequenced these strains' genomes and provided a list of genetic changes that arose in independent evolution experiments , creating a rich resource of information that can be used to continue studying the mechanisms by which organisms adapt to resource scarcity , as well as the apparent cost to being a “hunger artist” when resources are plentiful . Intriguing as these apparent trade-offs that we have identified may be , more work will be required to understand every mutation's mechanistic role ( biochemical , metabolic , regulatory , etc . ) in adaptation to prolonged resource limitation . We will then be in a position to generate further specific hypotheses as to which conditions should reveal the cost of particular adaptations and whether that cost is incurred as a result of antagonistic pleiotropy , mutation accumulation , or both .
Strains used in this study were Saccharomyces cerevisiae CP1AB with genotype MATa/α , gal2/gal2 , mel/mel , mal/mal [7] and evolved clones E1 , E4 , and E5 [8] and E2 and E3 [13] . The common reference strain used for the competition experiments was DBY11249 ( FY4 , with a d-Tomato/NatMX cassette replacing the dubious ORF YLR255c , strain courtesy David Gresham and Greg Lang ) . Cultures were stored in 15% glycerol at −80°C . Strains for chemostat cell cultures were grown in 1% YEP Dextrose and 1 mL aliquots were frozen in 15% glycerol at −80°C . The entire contents of a single 1 mL frozen aliquot of either the reference strain and evolved or ancestral isolates were used to inoculate an individual chemostat ( ATR SixFors fermentation apparatus , ATR Biotechnologies ) with working volume set to 400 mL of minimal ( SC ) media defined by [73] . Batch cultures were then grown for 24 hours to achieve saturation . After saturation was achieved , chemostat pumps were turned on to the desired dilution rate and 2–3 vessel volumes of turnover were allowed so cultures could reach steady state . 100 mL of each strain growing at steady state ( reference plus evolved or ancestor ) were transferred to a fresh chemostat and the dilution rate was set to 0 . 17 hr−1 for aerobic ( 0 . 08% glucose ) and anaerobic glucose limitation ( 0 . 08% glucose+420 mg mL−1 Tween 80+10 mg L−1 ergosterol ) and 0 . 05 hr−1 for aerobic acetate limitation ( 10 . 9 g/L sodium acetate ) . For aerobic ammonium limited ( 0 . 015% ( NH4 ) 2SO4 ) , glucose-sufficient ( 9 g/L glucose ) growth , chemostat dilution rate was also set to 0 . 17 hr−1 . Aerobic conditions were achieved by sparging with 25 L h−1 of sterile air and anaerobic conditions by sparging cultures with 25 L hr−1 sterile-filtered , humidified N2 ( g ) . 3 mL samples were taken at time = 0 ( immediately following transfer to fresh chemostat ) and every 6–8 hours for 2–3 days ( ∼15 generations ) . Time and volume of effluent were measured at each sample to determine generations . 1 mL of cells were resuspended in Phosphate Buffered Saline , sonicated for 10 s , and analyzed with flow cytometry to determine relative proportions of fluorescent ( reference ) to non-fluorescent ( experimental sample ) strains . 50 , 000 cells were counted to obtain accurate measurements of relative proportions . Regression analysis of generation time vs . ln ( experimental sample/common reference ) was used to calculate per-generation competition coefficients . This method is based on the method worked out by Alex Ward and David Gresham and similar to the method used in [74] . Similar procedures were used to compute selection coefficients of strains competed in serial dilution batch culture . The media employed in these experiments was that of Adams et al . with the addition of 4% dextrose ( wt/vol ) . Approximately equal numbers of the test and fluorescent reference strains were combined at an initial cell density of ∼105 cells mL−1 in 10 mL media . Samples were cultured for 24 h ( ∼6 . 5 generations ) at 30°C on a New Brunswick T-7 roller drum , then diluted to a similar cell density in fresh media and cultured an additional 24 h . Samples for FACS analysis were taken over 3 successive serial dilutions ( approximately 20 generations ) . Pairwise competition experiments were performed in triplicate . To determine specific growth rates in glucose non-limiting batch growth for evolved clones E1–E5 , ancestral clone CP1AB , and haploid segregants with or without the HXT6/7 amplification ( haploid segregants GSY2707-2714 were otherwise isogenic from parent diploid GSY1208 that was heterozygous only for the HXT6/7 amplification ) , multiple independent single colonies of these strains were grown overnight in 2% YEP dextrose and diluted 1∶50 into fresh medium in a 100 µL , 96-well optical plate ( Costar ) , sealed with optical sealing tape ( E&K Scientific ) , and grown for approximately 24 hours in a TECAN plate reader at 30°C . Specific growth rate was defined as the change in ln ( optical density ) per hour during exponential growth . To test the stability of the adaptively evolved HXT7/6 amplification under nutrient-rich conditions , we selected at random five colonies of the parent strain and ten colonies of adapted strain E1 and used these to found fifteen experimental populations . Populations were inoculated at a density of ∼105 cells mL−1 in 10 mL YEPD ( 2% glucose ) , cultured at 30°C in a New Brunswick T-7 roller drum , and serially propagated by diluting cells ∼100-fold on a daily basis in fresh media . Experiments were carried out for 15 days ( >100 generations ) ; 1 mL of each culture was archived every 25 generations as −80°C 15% glycerol stocks . Population samples from the last time-point of each experiment were spread onto YPD agar . Genomic DNA was prepared from three randomly chosen colonies on each plate using the YeaStar Genomic DNA Kit ( Zymo Research ) ; this material was used as template for quantitative PCR assay of HXT7/6 copy number using primers specific for the HXT6/7 locus and control primers on chromosome IV designed against the UBP1 locus ( primers in Table S5 ) , using the ΔΔCt method as described by [58] . To obtain physiological measurements , cultures were grown to steady state in individual chemostats in each of the three environments described above , under identical conditions . Experiments were performed in triplicate . Biomass estimates were determined by rapidly withdrawing 100 mL from fermentation vessels , and fast-filtering this volume through sterile , tared 47 mm , 0 . 45 mm Nylon filters ( Whatman ) . Filters were dried overnight in an 80°C oven and weighed the following day . Cell number was estimated by haemocytometry using an aliquot from 1 mL of sample treated with 10 ug mL−1 cycloheximide . Steady state optical density was measured spectrophotometrically at 600 nm . To isolate total RNA , 100 mL of sample was quickly filtered through 0 . 45 mm Nylon filters ( Whatman ) and flash frozen in liquid nitrogen . RNA for gene-expression measurements was isolated using the hot acid-phenol method described by [13] . To assay relative mRNA abundance , total RNA was isolated as described above . A pooled reference sample was created containing equimolar amounts of each of 36 samples ( 6 strains in 3 environments , using two of the three biological replicates ) . 325 ng of total RNA from samples or reference pool was used as the input for reverse transcription and labeling with Cy dyes ( Amersham ) using the Low RNA-input Linear Amplification Kit ( Agilent ) following manufacturer's instructions except that reaction volumes were halved . 1 . 5 µg each of labeled sample and labeled reference were hybridized to Yeast Gene Expression Arrays v2 8×15k ( Agilent ) for 17 hours at 65°C rotating at 10 rpm in a hybridization oven ( Shel Lab ) . Arrays were then washed according to manufacturer's instructions and scanned at 5 µm resolution on an Agilent Scanner . Data were extracted using Agilent Feature Extraction v9 . 5 . 3 . 1 , which uses linear-Lowess normalization and calculates log2 ratios . Following data extraction from the raw images , we averaged the data for both probes for each gene . Raw gene-expression data have been deposited in GEO with accession number GSE25081 . CP1AB and E1–E5 were streaked for single colonies from 15% glycerol stock solutions ( −80°C ) onto 2% YEP Dextrose plates . Single colonies were grown in 2% YEP Dextrose liquid cultures at 30°C and genomic DNA was extracted by spooling as described [75] . Paired-end libraries were created using the Illumina Genomic DNA Sample Prep Kit according to manufacturers instructions ( 5 µg input genomic DNA ) , and sequencing flow cells were prepared using the Illumina Standard Cluster Generation Kit . Samples were sequenced on the Illumina Genome Analyzer II , and image analysis and data extraction were performed using Illumina RTA 1 . 5 . 35 . 0 . Reads were mapped and variants were called using two different methods , with largely similar results . In the first method , reads with qualities ( FASTQ ) were aligned to the S288c reference genome ( SGD , as of Feb 2 , 2010 ) using BWA v0 . 5 . 7 [76] . Whole-genome pileup files were generated using SAMtools v0 . 1 . 7 [77] and SNPs and Indels were filtered using custom Perl scripts . Briefly , SNPs passed the filter if they were represented in at least 30% of reads in the evolved strain ( allowing for heterozygosity ) and at most 10% in the ancestor , or at least 80% in the evolved strain but less than 80% in the ancestor ( allowing for heterozygous to homozygous mutations ) . Additional heuristic filters included a confirming read from both strands , with at least 5 reads covering the position in both strains , and no more than one ambiguous SNP call ( “N” ) or deletion ( “*” ) at that position . Indels were filtered by requiring at least a 30% or greater allele frequency difference between ancestral and evolved strains , if they shared the same indel call . Additionally , if there were >2 indel calls at a given position , the number of reads supporting the two most common indel calls had to be > = 80% of the total reads covering that position . Raw coverage in evolved and ancestral strains at the given position must also have been at least 10× . In the second method , we mapped reads with qualities using Stampy [78] , and applied the Genome Analysis Toolkit ( GATK ) “Best Practice Variant Detection” [79] by first performing base quality score recalibration , indel realignment , and duplicate removal . We then performed SNP and indel discovery across all evolved and ancestral sequences simultaneously using standard hard filtering parameters [80] , and then used custom perl scripts to identify SNP or indel variant calls that were different between ancestral and evolved strains . Primers used to confirm or reject SNPs and Indels are in Table S5 . For determining copy number variation ( Figure S1 ) , a coverage-based approach was used as outlined by [37] . Briefly , raw sequencing coverage was averaged over 1 Kb intervals across the genome of each evolved clone and the ancestor . Log2 ( evolved/ancestor ) ratios were then calculated and normalized to the genome mean log2 ratio . Genome segments were identified using a circular binary segmentation algorithm implemented in the R software package DNAcopy [81] with parameters as follows: data . type [logratio]; smooth . region [3]; alpha sign . cutoff [ . 01]; min . width [5]; undo . splits [sdundo]; sdundo [4]; nperm [10000] . Raw sequence data have been deposited in the Sequence Read Archive ( SRA ) database with accession number SRA025083 . 1 . To determine the average probability of a mutation in a coding region effecting a non-synonymous coding change we wrote a custom perl script that calculates the average probability that a mutation would change the codon to encode a different amino acid . Briefly , for each codon in the genome , every possible mutation was generated ( 9 changes for each codon ) , and the fraction of those 9 possible mutations that created a non-synonmous codon was recorded . For example , a four-fold degenerate site at the wobble base of a given codon would yield a probability of 2/3 non-synonymous ( 6 out of 9 mutations change the codon ) . We then simply averaged this probability across all codons in the genome . | Microorganisms such as yeast have been used for decades to study adaptive evolution by natural selection . Thirty years ago in now seminal experiments , a strain of yeast was evolved multiple times under carbon limitation . The adaptive changes that gave rise to increases in fitness have previously been studied both phenomenologically and mechanistically but not in detail at the molecular level . To better understand the basis for these strains' fitness increase , we sequenced their genomes and identified putative adaptive mutations . We found that multiple mutational paths lead to these fitness increases . We also determined whether the evolved yeasts' gains in fitness under the original conditions in some cases diminished fitness under other conditions . We therefore evaluated their performance relative to the ancestral strain under the evolutionary and two alternative resource-limiting conditions by determining the ancestral and evolved strains' relative fitnesses and gene-expression levels under all three conditions . We found scant evidence among evolved strains for fitness trade-offs when nutrients were scarce , but discovered a cost was paid when nutrients were plentiful . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"carbohydrate",
"metabolism",
"organismal",
"evolution",
"genome",
"evolution",
"population",
"genetics",
"microbiology",
"mutation",
"model",
"organisms",
"microbial",
"evolution",
"metabolic",
"pathways",
"biology",
"evolutionary",
"theory",
"evolutionary",
"genetics",
"biochemistry",
"adaptation",
"natural",
"selection",
"yeast",
"and",
"fungal",
"models",
"saccharomyces",
"cerevisiae",
"genomics",
"evolutionary",
"biology",
"genomic",
"evolution",
"metabolism",
"evolutionary",
"processes",
"genetics",
"and",
"genomics"
]
| 2011 | Hunger Artists: Yeast Adapted to Carbon Limitation Show Trade-Offs under Carbon Sufficiency |
Pili have been identified on the cell surface of Streptococcus pneumoniae , a major cause of morbidity and mortality worldwide . In contrast to Gram-negative bacteria , little is known about the structure of native pili in Gram-positive species and their role in pathogenicity . Triple immunoelectron microscopy of the elongated structure showed that purified pili contained RrgB as the major compound , followed by clustered RrgA and individual RrgC molecules on the pilus surface . The arrangement of gold particles displayed a uniform distribution of anti-RrgB antibodies along the whole pilus , forming a backbone structure . Antibodies against RrgA were found along the filament as particulate aggregates of 2–3 units , often co-localised with single RrgC subunits . Structural analysis using cryo electron microscopy and data obtained from freeze drying/metal shadowing technique showed that pili are oligomeric appendages formed by at least two protofilaments arranged in a coiled-coil , compact superstructure of various diameters . Using extracellular matrix proteins in an enzyme-linked immunosorbent assay , ancillary RrgA was identified as the major adhesin of the pilus . Combining the structural and functional data , a model emerges where the pilus RrgB backbone serves as a carrier for surface located adhesive clusters of RrgA that facilitates the interaction with the host .
The Gram-positive bacterium Streptococcus pneumoniae , also known as pneumococcus , is one of the most important human pathogens causing respiratory tract infections such as sinusitis , otitis media , and community acquired pneumonia , but also invasive diseases such as septicemia and meningitis . Together with HIV , malaria , and tuberculosis the pneumococcus represents one of the four major infectious disease killers [1]–[4] . Even though pneumococcus is a devastating pathogen , it is also a member of the human commensal flora and is known to asymptomatically colonize the nasopharynx [1] . A major virulence factor of Streptococcus pneumoniae is the polysaccharide capsule , by which pneumococci are grouped into at least ninety different serotypes [5] . Other genetic factors , such as CbpA ( choline-binding protein A ) and pneumolysin , have been described to be of importance for virulence [6]–[8] . Infection by Streptococcus pneumoniae leads to invasive disease triggered by initial colonization of the nasopharynx , but the mechanisms of adhesion are not well understood [9] . Recently , pilus harboring pneumococci were discovered and results obtained indicate a key role for these structures in virulence and disease [10] , [11] . Furthermore , in a mouse model of intraperitoneal infection Gianfaldoni et al . [12] reported protective immune responses after active and passive immunization with recombinant pilus subunits of Streptococcus pneumoniae Type 4 strain TIGR4 ( T4 ) . Previously , similar pili-like surface structures had been identified in other Gram-positive bacteria , such as Corynebacterium diphtheriae [13] , [14] , Actinomyces spp . [15] , group A streptococci ( GAS ) [16] , group B streptococci ( GBS ) [17] and recently Mycobacterium tuberculosis [18] where they were shown to play an important role in the interaction with the host at different stages of infection . The Streptococcus pneumoniae pilus was found to be encoded by the rlrA pathogenicity islet [10] , [19] , initially discovered in T4 , a clinical , serotype 4 strain , of which the genome is known [20] . Sequencing of various pneumococcal strains revealed , that not all isolates contain this genetic element [21] , [22] . The rlrA operon encodes , besides a Rof-A-like transcriptional regulator ( RlrA ) , 3 sortases ( SrtB , SrtC and SrtD ) and 3 structural proteins RrgA ( Swiss-Prot Q97SC3 ) , RrgB ( Swiss-Prot Q97SC2 ) and RrgC ( Swiss-Prot Q97SC1 ) containing a LPxTG motif ( or variants thereof ) [10] , [19] , [23] . In contrast to Gram-negative pili , which are composed of non-covalently linked subunits , Gram-positive pili studied so far are thought to be extended polymers formed by a transpeptidase reaction involving covalent cross-linked subunit proteins containing specific amino acid motifs , which are assembled by specific sortases . Sortases are also responsible for the covalent attachment of the pilus to the peptidoglycan cell wall [24] . Fundamental work on this was done by Schneewind and co-workers studying Corynebacterium spp . pili [13] , [14] , [25] , [26] and recent reviews summarize the more general knowledge on Gram-positive pili [27]–[29] . In Corynebacterium diphtheriae , in addition to a N-terminal signal sequence and a C-terminal cell wall sorting signal , two motifs are considered to be important for the major pilus component , i . e . the so called pilus backbone forming protein: the pilin motif and the so called E-box [13] . Following the corynebacterial system , pneumococcal RrgB was proposed to form the backbone of the pneumococcus T4 pilus structure , as its sequence contains homologues of both motifs . For pneumococcal T4 RrgA and RrgC , a role as ancillary proteins was suggested [10] . These observations are supported by initial electron microscopy ( EM ) analysis on pneumococcal cells containing pili [10] , [11] . Although the precise mechanism of incorporation of RrgA , RrgB and RrgC into the pneumococcal pilus is not yet understood , one hypothesis is that incorporation of the three subunits is specifically catalysed by each of the 3 sortases present in the rlrA islet: in line with this are results found by LeMieux et al . [11] that showed that SrtD is needed for RrgA incorporation into the typical high molecular weight ( HMW ) structure . In addition , the incorporation of RrgA is dependent on the presence of RrgB but not RrgC . Whereas genetically based functional studies regarding Gram-positive pili such as those of Streptococcus pneumoniae are emerging , structural information of the native entire pilus in Gram-positives is lacking and its significance in infectious disease is not clear . Very recent data based on crystal structures of single pilus subunits of Gram-positive pili in Streptococcus agalactiae and Streptococcus pyogenes stimulated novel insights into Gram-positive pilus composition [30] , [31] . The elucidation of the structure of the native pilus is of great interest not only to increase our understanding of the biology of Gram-positive bacteria , but also as potential tool to develop proper therapeutics and vaccines against pathogenic bacteria like Streptococcus pneumoniae [32] , [33] . Our approach consists in using native , purified pneumococcal pili of a pathogenic T4 strain to study structure and properties of these Gram-positive surface appendages . We provide for the first time structural evidence of the pneumococcal pilus , which is composed of protofilaments arranged in a coiled-coil superstructure . Structural proteins RrgA , RrgB and RrgC localized to different regions of the same pilus , confirming RrgB as the major compound , followed by clustered RrgA and single RrgC molecules on the pilus surface . RrgA was identified as major adhesion protein towards selected extracellular matrix ( ECM ) compounds . Structural and functional data indicate the pneumococcal pilus to be a flexible carrier of functional groups able to cross pneumococcal polysaccharide capsule , promoting host cell interaction .
In order to study the pneumococcal pilus in detail , a multi step purification procedure was set up to obtain pure native pili preparations . Pneumococcal pili were isolated from strain T4 , bacteria that in low-dose EM showed individual pili and bundles of individual pili on the bacterial cell surface ( Figure 1 ) . Both types of appendages were distributed on the bacterial surface , the majority of which ( ∼65% ) belonging to the individual pilus type . The same appearance was found analysing purified pili preparations with cryo electron microscopy ( cryo-EM ) . Briefly , for pili purification , bacteria were grown on blood agar plates . Harvested bacteria were washed and subjected to mutanolysin treatment . N-acetyl muramidase treatment released pneumococcal pili into the supernatant . Subsequently , supernatants containing pili were applied to a sucrose gradient to separate them from other cellular impurities and to concentrate their relative amount in the sample preparation ( Figure 2A ) . Pili-positive gradient fractions were identified by dot blot analysis with anti-RrgB antibodies . Western analysis ( anti-RrgB ) of SDS-PAGE separated pili samples was performed to identify HMW material in fraction number 5–8 at the top of the separation gel . Dialysed sucrose pools were concentrated and applied to gel filtration to further purify the pili preparations . As shown in Figure 2B , size exclusion chromatography allowed the separation of HMW pili ( peak A ) from lower molecular weight material ( peak B and C ) as proven by SDS-PAGE and western blot analysis with anti-RrgB antibodies ( data not shown ) . As a control , the same purification procedure was performed with a pneumococcus T4 Δpil strain . As expected , HMW pili ( peak A ) that were eluted in the void volume were not found in the respective delta pilus preparation . A summary of the purification strategy of pneumococcal T4 pili is shown in Figure 2C: Silver stained SDS-PAGE analysis of the different purification steps and the correspondent western blot anti-RrgB analysis show purified HMW pili after gel filtration at the entry of the gel pocket . Purified pili were used for further analysis and to study the structure of pneumococcal pili . HMW pili , following size exclusion chromatography , were applied to SDS-PAGE and HMW band was subjected to mass spectrometry analysis and N-terminal sequencing ( Edman analysis ) . Tryptic peptide sequence of HMW pili were analysed by MALDI-TOF . Results identifying the cell-wall surface anchor family protein RrgB of Streptococcus pneumoniae T4 ( gi|15900379 ) were confirmed by MS/MS analysis: the fragmentation of a peptide with a mass of 2064 Da matched the peptide sequence LAGAEFVIANADNAGQYLAR that is part of pneumococcal T4 RrgB . The Edman analysis resulted in the peptide sequence AGTTTTSVTVHXL , which could be identified as part of T4 protein RrgB . The identified N-terminal starting amino acid corresponds to an alanine , which is located 30 residues downstream of the Met of the RrgB sequence , in agreement with the predicted cleavage site of the signal sequence . The detailed composition of pili was investigated by IEM with antibodies raised against recombinant HisTag-RrgA , -RrgB and -RrgC . Initially , single and double IEM were performed with different combinations of the three antisera on both bacteria and isolated pili in order to reveal the presence of all three pilus components . Triple immunogold staining was then performed on the same pilus preparation and on whole bacteria to observe the type of distribution and the relative amount of the 3 structural proteins . Figure 3A shows RrgB distributed evenly along the entire pilus polymer while RrgA and RrgC were present at non-regular intervals along the pilus shaft . An approximate estimation of the relative amounts of the three proteins based on triple , double and single IEM observations indicated that roughly 90% of the gold particles corresponded to RrgB . The remaining 10% circa were composed of RrgA and RrgC , with a higher occurrence of RrgA in comparison to RrgC . In particular , the IEMs showed that RrgA was organized in small clusters , as found by particulate aggregates of 2–3 anti-RrgA antibody units , distributed along the entire pilus surface . Interestingly , RrgC protein was found in single copies and often co-localized with the RrgA clusters . Triple IEM performed on whole bacteria ( Figure 3B ) confirmed that purified pili conserved the same structural characteristics as native pili attached to bacteria . Purified HMW pili preparations , observed by cryo-EM ( Figure 4 ) and freeze drying/metal shadowing techniques ( Figure 5A , 5B and 5C ) showed that pili were elongated structures of up to 1 µm in length . They were identified as elongated and adhesive structures with the tendency to form a net on the EM grids ( Figure 4 ) . Generally present as individual single pili of different thickness , they were also found to form bundles of individual pili ( data not shown ) . Among the individual single pili several classes could be identified based on their diameter and morphological variability ( Figure 5 ) . The most subtle filaments ( Figure 5A and 5D ) showed a linear morphology with no evident periodicity . A corresponding IEM displayed a linear distribution of gold particles binding to RrgB backbone proteins ( Figure 5D ) . We defined this type of filament as the pilus protofilament . Increasing filament widths ( Figure 5B and 5C ) resulted in an increasing complexity of the filaments , as clearly indicated by the higher number of gold particles decorating the filaments in a non-linear spacing ( Figure 5E and 5F ) . The majority of the pili could be assigned to the class of thin pili ( ∼63% ) ( Figure 5B and 5E ) with an observed average diameter of 9 . 5 nm , as calculated from cryo-EM data ( Figure 4 ) . The remaining ∼37% of the isolated pili were of larger diameters , the majority of which , individualised as class of thick pili , showed a width of about 10 . 5 nm ( data not shown ) . Further structural analyses were performed on the thin pili , more than 200 individual thin pilus segments were selected [34] from digitized micrographs ( Figure 4 ) obtained by cryo-EM on vitrified samples by using 300×300 pixel size boxes . Thus working with shorter segments that resulted to be approximately straight for the chosen box size was possible . Pili segments were treated as discrete single particles , 300 pixel in length and processed [35] by first aligning them rotationally and translationally to a reference cylinder centered into the image box . All segments that did not align with the reference were eliminated . Finally 124 segments of the thin pili ( Figure 6A ) were kept and used to generate averaged thin pili segments with an increased signal-to-noise ratio ( Figure 6B ) . Subsequently the diameter of the averaged thin pilus segments could be calculated from its density profile , by creating the profile ( IMAGIC5 ) [35] of the averaged segment generated from the 2D image of the pilus ( red line ) . The results showed a diameter of 9 . 6±0 . 3 nm for the thin pilus . Moreover the shape and the values of the density profile clearly indicated that thin pili were rather compact structures . Interestingly when a double Gaussian filter , where both , high and low frequency transmission were cut off , was applied on original thin pili data ( Figure 7A ) , the filtered 2D image ( Figure 7B ) , clearly showed that thin individual pili were composed of at least two protofilaments arranged in a coiled-coil superstructure . The Gaussian filtered image of the thin pilus showed zones where the 3 . 5 nm diameter protofilaments were tightly intersected ( crossovers ) , resulting in a pilus diameter of 6 . 8 nm . This was alternated by zones where protofilaments had a more relaxed intersection with a pilus diameter of 9 . 5 nm . The average distance between two neighbouring crossovers was approximately 13 nm . Preliminary results suggest that thick pili are also composed of similar protofilaments arranged in coiled-coil manner ( data not shown ) . In order to investigate adhesive properties of isolated pili and of single pilus subunits RrgA , RrgB and RrgC in vitro binding assays were performed to study the interaction to selected ECM components using this approach to provide a proof in principle [36] , [37] . In particular , fibrinogen , fibronectin , laminin , lactoferrin and collagen I were selected , as these cellular compounds are known to be recruited by pathogenic agents [9] , [38]–[40] . Apart from overall pili adhesive properties special interest was drawn to a potential role of distinctly surface exposed RrgA and RrgC . For this purpose , serial diluted samples of recombinant proteins RrgA , RrgB and RrgC , as well as native purified pili and a pilus negative control were added to 96-well plates coated with the selected ECM components . Binding was detected using polyclonal sera raised against the single recombinant pilus subunits and quantified by enzyme-linked immunosorbent assay ( ELISA ) . As demonstrated in Figure 8 ( lane A ) , RrgA showed very pronounced dose depending binding to most of the tested ECM compounds , whereas binding results obtained for RrgC and RrgB are negligible . In addition , RrgA binding was observed for lactoferrin and fibrinogen whereas no binding was detected to vitronectin coated plates ( data not shown ) . Bovine serum albumin ( BSA ) was used as negative control in all the assays . Binding studies performed using purified pili ( Figure 8 , lane B ) showed binding to ECM components clearly distinguishable from the Streptococcus pneumoniae delta pilus negative control .
Pili are considered important key players in bacterial pathogenesis and disease [10] , [27] . To date structural information of the native pilus in Gram-positive bacteria is lacking , therefore the elucidation of their structure and function are of great interest . Our approach consisted in obtaining native purified pili from a pathogenic strain of Streptococcus pneumoniae to study pilus structure and function . Special emphasis was drawn on the overall structural principle of the native pilus and the role of the individual structural proteins RrgA , RrgB and RrgC . As opposed to pili attached to the bacterial surface , isolated pili allow a broader spectrum of analyses and at the same time permit a comprehensive characterization of their structure in sufficient detail to describe the function at the quasi-molecular level . We developed a multi step purification procedure to obtain native pilus material that allowed to perform the desired analyses . T4 bacteria were examined by low dose negative stain EM , IEM and cryo-EM , showing that the bacterial surface is covered with elongated , flexible and rather sticky pilus-like appendages of up to 1 µm long . Interestingly , we observed pili of various morphologies: individual single pili , distinguishable into different classes by their diameter ( ranging from 9 . 5 nm up to 10 . 5 nm ) , and bundles of individual pili . Whether or not this has a physiological role has yet to be evaluated . The established purification method allowed for the isolation of pure HMW material that showed pili morphotypes having the same features as those found for wild-type pili expressed on whole bacteria . Structural analysis based on cryo-EM data of vitrified , purified single pili revealed that they are organised in coiled-coil superstructures made by at least two protofilaments . The observed range in pilus diameters could either reflect a difference in the degree of packaging of the identical protofilaments into the pili superstructures or a higher number of protofilaments composing the larger pili . The protofilaments of the thin pilus type are organized to form a rather compact superstructure . However no distinct internal cavity could be identified within the thin pilus structures . Preliminary results on the thick type of pilus suggest also a protofilament based structure . The picture of the individual pili that emerges from our analysis indicates that the Streptococcus pneumoniae pilus does not exist in a single structural state but rather in several structural states that are underlaying , among other things , the flexibility and elasticity of these polymers while keeping the same protein composition and proteins roles: RrgB forming the backbone , surface located clustered RrgA being the major ancillary protein involved in adhesion and RrgC as minor ancillary protein of still unknown role . Additional biochemical analysis of isolated pili supports RrgB as the main pilus building block: Mass analysis of native pili revealed clear signals only for peptides that could be assigned to structural protein RrgB . Neither RrgA nor RrgC related signals could be identified , which is probably due to their minor abundance and the overall hindered protease digestibility of the isolated HMW pili . Similarly , the determination of the N-terminal amino acid sequence of the purified pili by Edman analysis , matched only with the sequence following the predicted signal sequence of RrgB . The observation that purified pili show a free N-terminal part of RrgB starting exactly after the signal sequence may reflect properties of pilus biosynthesis and subsequently its structure . Purified pili of a Streptococcus pneumoniae delta RrgA background show a similar overall pili structure composed of protofilaments . This is in accordance with studies showing that a pneumococcal RrgA mutant strain is still able to form pili , whereas a ΔrrgB ΔrrgC strain is not [11] and fits with the detected structural organisation of RrgA clusters on a coiled-coil RrgB based scaffold . Gram-positive and Gram-negative pili differ substantially in their assembly mechanisms ( Gram-negative pili: non-covalently linked protein subunits versus Gram-positive pili: covalently linked subunits ) , interestingly both types of bacterial pili share a common arrangement , the coiled-coil superstructure . Our work supports that , also for Gram-positive bacteria , adhesive pili extending from the bacterial surface are the most appropriate structures to promote biological function like adherence to the host due to their structural arrangement leading to flexibility and elasticity . Until now this could be only observed in Gram-negative bacteria like Haemophilus influenzae type b pili and Escherichia coli P-pili [41] or Actinobacillus actinomycetemcomitans [42] . Results by Kang et al . [30] identified a novel principle of stabilization of long and thin pilus filaments by isopeptide linkage between pilus subunits of Gram-positive Streptococcus pyogenes . Further work will have to show whether similar design can also be found in other Gram-positive pili , like those of Streptococcus pneumoniae . Our results suggest that the coiled-coil arrangement of the protofilaments , forming the pneumococcal pili , might be an additional principle , other than isopeptide bond formation , to confer stability and flexibility to subtle surface structures in order to withstand mechanical rigors outside the cell . Research on bacteria and therefore also the study of the pneumococcal pilus should be seen in the context of bacterial life cycles within specific ecological niches and e . g . in the interaction with its host . Pneumococcal infection of the host occurs mainly via the mucosal route [1] , thus bacteria need to develop strategies to adhere and resist actions of the human immune system like mucosal clearance [43] . Studies performed using pilus negative mutants of T4 clearly demonstrate a positive correlation between bacterial virulence and colonization and the presence of the pilus [10] . We therefore wanted to study , if the structural data found for the isolated pili help us to better understand the functionality of pilus mediated pneumococcal behaviour within a host , and whether structural properties of the pneumococcal pilus could be function derived . How does the pneumococcal pilus mediate interaction with its host ? Our data suggest that pneumococcal pili are flexible protofilament-based structures composed of ancillary proteins RrgA and RrgC and the RrgB backbone ( Figure 9 ) . Recently , proteins of group B streptococcal pilus were found to facilitate the interaction with endothelial cells [44] . Our data elucidate the adhesive properties of RrgA to fibronectin , laminin and collagen , suggesting that the clusters containing RrgA are the adhesive regions of pili . In silico analysis of RrgA ( T4 ) sequence identified domains important for adhesion , like MSCRAMM motifs [19] and homologues of the von Willebrand factor A ( vWFA ) [45] . Interestingly , PapG , the adhesin of Escherichia coli P-pili , that binds to uroepithelial cells in its human host was also found to be located on the pilus surface , but only at the very distal end of the pilus fiber [46] . Streptococcus pneumoniae is a mucosal commensal , a mucosal pathogen and an invasive pathogen . Colonization of the nasopharynx by Streptococcus pneumoniae is a prerequisite for the development of pneumococcal disease and the result of a complex interplay between host and pathogen factors . Respiratory pathogens are known to release products which interfere with mucosal defences , causing epithelial disruption and cell death [47] , [48] . Streptococcus pneumoniae was seen to adhere in particular to damaged cells and extruded cells [47] , and bacteria were often found to be associated with damaged epithelium and exposed ECM [49] . Pathogen-ECM interactions have been found to be associated with adhesion and subsequent invasion of the pathogen [9] . Adhesive properties of pilus surface located ancillary protein RrgA to selected compounds of the ECM might therefore be part of the pilus mediated host-pathogen interplay . Flexibility of the pilus , as suggested by the protofilament-based structure , supports its functionality under in vivo conditions . Interestingly , recent work done by Nelson et al . [50] identified adhesive properties of pneumococcal pilus RrgA in cell-based assays . This together with data showing the impact of RrgA on pneumococcal virulence in mice [19] , [50] indicate that the polypeptide may function at more than one stage in the infection process . In summary , this report presents support for the structural composition of the Streptococcus pneumoniae pilus as an oligomeric appendage with adhesive properties and future work will help to further improve our understanding of the structure and function of the pilus and its main components .
Streptococcus pneumoniae type 4 strain TIGR4 has been described [20] . Mutants TIGR4Δpil ( rrgA-srtD ) [10] and TIGR4Δ ( rrgA ) [50] were kindly donated by B . Henriques-Normark ( Karolinska Institute , Stockholm ) . The pneumococcal strains were stored at −80°C in 12% glycerol and routinely grown at 37°C in 5% CO2 on Tryptic Soy Agar ( Becton Dickinson ) supplemented with 5% defibrinated sheep blood or in Tryptic Soy Broth ( Becton Dickinson ) . When appropriate , erythromycin ( Sigma-Aldrich ) as selection marker was used . Standard recombinant DNA techniques were used to construct all expression plasmids . Vector pET 21b+ was purchased from Invitrogen . Full length sequence of T4 pili proteins RrgA ( TIGR annotation No . sp0462 ) , RrgB ( TIGR annotation No . sp0463 ) and RrgC ( TIGR annotation No . sp0464 ) with exception of their N-terminal signal sequence and C-terminal cell wall sorting signal motif , hydrophobic stretch and charged tail was cloned into pET21b+: pellets of IPTG induced recombinant Escherichia coli BLR ( DE3 ) cultures , containing expressed His-tagged RrgA , RrgB and RrgC proteins respectively , were subjected to lysis by lysozyme in a BugBuster ( Novagen ) , Benzonase Nuclease ( Novagen ) solution containing proteinase inhibitors . After centrifugation at 35000 rpm for 1 h at 4°C , the soluble fraction was subjected to metal chelate affinity chromatography on His-Trap HP columns ( GE Healthcare ) equilibrated and eluted according to manufacturer's instructions . Pooled fractions were dialysed overnight ( ON ) against 0 . 9% NaCl and stored at −80°C until further use . Protein concentration and purity was determined by scanning densitometry of Coomassie Blue-stained SDS-PAGE using a BSA standard and measuring 280 nm absorption of the protein solution ( NanoDrop® ) . Streptococcus pneumoniae T4 was chosen as starting material as far as the bacteria belong to a clinical relevant serotype 4 isolate , the sequence of which is known [20] and it represents a well characterized pneumococcal strain . Streptococcus pneumoniae T4 glycerol stock ( −80°C ) was grown on Tryptic Soy Agar supplemented with 5% defibrinated mutton blood ( ON at 37°C in 5% CO2 ) . Fresh bacteria were used to incubate new agar plates and cultivated for about 12 h ( at 37°C in 5% CO2 ) . Harvested bacteria of about 10 plates were washed once in 35 ml PBS , and resuspended in 2 ml protoplast buffer PPB ( 10 mM MgCl2 , 50 mM NaPPi pH 6 . 3 , 20% sucrose ) containing protease inhibitors . About 450 U of mutanolysin in 100 mM NaPPi pH 6 . 3 were added to each half of the suspension and incubated at 37°C for about 5 to 8 h with gentle shaking until protoplast formation was detected ( microscopic control ) . Supernatant , containing digested pilus material was loaded on a sucrose gradient ( 25 to 56% in 10 mM MgCl2 , 50 mM NaPPi pH 6 . 3 ) and run for about 20 h at 38000 rpm ( 4°C ) . All further steps were performed at 4°C using buffers containing protease inhibitors . In addition , benzonase nuclease ( Novagen ) was added to remove DNA and RNA impurities . Collected gradient fractions were tested for pilus material using anti-RrgB antibodies . Pilus containing fractions were pooled and dialyzed against 10 mM MgCl2 , 50 mM NaPPi pH 6 . 3 for about one day to remove sucrose . When necessary , additional chromatography steps were added to reduce polydispersity and pooled pilus preparations were concentrated before loading them on a Superose 6 10/300 GL column ( Amersham Biosciences ) . Gel filtration resulted in separation of pilus containing material of different molecular weight . Purified pilus fractions were judged to be homogeneous based on EM and SDS-PAGE . Samples were stored at −80°C or liquid nitrogen until further use . Protein spots corresponding to HMW pili material were excised from SDS-PAGE gels ( 3–8% TA , Invitrogen ) , washed with 100 mM ammonium bicarbonate/ACN 50/50 v/v , and dried using a SpeedVac centrifuge ( Savant , Holbrook , NY , USA ) . Dried spots were digested for 2 h at 37°C in 12 ml of 0 . 012 µg/ml sequencing-grade modified trypsin ( Promega , Madison , WI , USA ) , in 50 mM ammonium bicarbonate . After digestion , 5 µl of 0 . 1% Trifluoroacetic acid ( TFA ) were added , and the peptides were desalted and concentrated with Zip-Tips ( C18 , Millipore ) . Samples were eluted with 2 µl of 5 g/l 2 , 5-dihydroxybenzoic acid in 50% ACN/0 . 1% TFA onto the mass spectrometer Anchorchip 384 ( 400 µm , Bruker Daltonics , Bremen , Germany ) , and allowed to air-dry at room temperature . MALDI-TOF spectra were acquired on a Bruker Ultraflex MALDI-TOF instrument ( Bruker Daltonics ) . Protein identification was carried out by both automatic and manual comparison of experimentally generated monoisotopic values of peptides in the mass range of 700–3000 Da with computer-generated fingerprints using MASCOT software running on proprietary databases . Identifications were confirmed by MS/MS analysis: after denaturing the samples in a MS-compatible detergent ( RapiGest SF , Waters ) and boiling for 15 min , in-solution digestion was performed by adding 2 µg of trypsin , and allowing digestion ON . MS/MS spectra were acquired using an ESI-q-TOF Micro mass spectrometer ( Waters ) , coupled to a nano-LC on a CapLC HPLC system ( Waters ) . A MS survey scan was used to automatically select multicharged peptides over the m/z range of 400–2000 for further MS/MS fragmentation . After data acquisition , the MS/MS spectra were combined , smoothed and centroided by MassLynx software , version 4 . 0 ( Waters ) . Search and identification of peptides were performed with a licensed version of MASCOT , in a local database , after converting the acquired MS/MS spectra in . pkl files . Identification of the N-terminal amino acid sequence of HMW pili material by Edman degradation was performed according to standard conditions . HMW pili material , following size exclusion chromatography , was applied to SDS-PAGE ( 3–8% TA; Invitrogen ) . After western transfer to PVDF membrane , HMW pili band was cut out and used for Edman analysis . SDS-PAGE analysis was performed using NuPAGE™ 3–8% Tris-Acetate Gels ( Invitrogen ) according to the instructions of the manufacturer . HiMark™ pre-stained HMW protein standard ( Invitrogen ) served as protein standard . Western analysis was done using standard protocols . Antibodies against recombinant HisTag-RrgB were used at 1/10000 dilution . Secondary goat anti-mouse HRP antibodies were used at 1/30000 . Antibodies against recombinant HisTag-RrgA ( mouse; guinea pig ) , -RrgB ( mouse ) , and -RrgC ( mouse; rabbit ) were produced in our lab and tested for specificity . Secondary goat anti-mouse HRP antibodies were obtained from Bio-Rad . Gold labelled antibodies for IEM were purchased of BBInternational: anti-mouse ( 5 nm ) , anti-rabbit ( 10 nm ) and anti-guinea pig ( 15 nm ) . 96-well MaxiSorp™ flat-bottom plates ( Nunc , Roskilde , Denmark ) were coated for 1 h at 37°C followed by an ON incubation at 4°C with 2 µg/well of respective ECM vitronectin ( from human plasma , BD Biosciences ) , lactoferrin ( from human milk , Sigma ) , collagen I ( from human lung , Sigma ) and fibrinogen ( from human plasma , Sigma ) and with 1 µg/well with laminin ( from human placenta , Sigma ) and fibronectin ( from human plasma , Sigma ) in phosphate-buffered saline pH 7 . 4 ( PBS ) . A BSA coated plate served as negative control . Plates were washed 3 times with PBS/0 , 05% Tween 20 and blocked for 2 h at 37°C with 200 µl of PBS/1% BSA followed by 3 washing steps with PBS/0 , 05% Tween 20 . Recombinant protein samples ( HisTag-RrgA , -RrgB and -RrgC ) were initially diluted to 4 µg/ml with PBS . 200 µl of protein solution or 100 µl of wild type pilus preparation ( 53 µg/ml ) and 100 µl T4Δpil sample ( 35 µg/ml ) , diluted in 200 µl total volume with PBS , and respective controls were transferred into coated-blocked plates in which the samples were serially diluted two-fold with PBS , obtaining a final volume of 100 µl/well . Plates were incubated for 2 h at 37°C and ON at 4°C . The plates were washed 3 times and incubated for 2 h at 37°C with respective primary mouse anti-HisTag-Rrg antibodies ( 1/10000 dilutions ) ; pilus coated plates were incubated with anti-HisTag-RrgB antibodies . After another 3 washing steps , antigen-specific IgG was revealed with alkaline phosphatase-conjugated goat anti-mouse IgG ( Sigma Chemical Co . , SA Louis , Mo . ) after 2 h of incubation at 37°C , followed by addition of the phosphatase alkaline substrate p-nitrophenyl-phosphate ( Sigma ) . Read out was performed at 405 nm by an ELISA plate reader . Formvar-carbon-coated nickel grids were charged with 5 µl of purified sample and let stand for 5 min . The grids were then fixed in 2% paraformaldehyde ( PFA ) in Phosphate Buffered Saline 0 . 1 M pH 7 . 4 ( PBS ) for 5 min , and placed in blocking solution ( PBS containing 1% normal rabbit serum and 1% BSA ) for 1 h at room temperature . The grids were then floated on drops of polyclonal antibodies α-RrgA ( guinea pig ) , α-RrgB ( mouse ) and α-RrgC ( rabbit ) at dilution of 1∶10 in blocking solution for 1 h at room temperature , washed with 5 drops of blocking solution for 5 min , and floated on secondary gold-conjugated antibodies ( goat anti-mouse IgG , 5 nm; goat anti-rabbit IgG , 10 nm; goat anti-guinea pig IgG , 15 nm ) diluted 1∶20 in blocking buffer for 1 h . The grids were then washed with five drops of PBS and fixed in 2% PFA/PBS for 5 min at room temperature . Finally samples were washed with 8 drops of distilled water . Grids were stained with 1% buffered phosphotungstic acid ( PTA ) ( pH 6 . 5 ) for 15 s , the excess of solution was soaked off by Whatman filter paper . The grids were examined in a CM10 Transmission Electron Microscope ( TEM , Philips Electronic Instruments , Inc ) operating at 80 kV . The command boxer from software EMAN was used to isolate and to count the single gold particles of different sizes . 20 µl of the solution containing purified pili were transferred onto a cover slip that had been previously cleaned by immersion in chromic acid solution followed by several rinses in distilled water . Pili were allowed to sediment on the glass surface then the cover slips were rinsed in distilled water to remove the excess of material . Immediately before freezing each cover slip was rapidly rinsed in distilled water , and a thin meniscus of solution was left on the glass to prevent dehydration of the samples . While the freezing machine was brought to its lowest temperature 4oK , the tiny glass was placed onto a thin slice of aldehyde fixed lung for support during freezing . This was accomplished by slamming the samples onto the liquid helium-cooled copper block of a quick freezing device ( Cryopress; Med-Vac , Inc . , St . Louis , MO ) . The frozen samples were freeze dried in a freeze etching unit ( Baf 301; Balzers S . p . A . , Milan , Italy ) for 20 min at −80°C . Pilus absorbed to the cover slip were rotary replicated with ∼2 nm of platinum applied from an angle of 24° above the horizontal and then backed with 25-nm-thick film of pure carbon . Replicas were separated from the glass by immersion in concentrated hydrofluoric acid then cleaned with sodium hypochlorite . After several rinses in distilled water replicas were picked up on 75-mesh formvar-coated microscope grids . Samples were viewed in a transmission electron microscope ( CM10; Philips Electronic Instruments , Inc , Mahwah , NJ ) operating at 80 kV . 5 µl aliquots of whole bacteria were applied to 200-mesh copper grids coated with a thin carbon film and let stand for 5 min . The grids were first washed by streaming several drops of PBS over the grids . They were subsequently negatively stained by two drops of 1% buffered PTA ( pH 6 . 5 ) . The last drop was left on the grids for 17 s . Finally the grids were washed with several drops of ddH2O , the excess of liquid was soaked off by Whatman filter paper and quickly air dried . The grids were observed using a CM200 FEG Philips Electron Microscope ( FEI , Eindhoven , The Netherlands ) , equipped with a GATAN GIF 2002 postcolumn energy filter ( Gatan , Pleasanton , California , United States ) , and images were collected at an accelerating voltage of 200 kV and a nominal magnification of 50000× , on Kodak SO163 film . 5 µl of purified pili preparation were loaded onto a glow discharged Quantifoil holey carbon grid with 2 µm holes . After being blotted from the front side with a slip filter paper ( Whatman No . 4 ) , the grid was flash frozen into liquid ethane as described [51] . Micrographs taken at 50000× of magnification were digitized on a IMACON 949 scanner at spacing of 7 . 95 µm resulting in a nominal sampling of 1 . 6 Å/pixel-1 . Pili were picked from digitized images using the command “helixboxer” from the software EMAN [34] . Digitized pili images were cut into individual repeats by using boxes of 300×300 pixels , with overlapping ends , using 10 pixel shift for each box , so that adjacent boxes had 90% overlap . The isolated repeats were treated as single particles . In a first analysis , the straightest pili segments were selected and pre-aligned interactively , subsequently the pre-aligned repeats were aligned using alignments with only limited angular ranges ( −5° , +5° ) , finally a vertical alignment has been performed using as a future-less reference the projection of a model cylinder followed by translational alignment perpendicular to the cylinder axis only . Aligned repeats were than subjected to high-pass and low-pass filtrations before the density profiles were calculated ( the densities across the filament axis of the pili were projected onto the short axis ) using different commands of IMAGIC 5 [35] and of Bsoft software [52] . All the aligned and filtered images were consistent: they all presented centred rods with similar diameters . The only major differences were the surrounding stain distributions . Swiss-Prot ( http://www . expasy . org/sprot/ ) accession numbers for pilus proteins mentioned in the text are: SP_0462 , RrgA ( TIGR4 ) Q97SC3 SP_0463 , RrgB ( TIGR4 ) Q97SC2 SP_0464 , RrgC ( TIGR4 ) Q97SC1 | Streptococcus pneumoniae ( pneumococcus ) is one of the most important human pathogens and a major cause of morbidity and mortality worldwide , causing respiratory tract infections , community acquired pneumonia , and invasive diseases . Although the pneumococcus is a well-studied bacterial pathogen , first described in the late 19th century , pili on its surface were discovered only recently . Pili are elongated structures extruding from the bacterial surface and were found to be important virulence factors of both Gram-positive and Gram-negative bacteria . Bacterial pili are considered to participate in bacterial adhesion to a host , a crucial step in bacterial infection . In contrast to Gram-negative pili , little is known about the structure of native Gram-positive pili . We used native purified pili of pathogenic pneumococcus TIGR4 to study its structural composition , mainly by the use of cryo EM techniques . Pili were found to be composed of protofilaments that are arranged in a coiled-coil , compact superstructure of various diameters . Adhesive properties of pilus surface located ancillary protein RrgA to selected compounds of the extracellular matrix might be part of the pilus mediated host–pathogen interplay . Analysis of native pneumococcal pili revealed structural basics of a Gram-positive pilus that could also serve as a basis for effective vaccine design . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biochemistry",
"biochemistry/macromolecular",
"assemblies",
"and",
"machines",
"microbiology",
"infectious",
"diseases/respiratory",
"infections",
"infectious",
"diseases/bacterial",
"infections",
"microbiology/cellular",
"microbiology",
"and",
"pathogenesis"
]
| 2008 | Pneumococcal Pili Are Composed of Protofilaments Exposing Adhesive Clusters of Rrg A |
Reports of therapeutic failure to meglumine antimoniate ( MA ) and miltefosine in cutaneous leishmaniasis ( CL ) varies between species , populations and geographic regions . This study aimed to determine the clinical , drug-related factors , and Leishmania species associated with treatment failure in children and adults with cutaneous leishmaniasis . A cohort study was performed with children ( 2–12 years old ) and adults ( 18–65 years old ) with CL , who have participated in clinical studies at CIDEIM Cali , Tumaco and Chaparral . Incidence of therapeutic failure was estimated by treatment and age groups . Descriptive , bivariate , and multiple logistic regression analyses were performed for the complete cohort and pediatric patients . Two hundred and thirty patients were included ( miltefosine: 112; MA: 118 ) , of which 60 . 4% were children and 83 . 9% were infected with L . V . panamensis . Overall incidence of therapeutic failure was 15 . 65% ( 95%CI: 10 . 92–20 . 38 ) , and was lower for miltefosine than for MA ( 8 . 92% , 95%CI: 3 . 59–14 . 26 versus 22 . 03% , 95%CI:14 . 48–29 . 58 , p = 0 . 006 ) . Treatment failure was associated with age ≤8 years ( OR: 3 . 29; 95%CI: 1 . 37–7 . 89 ) , disease duration ≤1 month ( OR: 3 . 29; 95%CI: 1 . 37–7 . 89 ) , regional lymphadenopathy ( OR: 2 . 72; 95%CI: 1 . 10–6 . 70 ) , treatment with MA ( OR: 3 . 98; 95%CI: 1 . 66–9 . 50 ) , and adherence <90% ( OR: 3 . 59; 95%CI: 1 . 06–12 . 11 ) . In children , higher Z-score of height/age was a protective factor ( OR: 0 . 58; 95%CI: 0 . 36–0 . 93 ) , while treatment with MA was a risk factor ( OR: 40 . 82; 95%CI: 2 . 45–677 . 85 ) , demonstrating significant interaction with age ( p = 0 . 03 ) . Clinical and drug-related factors determine therapeutic failure in CL . High risk of failure in children treated with MA indicates the need to reconsider this drug as first line treatment in this population . Clinical trial registration: NCT00487253 Clinical trial registration: NCT01462500 Clinical trial registration: NCT01464242
Cutaneous leishmaniasis is a public health problem with the number of cases annually ranging from 0 . 7–1 . 2 million worldwide [1] . Seventy-five percent of the global burden of disease is concentrated in ten countries , including Colombia , where cutaneous leishmaniasis is an important cause of morbidity [1 , 2] . In the Americas , cutaneous leishmaniasis is caused principally by species of the Viannia subgenus . Clinical manifestations range from mild , localized , self-healing lesions to numerous disseminated cutaneous lesions , severe chronic cutaneous and mucosal lesions . Spontaneous cure of cutaneous lesions , frequently leaving scars , occurs in less than 30% of cases , of which 20% result in relapse after initial healing [2] . Treatment involves administration of often toxic and poorly tolerated drugs [3] . For over 5 decades , first line treatment for all age groups has been parenteral antimonial drugs whose efficacy varies across regions , age groups , and Leishmania species . Treatment success with these drugs can be as low as 25% in children younger than 5 years old [4 , 5] . Another available and recommended option is oral miltefosine , which has comparable efficacy to meglumine antimoniate in children [6]; its cure rates also vary by species and geographic location [6–9] , and is as low as 60% in military populations [7] . Second line options include amphotericin B and pentamidine , which are more toxic and require parenteral administration [5 , 10] . The identification of determinants of treatment response in American cutaneous leishmaniasis provides the basis for determining high risk patients , to orient the selection of treatment regimens , and to design interventions for those factors that are modifiable . Together these measures contribute to the preservation of the useful life of current medications . Known determinants of treatment outcome in cutaneous leishmaniasis include adherence to the treatment , Leishmania species , number and location of lesions , duration of the disease , and age [11–13] . However , the studies that have identified these associations have focused on adult patients and treatment with antimonials . Consequently , little is known about factors associated with treatment response in pediatric populations , or regarding treatment with miltefosine . This study sought to determine the clinical and drug-related factors , and Leishmania species associated with treatment failure in children and adults with cutaneous leishmaniasis .
This study was approved and monitored by Universidad del Valle ( approval number: 012–014 ) and CIDEIM’s institutional review boards for ethical conduct of research involving human subjects , and followed national and international clinical research ethics guidelines . Waiver of informed consent for the use of data was requested and accepted . We designed a cohort study using secondary data from four clinical studies conducted by investigators and collaborators of the Centro Internacional de Entrenamiento e Investigaciones Medicas ( CIDEIM ) between 2007 and 2013 in three municipalities of Colombia . The largest study was a non-inferiority trial ( Clinical trial registration: NCT00487253 ) comparing miltefosine and pentavalent antimony in children; the second was a pharmacokinetic trial of miltefosine in children and adults ( Clinical trial registration: NCT01462500 ) , the third was an immunologic study of patients treated with pentavalent antimony , and the smallest was an add-on trial evaluation of pentoxifylline or placebo to the antimonial treatment ( Clinical trial registration: NCT01464242 ) . The participants were recruited within the original studies on an outpatient basis in three municipalities in the central and southwestern regions of Colombia: Tumaco ( 2°48´24” N , 78°45´53” W ) , which is located in the southern Pacific coast of Colombia and is an endemic region for L . panamensis and L . braziliensis; Cali ( 3°26′13″ N; 76°31′20″ W ) , which is a referral center for cases of cutaneous leishmaniasis from the southwestern region of the country , and Chaparral ( 3°43′23″ N; 75°28′59″ W ) , which is an endemic area located in the central region of Colombia and the site of a recent epidemic of cutaneous leishmaniasis caused by L . guyanensis [6] . Eligible participants in the clinical studies were children aged 2–12 years or adults aged 18–60 years with parasitologically confirmed cutaneous leishmaniasis ( positive direct smear , culture of lesion aspirates or biopsy , as described in the original studies ) . Parasite identification was performed using subgenus and species discriminating monoclonal antibodies [14] . For the present study , those who received treatment with pentavalent antimony or miltefosine and completed the study follow-up scheme ( minimum 13 weeks or 90 days if pentavalent antimony , and minimum 26 weeks or 180 days if miltefosine ) were included . Patients’ records that did not include assessment of therapeutic response or have missing data regarding weight or treatment information ( doses prescribed and received ) were excluded . All patients who met eligibility criteria were included in the analysis . The main outcome of this study is therapeutic failure , determined at or before 26 weeks following initiation of treatment , according to the criteria described by Rubiano and colleagues [6] . Cure was defined as complete re-epithelization and the absence of inflammatory signs for all cutaneous leishmaniasis lesions at day 90 ( week 13 ) , and maintained until the end of the follow-up . Therapeutic failure was defined as incomplete re-epithelization and/or the presence of induration , raised borders , or redness in any lesion after day 90 , relapse ( reactivation of lesions after initial cure ) or the appearance of new lesions . Exposure variables were classified as related to the host , drug or parasite , including age , sex , number , duration and location of lesions , concomitant adenopathy , Leishmania species , dose and type of medication . Height and weight data were used to calculate Z-scores in pediatric patients , using the Anthro plus software [15] . All variables were measured at baseline , except for the adherence to the antileishmanial drug which was defined at the end of treatment as the proportion of received doses over the total doses prescribed . We considered as compliant those patients with adherence ≥90% . Outcome and exposure variables were obtained from databases of the original studies , which followed similar operation procedures for measurement of these variables . All studies controlled adherence using diaries and product count . None of the outcome measurements were masked . Overall risk of therapeutic failure was estimated for the complete cohort . Quantitative exposure variables were compared using Student´s T test or Mann–Whitney test according to normal or skewed distribution , respectively . For qualitative variables , the Chi-squared test or Fisher´s exact test were used when appropriate . Relative risks and their corresponding 95% confidence intervals were calculated . Multiple logistic regression modeling was used to estimate the Odds Ratio of therapeutic failure using the backward selection technique; a p-value <0 . 05 was considered statistically significant . Interactions between treatment , age , and Leishmania species were assessed in the final model . Subgroup analysis was performed in children ( ≤12 years old ) and in a group of patients who completed follow-up to 26 weeks , as part of a sensitivity analysis , in order to evaluate the change in the OR when participants with follow-up to 13 weeks were removed from the model . This was used to evaluate the presence of outcome misclassification in patients who were treated with meglumine antimoniate and did not have follow-up at 26 weeks ( study number 3 ) . Age categories ( ≤8 years old vs >8 years old ) were used in the analysis of the complete cohort , considering reported differences in efficacy of meglumine antimoniate in children [6]; however , age was analyzed as a continuous variable for subgroup analysis in pediatric patients . Goodness of fit test , Area Under the Curve ( AUC ) , and Akaike information criteria were used for model selection . All the analyses were performed using STATA 10 .
A total of 248 patients were enrolled in the four clinical studies , of which 230 were eligible for the cohort study ( Fig 1 ) . The age distribution ranged from 2 to 60 years old , with a mean of 10 years old . Ninety-five were female ( 41 . 3% ) and 78 . 4% came from the Pacific region of Colombia . Species identification of Leishmania parasites was available in 62 . 1% of participants , of which 83 . 9% were L . V . panamensis . Regarding the clinical presentation , 84 . 6% of lesions were ulcers , with median diameter of 22mm ( IQR: 14 –32 ) , located mainly on the arms and legs ( 87 . 1% ) . The median duration of the oldest lesion was 2 months ( IQR: 1–3 ) , and the median number of lesions was 1 ( IQR: 1–3 ) . Fifty-one percent of patients received treatment with miltefosine . The proportion of non-adherence was 9 . 5% ( <90% adherence , 22 patients ) for all treatment regimens taken together . The proportion of non-adherence was higher in the group treated with antimonial versus miltefosine ( 14 . 41% vs 4 . 46% , p = 0 . 009 ) . Descriptive statistics and relative risk of failure were estimated ( Table 1 ) . Overall incidence of therapeutic failure was 15 . 6% ( 95%CI: 10 . 92–20 . 38 ) , with a higher incidence of failure in patients treated with pentavalent antimony compared to miltefosine ( 22 . 03% , 95% CI: 14 . 48–29 . 58 vs . 8 . 93 , 95%CI: 3 . 59–14 . 26 ) , p = 0 . 006 . In the univariate analysis , therapeutic failure was associated with duration of the oldest lesion being less than 1 month ( RR: 1 . 94 , 95%CI: 1 . 06–3 . 54 ) , regional lymphadenopathy ( RR: 1 . 94 , 95% CI: 1 . 02–3 . 68 ) , and treatment with pentavalent antimony ( RR: 2 . 46; 95%CI: 1 . 24–4 . 88 ) . Height/age Z-score was significantly lower in children with therapeutic failure than children who were cured ( p = 0 . 04 ) . The analyses did not reveal an association between treatment outcome and Leishmania species , previous episodes of leishmaniasis , other clinical or socio-demographic variables . There was no difference in the incidence of failure by year of recruitment ( p = 0 . 46 ) . Factors independently associated with treatment failure included: age ≤ 8 years old ( aOR: 3 . 29; 95%CI: 1 . 37–7 . 89 ) , duration of the oldest cutaneous lesion ≤ 1 month ( aOR: 2 . 85; 95%CI: 1 . 29–6 . 28 ) , regional lymphadenopathy ( aOR: 2 . 72; 95%CI:1 . 10–6 . 70 ) , treatment with meglumine antimoniate ( aOR: 3 . 98; 95%CI: 1 . 66–9 . 50 ) , and less than 90% adherence to the treatment ( aOR: 3 . 59; 95%CI: 1 . 06–12 . 11 ) ( Table 2 ) . No association of therapeutic response was evident with the Leishmania species isolated from the patient , and we did not find significant interactions between treatment , age , or species . However , given that 62 . 1% of patients had data regarding the identity of the infecting species , a model including this variable diminishes the sample size to 58% of the overall cohort , hence , statistical power to make inferences regarding parasite species . In the pediatric population ( Table 3 ) , we found that nutritional status , represented by the Z-score of height/age , was a predictor of treatment response because the odds of failure decreased 48% for every unit of increase in the Z-score ( aOR: 0 . 55; 95%CI: 0 . 31–0 . 86 ) . Presence of regional lymphadenopathy was also identified as a risk factor in this group . Treatment with pentavalent antimony in children showed a statistically significant association with failure , more so than in the rest of the overall study population ( aOR: 40 . 82; 95% CI: 2 . 45–677 . 85 ) . A significant interaction between age and treatment with meglumine antimoniate was found , where the odds of treatment failure decreased for each additional year in patients treated with this drug ( aOR: 0 . 63; 95% CI: 0 . 41–0 . 96 , p = 0 . 03 ) , as opposed to those patients treated with miltefosine ( effect of age: aOR: 0 . 99; 95% CI: 0 . 77–1 . 27 ) . We found neither a statistically significant association nor interaction between Leishmania species and the study outcome . The models shown in Tables 2 and 3 fit the data . The first model ( Table 2 ) with p = 0 . 29 in the Goodness of fit test and the area under the ROC curve ( AUC ) = 0 . 76 indicates a good data discrimination; the second model ( Table 3 ) had estimated values of p = 0 . 49 and AUC = 0 . 81 . Sensitivity analysis showed that after excluding the patients without follow-up at 26 weeks ( n = 43 ) , the explanatory variables in the model remained statistically significant , although the aOR of age ≤ 8 years and adherence increased to 8 . 37 ( 95% CI: 2 . 16–32 . 41 ) and 19 . 36 ( 95% CI: 3 . 14–119 . 15 ) , respectively ( S1 Table ) .
This study assessed risk factors for therapeutic failure in children and adults with parasitologically confirmed cutaneous leishmaniasis in Colombia . The overall incidence of therapeutic failure was 15 . 65% ( 95% CI: 10 . 92–20 . 38 ) , which was higher with pentavalent antimony than miltefosine [16] . Estimates of treatment failure with meglumine antimoniate ( 22 . 03% , 95%CI: 14 . 48–29 . 58 ) are similar to previous reports from Colombian studies [6 , 17] , although they are lower than reported in other regions of Latin America [11 , 12] . In this cohort , the proportion of treatment failure with miltefosine ( 8 . 92% , 95%CI: 3 . 59–14 . 26 ) was lower than reported in other studies conducted in children and adults [6 , 7] , although it was similar to the proportion reported by Soto et al in patients with L . panamensis infection in other regions of Colombia [8] . The lower proportion of treatment failure can partially be explained by the characteristics of this cohort of patients being enrolled in clinical studies [18] , under supervised or directly-observed treatment , which are interventions that have shown a positive effect regarding the therapeutic response to antimicrobials [19] . Due to the inclusion of patients with either pentavalent antimony or miltefosine medications , we were able to identify antimonial treatment as an independent risk factor for therapeutic failure ( OR: 3 . 98; 95%CI: 1 . 66–9 . 50 ) . This drug has been the first line treatment for over 70 years , with its efficacy ranging from 70–85% [6 , 20] . Although the efficacy of meglumine antimoniate in this study falls in this range , the efficacy of miltefosine is higher . Possible explanation of the high cure rate of miltefosine is related to the predominance of one Leishmania species in the study and its known parasite susceptibility profile . Variations in parasite susceptibility to antileishmanial drugs have previously been reported by species and geographic locations [5 , 6 , 20] . In this study , the largest proportion of participants were infected in the pacific coast of Colombia , where L . V . panamensis is predominant and has shown a better in vitro susceptibility to miltefosine compared with other species isolated from the eastern parts of the Andean and Orinoco regions of the country [21] , explaining , at least partially , the good clinical response to miltefosine . Age under or equal to 8 years old was identified as a predictor of failure in the complete cohort , independent of the other factors ( OR: 3 . 59; 95%CI: 1 . 06–12 . 11 ) . Among children , younger age was also associated with increased odds of therapeutic failure . This finding is consistent with previous studies [4 , 11 , 12] . Differences in the pharmacokinetics of antileishmanial drugs in children are a possible explanation . Clearance of pentavalent antimony in children is faster than in adults , and therefore the maximum plasma concentration ( Cmax ) and the Area Under the Curve ( AUC ) of the drug are lower compared to adults [22] . Regarding miltefosine , analysis using Monte Carlo simulations of PK data from patients with visceral leishmaniasis have described lower plasma concentrations in children under the current linear dosing regimen ( mg/kg ) [23] . Additionally , reports of a clinical trial aimed at evaluating the pharmacokinetics of miltefosine in children shows lower Cmax and AUC in children at the same dose regimen ( clinicaltrials . org number: NCT01462500 ) [24] . This is important because miltefosine seems to be a time-dependent antimicrobial drug , and the risk of failure in patients with visceral leishmaniasis is increased with the number of days below 10xEC50 miltefosine plasma concentrations [25] . Age effect also varies according to the prescribed drug in the study population , as shown with a statistically significant interaction between age and meglumine antimoniate ( p = 0 . 03 ) . This suggests that antimonial treatment in children is related to treatment failure , and is concordant with differences in treatment efficacy by age groups , as previously described [4] . Allometric dosing of miltefosine have been proposed as an alternative to improve drug exposure in children , and its safety is being evaluated currently for VL patients ( NCT02431143 and NCT02193022 ) ; however , little is known about feasibility or safety of alternative dosing regimens for antimonials . Despite these facts , meglumine antimoniate is still the first line treatment in all age groups in Colombia , which highlights the urgent need to reconsider the management of pediatric cutaneous leishmaniasis with improved dosing or alternative treatments [26] . Partial immunity attributed to antigen exposition in endemic areas and differences in immune response in children can be related to the effect of age on treatment response , as described in other parasitic infections [27] . This hypothesis is supported by the increment of prevalence of Leishmania infection by age groups in this endemic area [28] , which might imply a lower exposure to Leishmania in younger children . Another determinant of therapeutic outcome was disease duration . In cutaneous leishmaniasis , early treatment has been described as a risk factor for treatment failure [11 , 12] , and management with antimonials before 20 days of disease appearance did not prevent the ulceration of lesions , therefore being associated with worse prognosis [29 , 30] . This is explained partially by the role of immune response in treatment outcomes [30] , when patients are treated before reaching protective acquired immunity . Surrogates of immune response such as the diameter of Montenegro skin test ( MST ) , gamma interferon ( INF-γ ) , and TNF-α ( in supernatant of cell cultures ) were reduced in patients with early treatment in a previous study [29] . Additionally , induration of MST was described as a determinant of therapeutic outcome in CL in a recent study , where short duration of the disease was another risk factor for failure ( OR: 6 . 33; 95%CI: 2 . 52–15 . 90 ) , supporting the impact of variables related to the host immunity in response to medications [31] . Poor treatment adherence and regional lymphadenopathy were independent factors for failure . Among these , irregular treatment was reported by Rodriges et al . as a determinant of poor outcome in CL , ( RR: 1 . 85; 95%CI:1 . 33–2 . 56 ) [13] , and is probably related to the lower drug exposure in this group of patients . Regional lymphadenopathy is an early sign of Leishmania infection , especially L . V . braziliensis , which could be present before the ulceration of cutaneous lesions [30 , 32] . In this study , presence of lymphadenopathy was independent of disease duration ( presence of lymphadenopathy in patients with disease duration ≤1 month: 34 . 2% vs >1 month: 65 . 8% , p = 0 . 49 ) . Therefore , we hypothesized that it can be an indicator of disease severity or lymphatic dissemination of parasites , as described elsewhere [33 , 34] . It may also be an indicator of the relationship between the reticuloendothelial system as a site of parasite persistence [35] , although the nature of that relationship remains still unclear [35 , 36] . This is a new finding that warrants further investigation . Higher height/age z-score values in children were associated with a decrease in the odds of treatment failure ( OR: 0 . 52; 95%CI: 0 . 31–0 . 86 ) , suggesting that a better nutritional status is a protective factor . Low height/age z-scores can be used to identify children at risk of stunting [37 , 38] , and is a robust measure for population nutritional studies [39 , 40] . Few studies have confirmed the relationship between malnutrition and risk of Leishmania infection [41 , 42] and a small study in adults did not find a relationship between weight and time-to-cure in tegumentary leishmaniasis [43] . However , using this indicator , we provide evidence of the influence of nutrition on treatment response in children with CL , as has been described in other infectious diseases [44 , 45] , murine models of leishmaniasis [46] , and in a small descriptive study conducted in children treated with miltefosine [47] . Possible explanations include the negative effect of malnutrition in delayed immune response , due to deficiency in vitamins A , C , E and minerals such as Zinc , among other factors [44 , 45] . Influence of parasite species on clinical outcome is shown in several studies [5 , 11 , 12] , but we were unable to identify this association . Low proportion of isolates ( 62% ) and the predominance of L . V . panamensis species in over 83% of participants can explain this lack of association , which is a limitation of this study . Another important limitation of this study is the imputation of data from treatment outcomes in participants treated with meglumine antimoniate with follow-up at 13 weeks ( 90 days ) , which represents the end of follow-up in one of the studies included in the analysis . In order to measure its impact , we performed a sensitivity analysis excluding these patients , and the risk factors remained significant with similar force of association ( S1 Table ) . Moreover , attendance to follow-up visits can be as low as 5% at six months , which is probably the most common scenario in monitoring treatment response in CL due to barriers to accessing health facilities in endemic areas . Despite these limitations , the standardized measurement of treatment outcome , adherence , and baseline characteristics allows for comparisons between subjects , and overcomes some restrictions involved with assessing risk factors based on routine data . Moreover , the proportion of pediatric patients is large ( 60% ) , which enables stronger inferences in this group compared with other studies of treatment failure in CL . Although data were collected in clinical studies , we included different sites from two endemic areas of the country , which allowed us to generalize our findings to CL patients from central and southwestern Colombia with predominance of L . V . panamensis . In conclusion , we provide evidence regarding the risk factors for treatment failure in adult and pediatric populations with cutaneous leishmaniasis . Modifiable characteristics , including timing of treatment , nutritional status , and use of antimonials , were identified as potential interventions to improve therapeutic success . The study highlights the urgent need to reconsider pentavalent antimony as first line treatment in children in Colombia where L . V . panamensis predominates . | Cutaneous leishmaniasis ( CL ) is a parasitic disease that causes chronic , often ulcerated , skin lesions . Treatment require administration of systemic and poorly tolerated drugs , of which , the most commonly used is meglumine antimoniate ( MA ) injections during 20 days . Although children and adults might have different responses to these drugs , Colombian treatment guidelines recommends meglumine antimoniate for all age groups . In this study , we explored the factors that influence the therapeutic response in children and adults with CL treated with MA and miltefosine . We included 230 children and adults in the analysis , and we found that young age ( ≤8 years old ) , presence of regional lymphadenopathy , disease duration ≤1 month , poor adherence to treatment ( <90% ) were associated with increased odds of treatment failure . Additionally , being treated with MA was a risk factor for therapeutic failure , especially for children . A better nutritional status ( higher Z-score of height/age ) was a protector factor in pediatric patients . These results highlight the urgent need to reconsider MA as first line treatment in children in Colombia and to evaluate better treatment options for this population . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
]
| [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"tropical",
"diseases",
"geographical",
"locations",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"pediatrics",
"protozoans",
"signs",
"and",
"symptoms",
"pharmaceutics",
"leishmania",
"antimony",
"neglected",
"tropical",
"diseases",
"infectious",
"diseases",
"zoonoses",
"south",
"america",
"lesions",
"chemistry",
"protozoan",
"infections",
"people",
"and",
"places",
"chemical",
"elements",
"diagnostic",
"medicine",
"colombia",
"leishmaniasis",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"drug",
"therapy",
"organisms"
]
| 2017 | Risk factors for therapeutic failure to meglumine antimoniate and miltefosine in adults and children with cutaneous leishmaniasis in Colombia: A cohort study |
A major challenge in systems biology is to understand how complex and highly connected metabolic networks are organized . The structure of these networks is investigated here by identifying sets of metabolites that have a similar biosynthetic potential . We measure the biosynthetic potential of a particular compound by determining all metabolites than can be produced from it and , following a terminology introduced previously , call this set the scope of the compound . To identify groups of compounds with similar scopes , we apply a hierarchical clustering method . We find that compounds within the same cluster often display similar chemical structures and appear in the same metabolic pathway . For each cluster we define a consensus scope by determining a set of metabolites that is most similar to all scopes within the cluster . This allows for a generalization from scopes of single compounds to scopes of a chemical family . We observe that most of the resulting consensus scopes overlap or are fully contained in others , revealing a hierarchical ordering of metabolites according to their biosynthetic potential . Our investigations show that this hierarchy is not only determined by the chemical complexity of the metabolites , but also strongly by their biological function . As a general tendency , metabolites which are necessary for essential cellular processes exhibit a larger biosynthetic potential than those involved in secondary metabolism . A central result is that chemically very similar substances with different biological functions may differ significantly in their biosynthetic potentials . Our studies provide an important step towards understanding fundamental design principles of metabolic networks determined by the structural and functional complexity of metabolites .
Cellular metabolism is mediated by highly efficient and specialized enzymes catalyzing chemical transformations of substrates into products . Since the products of a particular reaction may serve as substrates for other reactions , the entirety of the biochemical reactions forms a complex and highly connected metabolic network . With the sequencing of whole genomes of an ever increasing number of organisms and the emergence of biochemical databases such as KEGG [1] , Brenda [2] or MetaCyc [3] , which are based on genomic information , large-scale metabolic networks have become accessible . The KEGG database , for example , holds biochemical reactions of several hundred organisms , forming a metabolic network with over 6000 reactions connecting over 5000 metabolites . Whereas the wiring principles of small metabolic systems such as single biochemical pathways or a small number of interacting pathways are generally easily comprehensible , elucidating the organization of large-scale metabolic networks still poses a major challenge in the field of systems biology . While a network of 6000 reactions is large in the sense that it is computationally challenging , this number represents only a tiny fraction of all theoretically possible , chemically feasible reactions . So why did enzymes evolve for these reactions but not for others ? And what were the selective pressures that lead to this particular selection ? While we are still far from answering these intriguing questions satisfactorily , it is plausible to assume that the selection was not random but a result of a long evolutionary process which must have left its imprint in the structure of the contemporary metabolic network . We use this assumption as our working hypothesis and identify an interesting hierarchical organization which seems to be an intrinsic property of metabolism and robust against moderate changes in network structure and other specific assumptions like the availability of particular chemicals . Our results inspire some speculations on the above raised questions and we outline some possible continuations of this work with the aim to get further insight into the principles that guided metabolic evolution . Several approaches to analyze the structure of large-scale metabolic networks have emerged in recent years . Graph theoretical approaches have revealed characteristic global features . It has been shown that metabolic networks exhibit a small world character [4] , possess a scale-free topology [5] and display a hierarchical organization [6] . However , all these approaches rely on a representation of a metabolic network as a graph . There are many alternative ways to construct a graph from a metabolic network ( see for example [7] ) . A characteristic of most of the applied approaches is that it is in general not possible to reconstruct the original metabolic network from the graph , since in the simplification process important biochemical information is lost . Moreover , graph theoretical results may strongly depend on the particular representation . For example , the small world property has been shown for a graph , in which the nodes represent metabolites connected by an edge if they participate together in a biochemical reaction . If , however , metabolites are only connected by an edge if there exists a reaction that transfers at least one carbon atom from one metabolite to the other , the small world property is lost completely [8] . The concepts of flux balance analysis [9] , elementary flux modes [10] or extreme pathways [11] all aim at characterizing the possible flux distributions through the biochemical reactions when certain external metabolites are either provided by the environment or can be released into extracellular medium . Such an approach is well suited for the investigation of metabolic networks of selected organisms for which the fluxes of metabolites over the cellular membrane have been well characterized , so that it is clear which biochemical compounds have to be considered as external . Based on flux balance analysis , it has been shown that experimentally measured flux distributions in E . coli correspond well to distributions calculated under the premise that biomass production is maximized [12] . For the analysis of the network comprising the entirety of all biochemical reactions , it is impossible to decide which metabolites should be considered as external , since the role may differ greatly among different organisms or within cells of different tissues . A novel strategy for the analysis of large-scale metabolic network , that is less dependent on the knowledge which particular metabolites are external , has recently been proposed . The so-called method of network expansion [13] , [14] is based on the basic biochemical fact that only those reactions may take place which use the available substrates and that the products of these reactions may in turn be utilized by other reactions . With a number of given substrates ( the seed ) , a series of metabolic networks is constructed , where in each step the network is expanded by those reactions that utilize only the seed and those metabolites which are products of reactions incorporated in previous steps . The set of metabolites within the final network is called the scope of the seed and , by construction , comprises all substances that the network may produce when only the seed compounds are available as external resources . The scope describes the biosynthetic potential carried by the seed compounds and thus in a natural way links structural and functional properties of metabolic networks . In the present work , we aim at elucidating the global organization of functional aspects of metabolism by comparing the biosynthetic potentials of the different metabolites . For this , we extend studies carried out by us previously [15] . There , we observed that many compounds exhibit very similar potentials and introduced the notion of a consensus scope , characterizing the biosynthetic potential of a large group of metabolites . Whereas in our previous studies [15] we focused on the technical aspects and compared different dimensionality reduction methods , we concentrate in this work on the generalization of our results and in particular on their interpretation in a biological and evolutionary context . We find that many compounds can be grouped into biologically meaningful clusters , displaying a typical biosynthetic potential . We demonstrate that these typical potentials also characterize the combined potential of sets of metabolites . Furthermore , we observe that a similar biosynthetic potential of metabolites can often be connected with common chemical properties . However , in some cases chemically similar substances may exhibit dramatically different biosynthetic potentials and , moreover , clearly distinct biological functions may be assigned to such metabolites . The paper is organized as follows: The Results section consists of three parts in which we describe i ) the results from the hierarchical clustering as well as the construction of consensus scopes , ii ) the chemical properties of compounds belonging to the same cluster , and the hierarchical organization of the biosynthetic potentials , and iii ) the generalization to combined biosynthetic potentials of sets of metabolites . For readability , some results and definitions from [15] have been included in the first two parts of that section . In the Discussion section , our results are discussed . And finally , in the Methods section , details about the applied calculations are provided .
The aim of this work is to identify organizational principles in the metabolic network which is spanned by the entirety of biochemical reactions . For our analysis , we have retrieved enzymatic reactions from over 200 organisms from the KEGG database [1] . After curation of this information ( see Methods ) , the network consists of 4811 reactions involving a total of 4104 metabolites . We characterize all biochemical compounds by their biosynthetic potential . The extreme variation in consensus scope sizes raises the question whether they may be ordered by increasing biosynthetic potential . In fact , some consensus scopes are contained in others , some are mutually disjoint and others partially overlap . We schematically visualize the pairwise overlaps in Figure 1 . The figure shows that the immensely complex metabolic network displays an intricate hierarchical organization with respect to the biosynthetic potentials of the participating compounds . In the following , we will analyze similarities and differences in the chemical structure of metabolites belonging to the same cluster and particularly address the question whether the identified hierarchy may be explained by chemical structure alone or whether the biological role of metabolites or clusters of metabolites is also reflected in the metabolic organization . The largest consensus scope is formed by the four compounds in cluster XIII . It is identical to the scope of adenylyl sulfate ( APS ) and contains as subsets all other consensus scopes except those of clusters VII and XI . The second largest consensus scope is reached by metabolites of cluster III . Eight of the remaining consensus scopes are subsets thereof . This cluster contains organic compounds consisting of heterocyclic bases , sugars and phosphate groups , for example nucleotides , deoxynucleotides ( except those with thymine as base ) , nucleotide sugars , coenzymes except coenzyme A , and second messengers such as cAMP and other nucleotide derivatives . Many compounds contained in the cluster , such as ATP or NADH , are involved in energy metabolism . They are necessary for typical metabolic reactions , such as phosphate group transfer and redox reactions . The consensus scope of cluster III is identical to the scope of ATP . Cluster VI has the largest consensus scope completely contained in the scope of ATP . The cluster consists predominantly of those deoxynucleotides and deoxynucleotide sugars with thymine as their base . Apparently , their biosynthetic potential is smaller than that of other deoxynucleotides . This is surprising in view of the fact that dUTP , a member of cluster III , and dTTP , a member of cluster VI , show very similar chemical structures . However , even though dTTP is slightly more complex than dUTP because it possesses an additional methyl group , its biosynthetic capacity is much lower . While 1549 compounds may be synthesized from dUTP , dTTP allows for the production of only 305 compounds . This finding demonstrates that the chemical complexity of a biochemical compound is not the only determinant for the biosynthetic potential it carries . The consensus scope of cluster IV , consisting mainly of sugar phosphates , is completely contained in the consensus scope of cluster VI . The reduced biosynthetic potential is easily explained by the fact that sugar phosphates appear as chemical subunits in larger metabolites contained in clusters III and VI . However , sugar phosphates do not contain nitrogen , therefore , from them alone , e . g . nucleotides cannot be produced . Sugars form cluster V . Obviously , since the phosphate group is not available , their biosynthetic potential is even smaller , and consequently the consensus scope is completely contained in the consensus scope of cluster IV . Most other inclusion relations can also be explained by the presence or absence of characteristic chemical groups . Cluster II consists of organic acids not containing nitrogen . Its consensus scope , identical to the scope of pyruvate , is completely contained in that of cluster VI , but only shows a small overlap with that of cluster IV . It completely contains the consensus scopes of clusters VIII and XII . The composition of cluster VIII is rather diverse , ranging from small molecules such as glyoxylate to relatively large secondary metabolites including polyketides and flavonoids . A common property of these metabolites is that they can be oxidized to CO , CO2 or small carboxylic acids . These products also form the small consensus scope ( size 12 ) of the cluster . Metabolites within cluster XII share the common feature that they contain an aromatic 6-carbon ring . Its small consensus scope ( size 14 ) is almost identical to the cluster itself . Interestingly , there are two clusters ( VII and XI ) , whose consensus scopes do not overlap with other consensus scopes . Metabolites within cluster VII are all derived from 20-carbon polyunsaturated essential fatty acids , known as eicosanoids . These are highly specialized compounds functioning as signaling molecules in mammals during inflammation and immune response [17] . All metabolites in the cluster possess identical scopes ( cluster radius zero , see Table 1 ) , indicating that only a very special group of chemicals can be produced from them and conversely , those chemicals can exclusively be produced from eicosanoids . It is intriguing that structural considerations alone reveal such a clear separation of this cluster from the rest of the metabolism , reflecting the specialized role of eicosanoid metabolism . Cluster XI represents a group of nitrogen heterocyclic compounds with the common feature that all contain an indol group . All of these compounds are involved in the indole and ipecac alkaloid biosynthesis pathways . Again , it is striking that the purely structural approach reveals a separate cluster , consisting of metabolites that play a highly specialized role in metabolism . Similarly to the eicosanoids mentioned above , indole alkaloids function as signaling molecules; however , they are predominantly present in plants . In Figure 2 , the hierarchical ordering of the consensus scopes is displayed in a tree form . The boxes contain a cluster representative ( a compound with a scope identical to the consensus scope ) , the cluster label and the consensus scope size , as well as the chemical elements present in most metabolites of the corresponding cluster . In the drawing , clusters with a large biosynthetic potential are positioned above clusters with a lower biosynthetic potential . A line between two clusters is drawn if the consensus scope of the cluster positioned below is a subset of the consensus scope of the cluster positioned above . For clarity , indirect inclusions are not depicted ( although the scope of glucose is naturally contained in that of APS , a line has not been drawn ) . Also partial overlaps of consensus scopes are not depicted . Because the consensus scopes of clusters VII and XI are disjoint from all others , they are represented by isolated nodes . Interestingly , many inclusion relations can be associated with a difference in the chemical elements within the metabolites . For example , the ATP consensus scope ( cluster III ) is completely contained in the consensus scope of APS ( cluster XIII ) . ATP and all other metabolites of cluster III contain the elements C , H , O , N , and P . The four metabolites in the APS cluster contain additionally sulfur . This observation indicates that the chemical complexity plays an important role in the determination of the biosynthetic potential of a biochemical compound . However , there are examples in which metabolites possess a similar chemical composition and structure , but the corresponding consensus scopes differ greatly . For instance cluster IX , containing Acetyl-CoA , has a consensus scope being a small subset of the consensus scope of cluster XIII , containing APS . The members of both clusters are , however , composed of the same chemical elements ( C , H , O , N , P , S ) . Thus , the obtained hierarchy of the metabolites according to their biosynthetic potential is not only determined by their chemical complexity . So far , we have determined a hierarchy based on the biosynthetic potentials of single substances . However , a direct biological interpretation is hindered by the fact that it is unrealistic to assume that an organism will be provided with exactly one external substance . Usually , several nutrients are available and the exact composition may vary greatly for different organisms and different environments . To improve the biological significance of the developed concept , it is therefore of relevance to study the biosynthetic potentials of combinations of metabolites . Since a systematic analysis of seeds of a larger size is not feasible , we perform a Monte Carlo simulation and calculate the scopes for a large number of seeds consisting of a varying number of randomly chosen metabolites . We call the biosynthetic potential of a seed containing multiple compounds the combined biosynthetic potential . The Monte Carlo approach is similar to that followed in [18] . There the authors also calculated a large number of combined scopes for randomly selected seeds . They studied the size distribution of the scopes and in particular the increase in scope size when systematically central metabolites such as ATP , NADH , Coenzyme-A or oxygen were added to the seed . Here , we address the question whether the identified combined biosynthetic potentials can unambiguously be assigned to the determined consensus scopes , thus confirming that the revealed hierarchical ordering is of a general nature . For each seed size between 2 and 20 , we generate 1000 random seeds and calculate the corresponding scopes . Based on the distance measure ( Equation 1 ) , we identify for each scope the most similar consensus scope and denote the similarity by d0 . To assess the quality of the assignment to the closest consensus scope , we also identify the second nearest consensus scope and denote the distance by d1 . The ratio α = d0/d1 quantifies the uncertainty of the assignment , with small values α≪1 reflecting unambiguous assignments and α≈1 indicating a large uncertainty , because in such a case the scope is equally similar to at least two consensus scopes . The average value α is plotted against the number of metabolites in the seed in Figure 3 ( black squares ) . It can be seen that the assignment to a cluster is more reliable for larger seed sizes . This is not surprising since larger seeds tend to exhibit a larger biosynthetic potential and , as is the case for the potentials of single metabolites , small scopes cannot reliably be grouped into clusters . Consistent with the choice of parameters in the hierarchical clustering process , in which we merge two clusters if they exhibit a distance of less than 0 . 2 , we only assign the scopes of multiple-compound seeds to a cluster if d0<0 . 2 . As expected from the decreasing uncertainties of cluster assignment , the percentage of assigned clusters increases strongly with increasing seed size ( red circles in Figure 3 ) . For a seed size of two , less than 40% of the scopes are assigned to a cluster . For a seed size of 20 , almost all scopes are unambiguously assigned to one of the thirteen clusters determined above . To analyze which particular consensus scopes can be reached from a combination of metabolites , we plot in Figure 4 the fraction of scopes that are assigned to a particular consensus scope in dependence of the seed size . Shown are the values for the five clusters with the largest consensus scopes ( XIII−APS , III−ATP , I−L-Glutamate , VI−dTTP , IX−Acetyl-CoA ) , all other clusters are assigned with negligible frequency . The frequency of assignment to the largest consensus scope increases strongly with increasing seed size . This is expected because the addition of new metabolites to the seed may only increase the biosynthetic potential , so that a randomly chosen large set of metabolites is more likely to display the full potential of metabolites from cluster XIII than a small set . However , the numbers provide further insight into the structural design of metabolism . For 20 randomly selected compounds , the chance that one of them belongs to the four compounds forming cluster XIII is still below 2% . On the other hand , more than half of the scopes for this seed size get assigned to the corresponding consensus scope . This indicates that the particular , chemically very rich , compounds from cluster XIII are not necessary to obtain the full biosynthetic potential characterized by the scope of APS . Instead , the same potential is contained in many combinations of smaller substances . This result generalizes the observation made in [14] that the scope of APS may also be reached if , for instance , CO2 , NH3 , phosphate , sulfate , water and oxygen are used as seed . The frequency of assignment to the second largest consensus scope also increases with increasing seed size , however , it does not change considerably for seed sizes larger than 10 . For the assignment to the consensus scope of cluster I , and in fact for the other clusters as well , the frequency also increases for small seed sizes but tends to decrease when the seed sizes become large . The reason is that for larger sets of seed compounds it becomes increasingly difficult to find such combinations which do not exhibit a large biosynthetic potential . Therefore , for larger seed sizes , the frequency of assignment is shifted towards the larger consensus scopes . These investigations demonstrate that the notion of consensus scopes is also meaningful in the context of combined biosynthetic potentials . Moreover , the larger the set of metabolites for which the combined biosynthetic potential is considered , the more combinations are assigned to one cluster . In fact , larger sets of metabolites tend to display a potential characterized by the largest consensus scopes . For large seed sizes ( σ>20 ) , this is true for more than half of all combinations . As a consequence , large sets of metabolites may not be distinguished by their combined potential , reflecting the fact that central metabolites such as nucleotide phosphates , amino acids and coenzymes may be built flexibly from many different resource combinations . We conclude that the hierarchical ordering of biosynthetic potentials which was determined for single metabolites is of a general nature and its significance even increases when combined biosynthetic potentials are studied .
By calculation of the overlap of the consensus scopes we derived a hierarchy of metabolites , providing a novel view on the global organization of metabolism . Some compounds ( as those of cluster III ) possess a high chemical complexity in the sense that they are composed of many distinct functional groups that are essential for the synthesis of a large number of chemical compounds . Other metabolites possess a less complex chemical structure , and therefore the cellular metabolism can produce only a small number of compounds from them . For some of these cases , the lower biosynthetic potential can easily be explained by the presence or absence of chemical groups . For example , it is intuitive that from sugars ( cluster V ) less can be synthesized than from sugar phosphates ( cluster IV ) . In other cases , however , it is not immediately evident from the chemical structure alone why the biosynthetic potential of some metabolites is smaller than for others . For example , from deoxynucleotides with thymine as base ( cluster VI ) , the metabolism can produce only a small subset from those which can be produced from deoxynucleotides with other bases ( cluster III ) . This is surprising in view of the fact that thymine is structurally similar to , for example , uracil . In fact , the chemical structure of thymin is even slightly more complex since it possesses an additional methyl group . The different scope sizes of dTTP and dUTP result from the fact that dTTP is included in the scope of dUTP , whereas the opposite is not true . Because in our analysis all reactions are considered to be reversible , this asymmetry does not arise from thermodynamic constraints , but is rather an intrinsic structural property of the metabolic network . The issue of interconvertibility is discussed in detail in [14] . We hypothesize that the differences in scope size are a consequence of different biological functions of these compounds . In particular , synthesis of DNA in the presence of high levels of dUTP promotes incorporation of uracil into DNA , since polymerases cannot discriminate between the deoxynucleotides dUTP and dTTP [19] . Uracil misincorporation compromises the stability of DNA , resulting in DNA damage and cytotoxicity [20] . In normal cells , accumulation of dUTP by phosphorylation from dUMP via dUDP is avoided by rapid reductive methylation of dUMP to dTMP . Because a direct conversion of dTMP to dUMP is not possible , the concentration of dUMP is kept at a low level . Our hypothesis concerning the asymmetry of interconvertibility of dTTP and dUTP is supported by experimental findings , in which the amoebozoa Physarum Polycephalum was grown on 14C-labeled nucleosides [21] . There , the authors observed that the 14C from thymidine only enters dTTP , whereas the 14C from other nucleosides was found in other ribonucleoside and deoxyribonucleoside triphosphates . In our analysis this is reflected by the presence of dTTP in the scopes of dATP , dGTP , dUTP , and dCTP , whereas the scope of dTTP does not contain any of these nucleoside triphosphates . This example demonstrates that chemically very similar substances may differ significantly in their biosynthetic potentials and that these differences may only be explained by consideration of the biological functions of the metabolites . Apparently , the complexity of a chemical substance may be described in two different ways . The structural complexity of a metabolite is determined by the types and numbers of chemical groups and elements and the bonds between them . Another determinant is the oxidation number of a metabolite since oxidation/reduction reactions play an important role in metabolism . In contrast , the functional complexity of a metabolite is determined by its biological role within cellular metabolism . It may depend on the availability of appropriate enzymes , the subcellular and tissue-level localization of metabolites and enzymes , and the kinetics and thermodynamics of biochemical reactions . In this work , we invoked the concept of scopes , characterizing the biosynthetic potential of a metabolite , to provide a quantification for the functional complexity . Our results have indicated that both types of complexity are in many cases correlated , however , this correlation is not strict and we identified chemically very similar metabolites exhibiting a drastically different functional complexity . It will be interesting to study the relation between structural and functional complexity of metabolites in further detail . The analysis of the consensus scopes revealed a hierarchical setting in which some consensus scopes are contained in others . Two clusters are not included in this hierarchy , namely clusters VII and XI , which display a disjoint consensus scope from all others . Both clusters are composed of compounds with particular chemical features exhibiting a specific biological function . The compounds in cluster VII ( eicosanoids ) , function as autocrine and paracrine mediators . Compounds in cluster XI , characterized by an indole group , act as signaling molecules in plants . As a general tendency , we found that compounds belonging to clusters with a large consensus scopes are primary metabolites , i . e . metabolites necessary for essential cellular functions and present in the metabolism of most organisms . In contrast , compounds within clusters with a small biosynthetic potential are often secondary metabolites , for example alkaloids , terpenoids or fatty acids . Such metabolites are species specific and are not directly involved in essential cellular processes . The chemical structure of many secondary metabolites is more complex than that of primary metabolites . Our investigations revealed that secondary metabolites within one cluster are structurally very similar and can be obtained from other members of the cluster through small chemical modifications such as methylation , hydroxylation or isomerization . In contrast , primary metabolites clustered in the same group often display large chemical differences . Despite this , the structure of the metabolic network ensures that such metabolites are still interconvertible , however , the pathway leading from one substrate to a product may require a large number of enzymatic reactions . In the first part of this work , we focused on the calculation of single scopes , i . e . sets of metabolites which can be produced if exactly one metabolite plus water and oxygen is available . We then asked whether our results are of a general nature and still hold true for combined scopes ( sets of metabolites which can be produced from a larger number of initial substrates , i . e . a larger seed size ) . We applied a Monte-Carlo approach , randomly selecting seeds of varying sizes , and measured the distance between the scopes of the multiple-compound seeds to the consensus scopes of the clusters I-XIII . We found that the larger the seed sizes , the more combined biosynthetic potentials can be assigned to the 13 clusters previously identified . Since larger seeds tend to exhibit a larger biosynthetic potential , the frequency of assignment to the largest consensus scope ( cluster XIII ) increases strongly with increasing seed size . Thus , many combinations of smaller substances can exhibit the same biosynthetic potential as the chemically complex compounds from cluster XIII . This hints at a redundancy principle in the design of the global metabolic network . Our findings demonstrate that the hierarchical ordering of biosynthetic potentials , originally determined for single metabolites , is of a general nature , and also meaningful for larger sets of nutrient seeds . Another study based on the Monte-Carlo approach [18] has shown that the sizes of the resulting scopes are concentrated in four disjoint regions , the largest scopes being produced only if oxygen is contained in the initial set of substrates . In the results presented above , we have always assumed that oxygen is present . We have tested whether our findings are dependent on this assumption and repeated our calculations for anaerobic conditions ( see Methods , Effect of Oxygen; and Table S3 and Dataset S2 in the supporting information ) . In agreement with the results of Raymond and Segrè [18] , we also observe that many metabolites possess a strongly reduced biosynthetic potential under anaerobic conditions , demonstrating that they can only deploy their full potential when oxygen is available . As a consequence , the clusters tend to be of a smaller size , but pairs of corresponding clusters can clearly be identified . In contrast , the consensus scopes , characterizing the typical biosynthetic potentials of the clusters , remain almost completely unchanged . It is remarkable that , while oxygen naturally has a strong impact on metabolism and possible synthesis routes , its absence or presence is not decisive for the hierarchical organization of the global metabolic network comprising enzymatic reactions from aerobic as well as anaerobic organisms . This fact supports the hypothesis that the computationally derived hierarchy is indeed of general nature and does not depend on specific assumptions . Interestingly , clusters I , II and VIII which exhibit a considerable reduction under anaerobic conditions , have consensus scopes containing metabolites with a higher oxidation state than compared to consensus scopes of other clusters . Thus , many compounds contained in these consensus scopes are obtained from oxidation reactions , which in many cases require oxygen . The fact that many oxidation reactions take place using NAD+ or FAD as oxidant provides an explanation that the consensus scopes can still be reached , however by a smaller number of compounds . We expect that a more detailed investigation of the effect of the average oxidation number within clusters and their consensus scopes on the cluster reduction under anaerobic conditions will provide new insight into the role of oxygen and alternative oxidants in cellular metabolism . Summarizing , by grouping metabolites with respect to their biosynthetic potential , the huge variability of biochemical compounds involved in metabolism can be represented in a relatively concise form . Apparently , there exist only a small number of typical sets of metabolites ( the consensus scopes ) which can be produced from one single precursor . These sets display a hierarchy which in some cases can be explained by the chemical groups contained in the precursors . In other cases , the underlying reasons for the hierarchical structuring have their origin in different biological functions of the compounds . Our results have been obtained from a computational study which is based on a database with necessarily incomplete and constantly changing content . Moreover , there is some degree of arbitrariness in the curation process used to derive the metabolic network . Despite these uncertainties , we could demonstrate that our results are only marginally different when based on database releases between which over two years have passed ( see Methods , Robustness Against Changes in Network Structure; and Dataset S2 in the supporting information ) . More importantly , the derived hierarchical structure proved stable . We thus assume that the hierarchy is indeed an intrinsic characteristic of the metabolic network itself . The catalyzing enzymes are a product of a long evolutionary process which was governed by selection and mutation principles . In total , they catalyze only a small fraction of all theoretically possible chemical transformations . The nature of the evolutionary driving forces which resulted in the selection of the particular set of enzymatic reactions found in contemporary organisms remains subject to speculation . Our analyses show that the network is extremely flexible in its resource requirements , exhibited by the fact that central metabolites ( e . g . ATP , NADH , Coenzyme-A , amino acids ) may be synthesized from many different combinations of substrates . Nevertheless , certain metabolic routes involving chemically similar compounds are to some degree separated . This indicates that for the chemically feasible reactions that could provide the link , enzymes have not evolved . Of course , it is possible that such enzymes do exist but have not been discovered yet and are therefore not yet included in the KEGG database . However , database versions almost three years apart did show the same cluster separations . With the present rate of increase of biochemical knowledge , this is a hint that such enzymes do indeed not exist in contemporary organisms or are extremely rare . Assuming that the observed separation of metabolism is a real feature of the contemporary metabolic network , what could have been a selective advantage that hindered the evolution of enzymes for connecting reactions ? Of course , under physiological conditions many similar chemicals , also such displaying an identical scope , may exhibit different biological functions . This is possible by mechanisms such as allosteric regulation and gene regulation to adjust production rates to the present demand , so a structural separation of metabolic routes does not seem necessary from a present-day view . However , it can be assumed that early during metabolic evolution , primitive precursors of contemporary enzymes have catalyzed biological reactions . Common mechanistical themes in diverse enzyme superfamilies [22] suggest that early enzymes displayed a significantly lower substrate specificity , and the modular structure separating catalyzing from regulatory domains in proteins [23] allows to speculate that domains for functional control have been less elaborated or non-existent in early enzymes . In such a scenario , a separation of the metabolic network on the structural level does indeed make sense , since in this way certain chemical conversion routes are principally excluded , providing a selective advantage by avoiding undesired interactions . It is remarkable that the clearest separation involves nucleotide phosphates , which , as prerequisite for the genetic code , assumably have acquired a central role early during metabolic evolution . Moreover , the particularly similar nucleotides with strikingly different potentials , uracil and thymin , are exactly those structural elements which distinguish the related , information coding macromolecules DNA and RNA . We therefore hypothesize that the observed separation of clusters of similar compound is a relict of the early phase of metabolic evolution , when regulatory mechanisms had not yet evolved to their present-day elaboration . Under such conditions , it might have provided a selective advantage to fundamentally separate metabolic routes , which is most drastically achieved by a separation on a structural level . One aspect poorly understood is how a large chemical diversity of more than 200 . 000 secondary species-specific metabolites has evolved from primary metabolic pathways [24] . The high plasticity of secondary metabolism involving enzymes with broad substrate specificity might have enabled organisms to adapt easily to environmental changes . A model has been proposed to explain the increased generation of chemical diversity after a mutational event assuming a broad substrate specificity of the enzymes [25] . In this model , a substrate A is sequentially converted by a series of distinct enzymes into other compounds B , C , D and E . A mutational event could give rise to a new variant of the organism , producing a compound A' that is structurally similar to A . The same enzymes acting on the pathway A→E could generate the new compounds B' , C' , D' and E' . Our results are consistent with such a model; secondary metabolites are confined to small clusters and the majority of these metabolites are interconvertible , being also found in the consensus scopes . Likely , most enzymes that might have catalyzed the transition from B to B' have not evolved , which explains the disjoint consensus scopes of these clusters . We expect that a clustering analysis of organism-specific networks may bring some insight in our understanding of the evolutionary transition from primary to secondary metabolism . We hypothesize that in early-evolved organisms some secondary metabolites or related compounds will be found in larger clusters functioning as primary metabolites . While the presence or absence of oxygen is not influencing the hierarchical organization of the global network discussed here , we expect that the effect of oxygen on organism-specific hierarchies is more pronounced . Metabolism under aerobic and anaerobic conditions may differ considerably between organisms . For example , earlier organisms , which appeared before oxygen was present at a high concentration in the atmosphere , prefer fermentation ( anaerobic ) to oxidative phosphorylation ( aerobic ) . We expect that i ) the organism specific hierarchies will reflect the growth conditions ( aerobic versus anaerobic ) and ii ) the effect of oxygen will stronger influence the structure of the metabolism for aerobic than for anaerobic organisms . From such comparative studies we expect to gain further clues about which underlying principles may have guided the evolution of metabolic networks .
We have retrieved the global metabolic network from the KEGG database in the following way . From the LIGAND subdivision , the complete list of reactions has been imported . The reactions have been checked for consistency and those were rejected which showed an erroneous stoichiometry , by which we mean that some atomic species occurred in different numbers on both sides of the reaction . The inclusion of such erroneous reactions could result in absurd events such as the creation of chemical elements or groups . We identified compounds possessing ambiguous structure information , such as chains of chemical groups of unspecified length ( e . g . Ubiquinol , KEGGID: C00390 , C14H20O4[C5H8]n ) or compounds with unspecified residues ( e . g . Amino acid , KEGGID: C00045 , C2H4NO2R ) . We rejected all reactions involving such metabolites . Further , we did not include reactions involved in glycan synthesis because the focus of our investigation lies on the metabolism of small chemical species , which also does not include the formation of complex structures such as proteins or RNA and DNA molecules . This curation process has been applied to two different database releases , one dating back to January 2005 , and a recent version from December 2007 . The older release resulted in a network comprising a set of 4811 enzymatic reactions connecting 4104 biochemical compounds . Results presented in the main text have been obtained for this network . To ensure that our results are not critically influenced due to changes in the database , calculations have been repeated for the network derived from the recent database release , comprising 5529 reactions with 4668 reactants ( see below ) . We have also tested whether the curation process applied to extract the metabolic network from the database is critical for our results . For this purpose , we built two more networks from the recent database version , one with relaxed and one with stringent criteria . For the former , all 6003 reactions were included , even those showing erroneous stoichiometry . For the latter we demanded absolute correctness , risking the exclusion of valid reactions , leaving 4257 reactions in the network . For all networks , the complete reaction lists are provided in the supporting information ( Dataset S2 ) . It is possible that by removing reactions during curation the resulting network contains parts not connected to the rest of the network . This was indeed observed; however , in most cases this concerns single or groups of a small number of reactions . We did not put any effort in deriving a fully connected network , since small disconnected components are unproblematic for the kind of analysis presented here . In principle , the KEGG database also provides information on the reversibility of biochemical reactions . This information is contained in XML files which define the organism-specific pathways . We found , however , that for many reactions ( over 200 ) , this information is ambiguous . Further , the direction in which a reaction actually proceeds under physiological conditions is strongly dependent on the metabolite concentrations and therefore may vary for different organisms , tissues or environmental conditions . To account for this and considering that in principle every enzymatic reaction may also proceed in the reverse direction , we have considered all reactions to be reversible . To assess the synthesizing capacities of a metabolic network when it is provided with a particular substrate , we apply the method of network expansion which is in detail described in [14] . We give here a short outline of the algorithm: The resulting set M is called the scope of compound X . We use the scope as a measure for the biosynthetic potential carried by metabolite X . Naturally , this algorithm can be initialized with any combination of seed compounds . For the analysis of combined biosynthetic potentials , we randomly select several metabolites as seeds and apply the described algorithm . We apply a hierarchical clustering with the distance measure ( Equation 1 ) . The advantage of hierarchical clustering methods is that they provide information about clustering of the data at all scales , from fine to coarse . The disadvantage is the computation time which scales with O ( n2 ) , because the distance between every pair of data has to be computed . We choose a nearest neighbor group-average clustering algorithm [26] . Nearest neighbor clustering is a bottom up clustering method where iteratively the elements or clusters with the smallest distance are joined . Group-averaging refers to the method of defining the distance between two clusters as the average over all distances between pairs of the corresponding cluster elements . We denote the elements ( scopes ) of cluster i by and of cluster j by , and the sizes of clusters i and j by si and sj respectively . The distance dij between the two clusters i and j is then defined as the group-average of the distances between all elements and , for k = 1 , 2 , …si and l = 1 , 2 , …sj , ( 2 ) where the average is over all k and l . The algorithm is implemented as follows: The result obtained in this procedure is a clustering of the data on various scales . At the first iterations only very similar elements obtain the same cluster label and the clustering is very fine . Towards the end elements or clusters with large distances are joint , resulting in a coarse clustering with a smaller number of clusters . Figure 5A shows the increasing distances at which elements or clusters are merged at subsequent iterations of the nearest neighbors clustering . In the beginning elements are clustered at very small distances , in fact there is quite a large number of identical scopes . In the next region the distances increase linearly to the maximum value . The following iterations then join elements that do not have even a single common substance in their scopes . All the very small scopes of compounds with zero synthesizing capacity are assigned to clusters in this last phase . After the clustering is finished , the next step is to decide which scale of the clustering is reasonable for analysis . Our choice is based on the distances themselves on one hand , and on the robustness of the clusters on the other . Naturally , elements of the same cluster should be very similar , therefore the coarse scale , where clusters are joined although the distance between them is large ( let's say , dml>0 . 5 ) is not feasible . The robustness of the clusters is measured as the variation in the amount of larger clusters when varying the joining distances . In Figure 5B we plot the number of clusters of a given minimum size k for various k as a function of the distance at the current iteration of clustering . The length of a plateau in one of the curves then gives the robustness of the clustering for clusters of size larger than k . We are interested in larger clusters and therefore consider only the three lower curves . There is a smaller plateau for distance values between 0 . 1–0 . 2 and a long one between 0 . 3–0 . 7 . Because of the requirement that the distances should be rather small , we focus on the first plateau , and therefore choose the clustering level where elements are joint with a distance of at most 0 . 2 . Finally , we measure the quality of the clustering to assure that the elements within the same cluster are similar and the clusters well distinguishable . Generally , a clustering is considered good , if the distances of the elements within a cluster are small , and the distances between elements of distinct clusters large . To quantify the quality of the clustering we compute for every cluster ( I-XIII ) the distance between all scopes contained in the cluster to the consensus scope . The maximum of these distances can be regarded as a cluster radius , denoted by . Furthermore , we compute the distance between all scopes in the cluster to the second nearest consensus scope . The minimum of these , , provides a measure of the cluster separation . Since the distance is based on the Jaccard coefficient , and ≤1 . A cluster is well defined if is small and large . Figure 6 shows that for all clusters is much smaller than , except for cluster II . This can be regarded as a consequence of the similarity between the clusters II and IX . Cluster II is fully embedded in cluster IX , while their consensus scopes differ only by about 25% . To obtain a single parameter quantifying the uncertainty of the assignment of metabolites to a cluster , we compute the ratio . The quality of the clustering is good for β≪1 . For β≈1 , the assignment is uncertain , as in this case the scope is equally similar to at least two consensus scopes . The results presented in this work have been obtained for a metabolic network reconstructed from a KEGG database version dating back to the year 2005 . Due to our rapidly increasing biochemical knowledge , it lies in the nature of databases as KEGG that their content is constantly changing . It is therefore a crucial question whether our results , and therefore our biological conclusions presented here , are still valid if new reactions enter the database or erroneous reactions are removed . An indication that the results should not be drastically influenced is given by the robustness studies of single scopes in [14] . Here it was shown that deletions of single reactions from the network alter the scopes only in rare cases considerably . Since it is unclear how the clusters and the consensus scopes are affected when the network structure is altered by several hundreds of reactions , we have repeated our calculations for a network derived using a database version from December 2007 ( for the detailed results , see Dataset S2 in the supporting information ) . For the new network we compared the obtained clusters , consensus scopes and resulting hierarchies to the results above . For all the clusters I−XIII , corresponding clusters can easily be identified for the new network ( see Tables S1 and S2 , and Figure S1 ) . The similarity between the clusters I−XIII and the corresponding new clusters is very high . The Jaccard coefficients of pairs of corresponding clusters have in most cases values above 0 . 7 . The corresponding consensus scopes are even more similar , Jaccard coefficients of pairs of corresponding consensus scopes have values of 0 . 8 or higher . As expected from the increase in network size , the cluster sizes as well as the consensus scopes have a tendency to increase in size . Apart from identifying corresponding clusters , we also obtained new clusters from the recent database version . Two clusters are formed by carotenes and oxylipins derived from linolenic acid , containing 18 and 10 metabolites , respectively . Their consensus scopes ( sizes 28 and 20 , respectively ) do not overlap with other consensus scopes . Again we see that compounds contained in isolated clusters are secondary metabolites . The effect of the specific network curation strategy has been tested with two more networks derived from the recent KEGG database version ( see above ) , termed relaxed and stringent . Not surprisingly , we found that for the relaxed network our results change considerably . Erroneous stoichiometries lead to absurd events , like the creation of new chemical elements . One example is that for this network the scopes of ATP and APS are identical even though APS possesses a sulfate group while ATP does not . Due to such obviously meaningless results , we do not consider the relaxed network further . However , such observations demonstrate how important the process of database curation is to derive consistent network models . For the stringent network the scope sizes–and therefore consensus scope sizes–are sometimes drastically reduced , however , most of the previously identified clusters are again present , and the obtained hierarchy is still structurally conserved ( see Figure S3 ) . The major exception is that cluster III , represented by ATP , splits into two clusters exhibiting consensus scopes identical to the scopes of ATP and UTP , respectively . Moreover , some of the smaller clusters disappear . We explain these changes by the fact that some important reactions are missing due to the very harsh criteria . The detailed results can be found in the supporting information ( Dataset S2 ) . We have performed our calculation assuming that oxygen is always available . This assumption is motivated by the fact that atmospheric oxygen , a highly reactive chemical , has been abundant for approximately 2 . 8 billion years and therefore the metabolic network that we see today has to a large extent evolved under aerobic conditions . However , there is a certain arbitrariness in our assumption , since for similar reasons other compounds , such as CO2 , could be included in the seed . To verify whether our calculations critically depend on the availability of oxygen , we repeated the cluster analysis based on biosynthetic potentials of all metabolites under the premise that only water is additionally available . Corresponding clusters with a high overlap can easily be identified between the results obtained with and without oxygen . The results are summarized in Tables S2 and S3 , details are found in Dataset S2 and Figures S2 and S4 . Some clusters remain completely unchanged ( IV , X , and XIII ) , whereas others are slightly reduced in size . Interestingly , while absence of oxygen does alter the cluster composition , most consensus scopes remain completely identical . A significant change in size is observed for cluster VII , which is almost halved in the absence of oxygen , while the corresponding consensus scope is unaffected . A closer inspection reveals that this cluster is in fact split into two subclusters , a smaller one formed by prostaglandins and thromboxanes whose consensus scope is completely contained in that of the larger subcluster formed by lipid hydroperoxides , leukotrienes and oxilins . The reduced biosynthetic capacity of prostaglandins and thromboxanes is due to an oxidation reaction of prostaglandin H2 yielding prostaglandin G2 , which does not occur under anaerobic conditions . The most dramatic change is observed for cluster VIII , which collapses completely . The metabolites within this cluster are very diverse but possess the common property that they can be oxidized to CO , CO2 or small carboxylic acids . The collapse of this cluster can be accounted to the fact that these oxidizing reactions cannot take place in the absence of oxygen , which also explains the strong reduction of the size of the corresponding consensus scope . | Life is based on the ability of cells to convert raw materials into complex chemicals like proteins or DNA . This ability is obtained through the interplay of a large number of enzymes , which are specialized proteins , each facilitating one specific chemical transformation . Since the products of one reaction can again be substrates for others , the entirety of all reactions forms a large and complex network in which important substances can be produced from many different combinations of simple chemicals and through a variety of pathways . The aim of our work is to gain understanding of the structural design of these networks and the evolutionary principles shaping them . We propose a computational strategy which allows us to pinpoint characteristic structural and functional properties distinguishing networks characterizing living processes from those that may occur in inanimate matter . Our approach reveals an intricate and unexpected hierarchical organization of the network , and gives rise to new hypotheses regarding the evolutionary origins of metabolism . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
]
| [
"computational",
"biology/metabolic",
"networks"
]
| 2008 | Biosynthetic Potentials of Metabolites and Their Hierarchical Organization |
The type VI secretion system ( T6SS ) has emerged as an important mediator of interbacterial interactions . A T6SS from Pseudomonas aeruginosa targets at least three effector proteins , type VI secretion exported 1–3 ( Tse1–3 ) , to recipient Gram-negative cells . The Tse2 protein is a cytoplasmic effector that acts as a potent inhibitor of target cell proliferation , thus providing a pronounced fitness advantage for P . aeruginosa donor cells . P . aeruginosa utilizes a dedicated immunity protein , type VI secretion immunity 2 ( Tsi2 ) , to protect against endogenous and intercellularly-transferred Tse2 . Here we show that Tse2 delivered by the T6SS efficiently induces quiescence , not death , within recipient cells . We demonstrate that despite direct interaction of Tsi2 and Tse2 in the cytoplasm , Tsi2 is dispensable for targeting the toxin to the secretory apparatus . To gain insights into the molecular basis of Tse2 immunity , we solved the 1 . 00 Å X-ray crystal structure of Tsi2 . The structure shows that Tsi2 assembles as a dimer that does not resemble previously characterized immunity or antitoxin proteins . A genetic screen for Tsi2 mutants deficient in Tse2 interaction revealed an acidic patch distal to the Tsi2 homodimer interface that mediates toxin interaction and immunity . Consistent with this finding , we observed that destabilization of the Tsi2 dimer does not impact Tse2 interaction . The molecular insights into Tsi2 structure and function garnered from this study shed light on the mechanisms of T6 effector secretion , and indicate that the Tse2–Tsi2 effector–immunity pair has features distinguishing it from previously characterized toxin–immunity and toxin–antitoxin systems .
The type VI secretion system ( T6SS ) is a multifaceted protein export pathway that is widely distributed in Gram-negative Proteobacteria [1] , [2] . With a minimal functional requirement for the products of at least 13 genes , this secretion machine rivals the complexity of the more extensively characterized type III and IV systems [3] , [4] . Among the conserved components of the T6SS are a AAA+ family ATPase , ClpV , two proteins with sequence similarity to the type IVB secretion proteins IcmF and DotU , TssM and TssL , and several proteins with sequence or structural similarity to non-filamentous phage proteins [5] . The latter group of proteins includes Haemolysin co-regulated protein ( Hcp ) and Valine glycine repeat protein G ( VgrG ) , which bear structural similarity to the tail protein of lambda phage ( gpV ) and the puncturing device of T4 phage ( gp27/gp5 ) , respectively [6]–[8] . Hcp and VgrG proteins are exported in a co-dependent fashion from nearly all T6SSs characterized to date . In combination , these observations have led to a prominent structure-function model in which the T6S apparatus resembles outward facing phage on the bacterial cell surface [9] . Early investigations of the T6SS focused on its role in modulating bacterial-host cell interactions . These efforts yielded information about the genetic requirements for T6S function and provided evidence that a subset of T6SSs influence pathogenesis by specifically mediating bacterial interactions with eukaryotic cells [10] . In addition to mediating host cell interactions , the T6SS has been shown to regulate gene expression and contribute to biofilm formation [11] , [12] . It is not currently understood how the system mediates such diverse phenomena , nor is it known in all cases whether the effects observed are a direct or indirect result of its function . Recently it has become clear that the T6SS plays a key role in mediating interactions between bacterial cells [2] . This was first observed in P . aeruginosa , where the Hcp secretion island I-encoded T6SS ( H1-T6SS ) was shown to target an effector protein , Tse2 ( type VI secretion exported 2 ) , to other P . aeruginosa cells [13] . Recipient cells lacking a Tse2-specific immunity protein , Tsi2 ( type VI secretion immunity 2 ) , were found to be at a competitive disadvantage relative to donor cells possessing Tse2 . Although the mechanism of action of Tse2 remains unknown , the fitness advantage bestowed by the protein requires a functional T6SS in the donor cell and close association of donor and recipient cells . The H1-T6SS exports at least two additional effector proteins , Tse1 and Tse3 [14] . These proteins are targeted by the T6SS to the periplasm of recipient cells where they degrade peptidoglycan and thereby provide a competitive fitness advantage for P . aeruginosa donor cells . P . aeruginosa protects its own cells from the action of these toxic proteins by synthesizing cognate periplasmic immunity proteins , Tsi1 and Tsi3 . Tsi2 differs from Tsi1 and Tsi3 in several respects . For instance , Tsi2 is an essential protein in P . aeruginosa , whereas Tsi1 and Tsi3 are dispensable for growth [14] . This reflects the differences in the subcellular sites of action of the associated cognate toxins . Because the T6S export mechanism avoids periplasmic intermediates , immunity proteins for periplasmically-targeted effectors ( Tse1 and Tse3 ) are required only for resisting intercellular self-intoxication . Conversely , Tse2 appears to act in the cytoplasm; therefore , in addition to resisting Tse2 delivered to the cytoplasm via intercellular self-intoxication , Tsi2 is essential for protecting against endogenous cytoplasmic intermediates of Tse2 . Owing to their localization in the same subcellular compartment , Tse2 and Tsi2 are able to complex prior to toxin export . In the case of Tse1–Tsi1 and Tse3–Tsi3 , the physical separation of the toxins ( cytoplasmic ) from their immunity proteins ( periplasm ) prevents such interactions . This likely imparts a unique requirement for Tse2 export – it must be recognized by the secretory apparatus in the context of a protein complex . Since Tsi2 is not exported by H1-T6SS , it must also be dissociated prior to or during the secretion of Tse2 . In this way , Tsi2 is analogous to specialized dedicated secretion chaperones involved in the export of effectors from other alternative secretion pathways such as the type III and IV systems [3] , [15] . Secretion chaperones are known to function critically both in stabilizing cognate effectors prior to export and in targeting effectors to the secretion apparatus . It is apparent that bacterial genomes possess an enormous diversity of toxin-immunity modules outside of T6S-associated Tse–Tsi pairs [16]–[18] . Perhaps the most abundant and thoroughly characterized of these are the toxin-antitoxin ( TA ) systems [19] , [20] . Growing evidence supports a general role for TA systems in resistance to stress and persister cell formation [21] . Type II TA systems consist of a cytoplasmic toxin that is maintained–under favorable conditions–in an inactive state by direct binding to a specific cognate antitoxin protein . Upon activation of cellular stress-response pathways , the antitoxin , which is typically less stable than the toxin , is rapidly degraded by cellular proteases including Lon ( Long Form Filament ) , allowing the toxin to act on its target ( s ) . Toxins vary in their mechanism , however most act as either ribosome-dependent or -independent mRNAses [22] , [23] . The properties of the Tse2–Tsi2 pair that make it unique among T6S effector–immunity proteins are the same as those that offer analogy to effector–chaperone and toxin–antitoxin systems . In this report , we sought to ascertain the degree of similarity between these systems by interrogating the structure and function of Tsi2 . Our results define properties of Tse2–Tsi2 that are shared with both TA and effector-chaperone systems , however we find that the Tse2–Tsi2 system is altogether functionally , structurally , and mechanistically distinct .
We have reported that P . aeruginosa donor cells capable of delivering Tse2 by an active H1-T6SS to P . aeruginosa recipient bacteria lacking tsi2 have a pronounced competitive fitness advantage [13] . However , absolute colony forming units ( CFU ) of competing bacteria were not determined in these experiments , which precluded defining whether Tse2 causes cell death or stasis in recipient cells when delivered by the T6SS . Lacking this information , the physiological role of Tsi2 – the subject of our current study – in the context of an interbacterial interaction was also not known . To investigate the role of Tsi2 in resisting T6S-dependent Tse2-based intoxication , we monitored changes in donor and recipient CFU during interbacterial competition experiments between P . aeruginosa strains . Both donor and recipient strains were generated in the P . aeruginosa ΔretS background . The deletion of retS relieves tight negative posttransciptional regulation of the H1-T6SS and reveals a robust T6S- and Tse2-dependent competitive fitness advantage between strains . Recipient strains bore an additional deletion of the tse2 tsi2 bicistron , which renders them sensitive to Tse2 . Both tse2 and tsi2 must be deleted in this strain , as the deletion of tsi2 alone is lethal in the presence of tse2 . Donor strains were distinguished from recipients by chromosomal lacZ expression from the neutral attB site . Interestingly , we found that while total CFU of the donor strain increased exponentially over the course of the competition experiment , CFU of the recipient remained constant ( Figure 1A ) . Consistent with our earlier findings , this inhibition of proliferation required tse2 in the donor and the absence of tsi2 in the recipient ( Figure 1B & 1C ) . We considered three explanations for our finding that the overall population of recipient cells lacking Tse2 immunity did not change during competition experiments against donor cells actively exporting Tse2 by the T6SS: 1 ) recipient cells are efficiently targeted ( approaching 100% ) and Tse2 is always bacteriostatic , 2 ) recipient cells are inefficiently targeted and Tse2 is bactericidal , and 3 ) recipient cells are efficiently targeted , but differentially affected by Tse2 ( unaffected , growth-inhibited , or killed ) . In the latter two scenarios , the balance between proliferation and death ( 2 ) , or between proliferation , non-proliferation , and death ( 3 ) , could produce the stable overall population of recipient bacteria observed . For either of these possibilities , we would expect to observe elevated cell death that is Tse2-dependent in competition experiments between a donor bacterium and a sensitive recipient . However , we found equivalent fractions of dead cells when a sensitive strain was competed against a donor strain capable of delivering Tse2 or one lacking Tse2 ( Figure 1D ) . From these data , we conclude that recipient cells are efficiently targeted by the T6SS , and that the function of Tsi2 is to protects cells from stasis induced by Tse2 . The substrates of many bacterial secretion pathways require dedicated chaperones for their export . We hypothesized that in addition to its immunity function , Tsi2 might also serve as a dedicated chaperone for Tse2 . Although our earlier work has shown that Tse2 and Tsi2 interact in P . aeruginosa , whether the proteins bind directly was not determined [13] . As a first step toward investigating the involvement of Tsi2 in Tse2 export , we probed for direct interaction between the proteins using an E . coli bacterial two-hybrid ( B2H ) assay [24] . In the system we employed , fusions of candidate interaction partners are made to a zinc-finger DNA-binding protein , Zif , and the ω subunit of RNA polymerase . Association of the fusion proteins promotes transcription of a lacZ reporter gene as described by Dove and colleagues [25] . The broadly toxic nature of Tse2 is a complicating factor for analyzing the protein in heterologous systems such as E . coli . Indeed , upon induction of its synthesis from B2H vectors , we found that cellular physiology was rapidly modified , obscuring interpretation of results ( data not shown ) . To facilitate the study of Tse2 in the B2H , we used an allele of the gene encoding a non-toxic Tse2 variant , Tse2T79A S80A ( Tse2NT ) . A description of this mutant is provided below . E . coli strains expressing Tsi2 C-terminally fused to the vesicular stomatitis virus glycoprotein ( VSV-G ) epitope followed by the Zif protein ( Tsi2–V–Zif ) and Tse2NT–ω showed significantly enhanced LacZ activity over control strains , indicating that Tsi2 and Tse2 directly interact ( Figure 2A ) . The interactions between dedicated secretion chaperones and their effector substrates often enhance stability of the effector in the bacterial cytoplasm [26] . However , if Tsi2 behaves analogously to typical antitoxin components of TA modules , even in the absence of stress it would be expected to have a shorter lifetime than the toxin – leaving it unable to act directly in stabilizing Tse2 [27] . Therefore , prior to determining if Tsi2 influences the stability of Tse2 , we queried the relative stabilities of the two proteins in P . aeruginosa . Western blot analyses of cells following treatment with the protein synthesis inhibitor tetracycline showed an almost complete loss of intact Tse2–V after one hour ( Figure 2B ) . However , no significant decrease in Tsi2–V levels was observed in the cells over the same time period . This was not the result of Tsi2 stabilization by its C-terminal VSV-G fusion , as an N-terminal VSV-G-fused Tsi2 ( V–Tsi2 ) displayed similar stability ( Figure S1 ) . The finding that Tsi2 is more stable in cells than its cognate toxin motivated us to further investigate its potential chaperone activity . Next we sought to ascertain the influence of Tsi2 on the stability of Tse2 in P . aeruginosa . Since Tse2 is toxic in cells lacking Tsi2 , this experiment required the use of a non-toxic variant of Tse2 . Furthermore , it was necessary that this variant bore only minimal perturbations , such that its stability and overall structure accurately reflected that of the native protein . In light of the additional objective of the study to probe the involvement of Tsi2 in Tse2 secretion , we added the requirement that the non-toxic variant was competent for export through the H1-T6S apparatus . The mechanism of action of Tse2 is not known and our analyses have so far failed to identify sequence motifs that would facilitate the prediction of residues essential for its function . Therefore , we adopted a scanning mutagenesis approach for defining minimal inactivating mutations . Using site-directed mutagenesis , we generated 15 tse2 alleles encoding adjacent double alanine substitutions at ten amino acid intervals along the length of the protein . In the four cases wherein one of these positions already encoded an alanine , only one substitution was made . Toxicity of the mutant alleles was determined by monitoring the effect of their ectopic expression on the growth of PAO1 Δtse2 Δtsi2 . Two alleles , tse2T79A S80A –V ( tse2NT–V ) and tse2V109A K110A–V , displayed a marked decrease in cytotoxicity relative to the wild-type protein ( Figure 3A ) . Western blot analysis showed that the mutant proteins accumulated to levels similar to those of the wild-type protein , suggesting that their lack of toxicity is due to inactivation of the toxin rather than poor expression or decreased stability ( Figure S2 ) . The sequence and structural determinants for effector export by the T6SS remain unresolved; therefore , we proceeded to empirically determine whether the non-toxic Tse2 variants retained H1-T6SS-dependent secretion . Expression plasmids directing the synthesis of Tse2NT–V and Tse2V109A K110A–V , or Tse2–V were introduced into P . aeruginosa ΔretS Δtse2 . The genes inactivated in this strain lead to constitutive export of effectors by the H1-T6SS ( ΔretS ) and avoid potential competition between native Tse2 and the ectopically-produced protein for the secretory apparatus ( Δtse2 ) . As an additional control , we also transformed a plasmid directing the synthesis of Tse2–V into ΔretS Δtse2 ΔclpV1 . The clpV1 gene encodes a AAA+ family ATPase that is an important determinant of effector export by the H1-T6SS . Consistent with our earlier finding that Tse2 is a substrate of the H1-T6SS , the wild-type protein was detected in concentrated culture supernatants in a manner dependent on clpV1 ( Figure 3B ) . The level of secreted Tse2NT–V was similar to that of the wild-type protein , whereas secretion of Tse2V109A K110A–V was not detected . From these data , we conclude that Tse2NT is a non-toxic substrate of the H1-T6SS . With a non-toxic and secreted Tse2 mutant in hand , we were able to test the involvement of Tsi2 in Tse2 stability and secretion by the H1-T6SS . The contribution of Tsi2 to Tse2 stability was determined by comparing the lifetime of Tse2NT–V when co-expressed in cells with Tsi2–V versus when expressed in cells devoid of Tsi2 . For co-expression , Tse2–V and Tsi2–V were produced in their native bicistronic configuration under the control of an inducible promoter . Our data showed that the presence of Tsi2–V significantly extends the lifetime of Tse2NT–V in P . aeruginosa . In the absence of Tsi2–V , intact Tse2NT–V was not detected beyond 15 minutes following the inhibition of protein synthesis; however , it persisted for 60 minutes in the presence of the immunity protein ( Figure 3C ) . The short cellular lifetime of Tse2–V was not due to either its fusion to the VSV-G epitope tag nor to secretion via the H1-T6SS ( Figure S1 ) . To determine the contribution of Tsi2 to Tse2 export by the H1-T6SS , levels of Tse2NT–V in culture supernatants from P . aeruginosa ΔretS strains with or without tsi2 were compared using quantitative western blotting . This analysis clearly demonstrated that Tsi2 is not required for Tse2 secretion by the H1-T6SS ( Figure 3D ) . Therefore , despite the direct interaction of Tsi2 with Tse2 , and the role of this interaction in stabilizing intracellular Tse2 , Tsi2 appears to have no role in targeting Tse2 to the secretion apparatus . To gain additional mechanistic insights into Tsi2 function , we solved its X-ray crystal structure to a resolution of 1 . 00 Å ( Table S1 , Figure 4A ) . Phasing of the structure was obtained using the multiwavelength anomalous diffraction method on a 1 . 68 Å resolution dataset collected from a crystal containing selenomethionine-substituted , C-terminal hexahistidine-tagged Tsi2 ( Tsi2–H6 ) [28] . A molecular model fit to a 1 . 68 Å resolution electron density map was used to calculate phases for a 1 . 00 Å data set collected from an isomorphous crystal of native Tsi2–H6 . Two monomers interacting through extensive contacts were modeled into the crystallographic asymmetric unit ( Figure 4B ) . Each Tsi2 monomer consists of two large α-helices ( α1 , amino acids 4–26 ( based on monomer A ) ; α2 , 30–62 ) arranged as an anti-parallel coiled-coil connected by a short turn ( T1 , 27–29 ) ( Figure 4C ) . The remaining C-terminal end of the protein is composed of a short helical segment ( α3 , 67–72 ) located between two extended loops ( L1 and L2 ) . In the observed Tsi2 dimer , the long axes of the two monomers are arranged approximately perpendicular to each other and the molecules pack via their coiled-coils . This interaction involves a large ( 725 Å2 , 13% total surface area ) and predominately hydrophobic ( 68% ) surface area , indicative of a physiologically relevant interface . In agreement with this , the molecular weight of purified Tsi2 measured by gel filtration chromatography was found to closely approximate that of the dimer ( calculated , 19 . 16 kDa; measured , 19 . 46 kDa ) , and we observed strong interaction between Tsi2 monomers using the B2H assay ( Figure 4D and Figure S3 ) . Superimposition of the Tsi2 monomers showed that overall the two subunits adopt highly similar structures ( Cα r . m . s . d , 1 . 1 Å ) . As Tse2 has proven recalcitrant to in vitro reconstitution , a direct biochemical study of the Tse2–Tsi2 interface has not been feasible . As an alternative strategy , we mutagenized 27 solvent-accessible Tsi2 residues to alanine and probed for effects on toxin immunity as a proxy for functional interaction with Tse2 . None of these substitutions , nor a truncation of Tsi2 lacking residues C-terminal of α3 , showed a measurable impact on Tse2 interaction ( Figure S4 ) . We did not attempt to analyze more extensive truncations of Tsi2 , as removing residues from α1 or α2 would likely disrupt its overall fold . From these results we conclude that the interaction of Tse2 with Tsi2 does not require the C-terminal loops and helix of Tsi2 , and that interaction is resilient to minor perturbations of the Tsi2 surface . The difficulty we encountered rationally dissecting the Tse2 binding site of Tsi2 led us to pursue a genetic screening strategy . The screen we designed exploited our ability to detect Tsi2 homodimerization and Tse2–Tsi2 association using the B2H ( Figure 5A ) . A random PCR-generated mutant library of tsi2 was inserted into the pACTR::V–zif B2H vector such that clones lacking nonsense mutations would generate N-terminal fusions to V–Zif ( Tsi2*–V–Zif ) . Next , the B2H was used to identify those pACTR::tsi2*–V–zif clones that did not activate lacZ expression when co-transformed into cells expressing Tse2NT–ω . After cultivation of positive clones , pACTR::tsi2*–V–zif plasmids were isolated , pooled and transformed into cells expressing Tsi2–ω . At this stage , clones of pACTR::tsi2*–V–zif that retained homotypic interaction , and therefore expressed high levels lacZ in the presence of Tsi2–ω , were selected for further characterization . While this second stage of our screen was critical for removing major sources of false positives , including tsi2 nonsense mutations and mutations that abrogated tsi2 expression , we are also aware of the caveat that it systematically eliminated the potential to recover tsi2* clones that affect both Tse2 and Tsi2 interaction . This issue is addressed below . Despite screening approximately 20 , 000 Tsi2* clones , we were able to identify only one single amino acid substitution , Tsi2E38K , that specifically abrogated Tse2 interaction when reconstructed and retested in the B2H ( Figure 5B ) . Interestingly , modeling of the electrostatic surface potential of Tsi2 showed that residues in the vicinity of Glu38 , including Asp23 , Asp45 and Asp49 , generate a prominent negatively charged surface feature ( Figure 5C ) . Based on these findings , we hypothesized that this acidic patch on the surface of Tsi2 contributes directly and critically to Tse2 binding . Although our structure of Tsi2 indicates that Glu38 does not mediate intramolecular interactions , we sought to rule-out the possibility that its non-conservative substitution with Lys perturbs native Tsi2 structure and indirectly leads to a loss of Tse2 binding . To this end , we purified Tsi2E38K–H6 and Tsi2–H6 from E . coli and compared their secondary structure by circular dichroism spectroscopy ( CD ) . Consistent with our X-ray crystal structure of Tsi2 , the CD spectrum of the wild-type protein showed strong helical character ( Figure S5 ) . The CD spectrum of Tsi2E38K–H6 displayed close agreement with the wild-type , suggesting that the Lys substitution does not significantly alter Tsi2 structure ( Figure S5 ) . Tsi2 is a strongly acidic protein ( calculated isolectric point , 4 . 1 ) with several solvent-exposed negatively charged amino acids located outside of the Glu38-containing acidic surface patch ( Figure 5C ) . To further investigate the specific involvement of this region on Tse2 interaction , we compared the effects of substituting Glu and Asp with Lys within , and outside of , its boundary . In total , we constructed nine additional lysine substitution mutants: three within the acidic patch ( D23K , D45K , D49K ) and six outside ( D30K , E36K , E56K , D69K , E73K , E74K ) . Using the B2H assay , each Tsi2 substitution mutant was probed for its capacity to both dimerize and associate with Tse2 . The nine variants expressed to similar levels as the wild-type protein , and , as expected , none of the substitutions affected Tsi2 dimerization ( Figures 5B and S6 ) . Interestingly , while substitutions outside of the acidic patch had no effect on Tse2 interaction , Tsi2 bearing a lysine at position 45 ( Tsi2D45K ) , located within the acidic patch , was incapable of binding to Tse2 ( Figure 5B ) . CD spectroscopy confirmed that Tsi2D45K–H6 retained the secondary structure of the wild-type protein ( Figure S5 ) . Surprisingly , no effect on Tse2 binding was caused by the Asp23Lys or Asp49Lys substitutions , suggesting that the residues critical for Tse2 binding within the acidic patch of Tsi2 are Glu38 and Asp45 . The observed lack of Tse2 binding by Tsi2E38K–V and Tsi2D45N–V was also reflected in the immunity properties of the proteins . We observed that both proteins fail to rescue Tse2-based toxicity when expressed in E . coli ( Figure 5D ) . The prominent role of Glu38 and Asp45 in Tse2 binding is further supported by the results of additional mutagenesis studies . Conservative substitutions introduced at these positions displayed a synergistic effect on Tse2 binding . Neither Tsi2 substitutions Glu38Gln nor Asp45Asn alone had a measurable impact on Tse2 interaction , however their combination reduced Tse2 binding by approximately 50% , as determined using the B2H assay ( Figure 5E ) . From these data , we conclude that Glu38 and Asp45 of Tsi2 are major determinants of Tse2 interaction . Furthermore , the substantial loss of Tse2 binding observed upon mutation of these residues to glutamine and asparagine , respectively , suggests that the Tse2–Tsi2 interface is stabilized in part by electrostatic interactions . As mentioned above , one caveat of our screen for Tse2-binding determinants of Tsi2 is that it excluded those mutations that also disrupt Tsi2 dimerization . It is conceivable that disruption of the Tsi2 dimer has a generally negative impact on Tse2 binding . Such a scenario could explain the difficulty we encountered in identifying Tsi2 substitutions that lose Tse2 , but not Tsi2 interaction . To address this issue , we probed the requirement for Tsi2 dimerization in the interaction of the protein with Tse2 . Inspection of the dimer interface permitted the identification of several candidate non-conservative single amino acid substitutions that we predicted could destabilize the Tsi2 dimer . To minimize the probability that our mutations – if successful in disrupting the Tsi2 dimer – would not disrupt Tse2 binding , we limited our mutagenesis to single substitutions . Our initial attempts focused on Arg18 and Glu21 , which form a network of polar intersubunit interactions at the origin of the non-crystallographic two-fold rotation axis relating the two subunits in the dimer ( Figure 6A ) . However , B2H analyses of alanine substitution mutants at these positions showed that disruption of this network does not significantly destabilize the overall dimer interface ( Figure S7 ) . Together these residues constitute the most significant polar intersubunit contacts , thus we concluded that interfering with hydrophobic interactions would be necessary to disrupt the Tsi2 dimer . Hydrophobic interactions between Tsi2 monomers are extensive . To increase the likelihood that single amino acid substitutions in non-polar residues at this interface resulted in significant perturbation , we replaced selected residues with the large polar amino acid glutamine . Three spatially distributed small hydrophobic interface residues ( all >90% solvent inaccessible ) , Val10 , Cys14 , and Ala47 , were selected , substituted with glutamine , and tested for homodimerization ( Figure 6A ) . Of the three mutant proteins , only Tsi2A47Q displayed decreased activity in the B2H ( Figure 6B ) . Levels of this protein were similar to the other mutant proteins and the wild-type , consistent with the diminished activity observed specifically resulting from a failure to efficiently homodimerize . As an independent measure of dimer formation by Tsi2A47Q , we employed an in vitro cysteine accessibility assay . This assay is based on our observations that Tsi2 possesses only one cysteine residue , and that this amino acid is solvent inaccessible at the dimer interface ( Figure 6A ) . Destabilization of the interface is expected to increase the reactivity of the Cys14 sulfhydryl side chain to small soluble maleimide-containing probes . To test for differential reactivity at this site , we purified Tsi2–H6 , Tsi2A47Q–H6 , and Tsi2C14A–H6 and Tsi2E21A–H6 as controls , and incubated the proteins with biotin-maleimide . Reactions were separated by SDS-PAGE and protein-biotin-maleimide conjugates were visualized by probing with Neutravidin-HRP . Consistent with our B2H data , Tsi2A47Q–H6 reacted more rapidly than the wild-type protein or Tsi2E21A–H6 ( Figure 6C ) . Tsi2C14A–H6 displayed no observable reactivity under these conditions , indicating that the products observed were specifically due to the reactivity at this site . Taken together with our B2H data , these data strongly suggest that the native Tsi2 dimer interface is significantly disrupted in Tsi2A47Q . With a Tsi2 dimer-defective variant in hand , we next sought to measure the impact of dimer disruption on heterotypic interactions of Tsi2 with Tse2 . First , using the B2H assay , we observed no significant difference between Tsi2–V and Tsi2A47Q–V binding with Tse2 ( Figure 6D ) . As a second , functional measure of Tse2–Tsi2 interaction , we also tested the capacity of Tsi2A47Q to provide immunity to Tse2 . Our data showed that Tsi2A47Q–V , like Tsi2–V , provides full protection against Tse2-based toxicity in E . coli ( Figure 6E ) . In total , these data show that the Tse2–Tsi2 interaction is insensitive to the oligomeric state of Tsi2 . Taken together with our screening data , we conclude that the regions of Tsi2 important for Tsi2 homotypic versus Tse2 heterotypic interactions are topologically distinct; Tse2 binding occurs at the face of Tsi2 opposite the site of homodimerization .
This study has shown that the Tse2–Tsi2 system has a unique set of properties that do not neatly conform to existing paradigms for toxin-antitoxin and effector-chaperone systems . For example , unlike canonical antitoxins , Tsi2 is more stable than its cognate toxin . This may reflect different physiological functions of the two systems . While the roles of TA systems are variable , and in certain instances remain a matter of debate , it is well established that they serve important functions in gene maintenance and response to stress [20] . For both of these functions , the activity of the TA system involved is mediated by modulation of toxin activity through antitoxin degradation . Our finding that Tsi2 stability greatly exceeds that of Tse2 suggests that this system has not evolved to conditionally release the toxin . Therefore , assuming an adequate expression level , P . aeruginosa strains endowed with tsi2 are likely to possess stable , non-dynamic immunity to growth inhibition by Tse2 . In this way , Tsi2 is more akin to certain colicin immunity proteins , which bind their cognate toxin with extraordinary affinity and provide complete protection against both endogenous and xenogenous cognate toxin [17] , [29] . Despite the functional disparities between the Tse2–Tsi2 and TA systems , they do possess notable parallels . One common property of TA systems is that the components have strongly opposing electrostatic character [19] . In the majority of instances , the antitoxin is more acidic than its cognate toxin . This is also the case for the Tse2–Tsi2 system , wherein Tsi2 is highly acidic and the difference in the calculated isoelectric points of the two proteins is 2 . 6 . This analogy holds , and is even more striking , when one considers the other Tse-Tsi pairs . The differences in toxin and immunity isoelectric points are greater in magnitude for these proteins ( Tse/i1 , 4 . 1; Tse/i3 , 3 . 6 ) , and in both cases the immunity proteins are acidic . A second shared physical attribute of TA and the Tse2–Tsi2 systems is that their antitoxin and immunity components , respectively , display modularity in their homotypic and heterotypic interactions . Although type II antitoxins are highly diverse at the sequence level , recent biochemical and structural analyses of these proteins indicate that they often exist as dimers and , despite their small size , homomeric contacts occur at a site physically removed from the site of cognate toxin interaction [30] . Our discovery that amino acids positions of Tsi2 critical for Tse2-binding reside on the face of the protein opposite from those involved in its homodimerization , taken together with our ability to readily generate specific loss-of-function mutations at either of these sites , strongly suggest an analogous configuration of the Tse2–Tsi2 complex . We found that co-expression of tsi2 with tse2 leads to a significant increase in the stability of the toxin , suggesting that the two proteins closely interact . Despite this , cells lacking Tsi2 have no detectable defect in Tse2 secretion , indicating that Tsi2 does not–in addition to its immunity properties–play a role in targeting Tse2 to the secretion apparatus . The specialization of Tsi2 as an immunity protein is in line with our current understanding of the function of other T6SS effector immunity proteins , Tsi1 and Tsi3 . These proteins reside in the periplasm and therefore are unavailable to assist in ushering their cognate toxins to the H1-T6SS [14] . This leaves open the question of how T6S effectors are recognized by the apparatus . One possibility is that effectors are bound by yet unidentified specialized secretion chaperone ( s ) . In this case , such a protein might remove Tse2 from the Tse2–Tsi2 complex prior to export . Since we observed no impact of Tsi2 on Tse2 secretion , we would expect that such a protein would either bind Tse2 with higher affinity than Tsi2 , or that it would bind a region of Tse2 not involved in Tsi2 binding and target the protein to the secretion machine , where Tsi2 would be readily removed . An alternative explanation for our finding that Tsi2 has no impact on Tse2 export is that the Tse proteins are exported co-translationally . In this model , Tsi2 , like Tsi1 and Tsi3 , might be present largely to protect against cognate Tse proteins arriving in trans via the T6SSs of adjacent bacteria . Co-translational export of the Tse proteins could also help reconcile how periplasmic effectors , in particular Tse1 , which possess numerous cysteine residues , avoid misfolding in the reduced cytoplasmic environment . Based on this model , one would predict that the H1-T6SS and its effectors would be tightly co-regulated . In P . aeruginosa , expression of tse and HSI-I genes ( encode the H1-T6S apparatus ) are coordinately co-regulated by the Gac/Rsm pathway [7] , [31] . Interestingly , this pathway stringently controls expression at the posttransciptional level , which would appear logical if a build-up of cytoplasmic effector was undesirable . Teng and colleagues recently reported the crystal structure of the N-terminal three-helix bundle ( Habc ) domain of the yeast SNARE ( soluble N-ethylmaleimide-sensitive factor activating protein receptor ) Vti1p bound to its epsin-like adaptor protein Ent3p [32] . According to analyses using DALI , the Vti1p Habc domain is the most similar structure to Tsi2 in the current protein databank ( Z score , 7 . 8; Cα r . m . s . d , 1 . 2 Å ) . Specifically , a close match of the length and curvature of the two large helices of Tsi2 to helices A and B of the Habc domain leaves these regions of the two structures nearly indistinguishable ( Figure 7 ) . Not only are these proteins structurally related , they also appear to interact with binding partners in a spatially and chemically similar fashion . Two acidic residues located on helix B of the Habc domain , Glu42 and Asp46 , were identified by Teng and colleagues as critical determinants of Ent3p binding . Substitution of either residue with arginine severely disrupted the interaction of Vti1p with Ent1p and led to mislocalization of Vti1p in yeast [32] . Interestingly , in an overlay of the two structures , these residues are found in close proximity to the acidic residues of Tsi2 discovered in our study to mediate Tse2 interaction ( Figure 7 ) . Although the simple structure of Tsi2 necessarily reduces confidence in interpreting the significance of structural similarity to the protein , we find the extent of structural and functional similarity between Tsi2 and the Habc domain striking . Our overall limited understanding of T6S makes it difficult to reconcile the relatedness of Tsi2 to the Habc domain , however it is worth noting that Tsi2 is now the second T6S protein shown to possess structural homology to an N-terminal regulatory domain of a SNARE protein . We recently reported that TagF , a negative posttranslational regulator of the T6SS , displays significant similarity to the N-terminal longin domain of Sec22b [33] . This domain has no structural homology to the Habc domain , however both can function analogously in directing subcellular localization by mediating interactions with adaptor proteins [34] . It will be of interest to determine the evolutionary mechanisms underlying the relationship between Tsi2 and TagF and proteins involved in vesicle trafficking . The structure of Tsi2 marks an important first step toward a complete molecular characterization of the Tse2–Tsi2 T6S toxin-immunity pair . However , many key outstanding questions remain . Foremost among these remains the mechanism of action of Tse2 . Our analysis of the effects of Tse2 on P . aeruginosa cells during intraspecies competition suggests that the protein acts efficiently and specifically to cease growth , and avoid killing targeted cells . Such effects have been observed for TA system toxins that act by cleaving mRNA , such as RelE [35] . Strong evidence that Tse2 also functions as a ribonuclease is lacking , however there are noteworthy indications . For example , the Phyre ( Protein Homology/AnalogY Recognition Engine ) structure prediction algorithm reports similarity between Tse2 and enzymes that bind and hydrolyze nucleic acids [36] . We found that an acidic patch of amino acids located on the surface of Tsi2 mediates interaction with , and immunity against Tse2 . If like most antitoxins , Tsi2 inhibits its cognate toxin by active site occlusion , it is conceivable that the negatively charged character of Tsi2 could engage basic residues of Tse2 that would otherwise participate in nucleic acid binding [37] .
Pseudomonas aeruginosa strains used in this study were derived from the sequenced strain PAO1 [38] . P . aeruginosa strains were grown on Luria-Bertani ( LB ) media at 37°C supplemented , when appropriate , with 30 µg/ml gentamycin , 25 µg/ml irgasan , 40 µg/ml X-gal , and stated concentrations of Isopropyl β-D-1-thiogalactopyranoside ( IPTG ) . Escherichia coli strains used in this study included DH5α for plasmid maintenance , SM10 for conjugal transfer of plasmids into P . aeruginosa , and BL21 pLysS for expression of Tse2 and Tsi2 . The tse2 , tsi2 genes and tse2 tsi2 and tse2–vsv-g tsi2 bicistronic sequences were PCR-amplified from P . aeruginosa genomic DNA and cloned into pPSV35CV [13] , pET29b+ and pET21a+ vectors ( Novagen ) . Site-directed mutants of tsi2 and tse2 were generated using either QuickChange ( Stratagene ) or Kunkel mutagenesis procedures [39] . Chromosomal fusions and in-frame gene deletions were generated as described previously and were verified by DNA sequencing [7] , [40] . The ΔHSI-I strain was constructed such that all sequence between nucleotide 2015 of PA0074 ( ppkA ) and nucleotide 754 of PA0091 ( vgrG1 ) were deleted . P . aeruginosa cultures were grown overnight at 37°C in LB broth containing 0 . 01% L-arabinose . In each experiment , the donor strain contained lacZ inserted at the neutral phage attachment site [41] . LacZ-labeled donor and non-labeled recipient strains were mixed at a ratio of 1∶1 and spotted onto a 0 . 2 µm nitrocellulose membrane ( Whatman ) on a 3% LB agar plate containing 0 . 2% L-arabinose . Competitions were incubated at 37°C and harvested at the indicated time points by resuspending bacterial cells in LB and plating onto LB plates containing 40 µg/ml X-gal for CFU enumeration . Growth competition assays of P . aeruginosa ΔretS and ΔretS Δtse2 against P . aeruginosa ΔretS Δtse2 Δtsi2 were performed on filters as described above . At 4 hrs after initiating the experiment , the filters were removed from agar plates and resuspended in 3 ml LB broth . The cells were collected by centrifugation , washed once with 1× phosphate buffered saline ( PBS ) and resuspended in 100 µl PBS . The bacterial suspension was stained with the LIVE/DEAD BacLight Bacterial Viability Kit ( Molecular Probes ) according to the protocol of the manufacturer . Viability was measured using a fluorescence microscope equipped with FITC and mCherry filters . The ratio of live/dead cells was determined by calculating the green/red fluorescent cells for 12 random fields per competition . Three independent experiments were performed . C-terminal hexahistidine-tagged Tsi2 ( Tsi2–H6 ) proteins were overexpressed in E . coli BL21 pLysS . Overnight cultured cells were back-diluted 1∶1000 into fresh 2× Yeast Tryptone ( YT ) media or defined SelenoMet medium base and SelenoMet nutrient mix medium ( Athena Enzyme Systems ) . Expression was induced at an OD600 of 0 . 5 with 0 . 1 mM IPTG for 16 hrs . at 20°C . Cells were harvested by centrifugation ( 8000× g; 20 min , 4°C ) and resuspended in lysis buffer [50 mM Tris-HCl , pH 7 . 5 , 0 . 5 M NaCl , 1% ( v/v ) Triton X-100 , 5% ( v/v ) glycerol , 1 mM DTT , and protease inhibitor cocktail ( Roche Diagnostics ) ] . Tsi2–H6 was purified by affinity chromatography using a HisTrap FF column ( GE Healthcare ) followed by size-exclusion chromatography on a HiPrep 16/60 Sephacryl S-200 high-resolution column ( GE Healthcare ) using the AKTA Explorer FPLC System . Purified proteins were stored in a buffer containing 50 mM Tris-HCl pH 7 . 5 , 500 mM NaCl , 1 mM DTT , and 5% ( w/v ) glycerol and dialyzed into a buffer containing 5 mM Tris-HCl pH 7 . 5 , 5 mM NaCl , and 1 mM DTT for crystallization . Crystals of Tsi2-H6 were grown at 25°C by hanging drop vapor diffusion . An equal volume of 10 mg/ml protein sample was mixed with the crystallization solution ( 0 . 1 M sodium acetate , pH 5 . 0 and 8% polyethylene glycol ( PEG ) 4000 ) . Crystals were cryo-protected in reservoir solution containing 25% PEG 4000 and flash frozen in liquid nitrogen . Diffraction data were collected at the Lawrence Berkeley National Laboratory Advanced Light Source ( ALS ) Beamline 8 . 2 . 1 ( University of California , Berkeley ) . Data were reduced using HKL2000 [42] . Phases were obtained experimentally with data obtained from selenomethionine-substituted Tsi2-H6 for structure solution by multi-wavelength anomalous dispersion ( MAD ) using the SOLVE program [43] . The final model was built by iterative model building and maximum likelihood refinement with Refmac-5 [44] . Finally , 123 well-defined water molecules were added , and refinement was continued until the R-value converged to 0 . 144 ( Rfree = 0 . 176 ) for all reflections to 1 . 00 Å resolution . The CCP4 [45] suite and XtalView [46] were used for crystallographic calculations . Molecular figures were generated with PyMOL [47] and CCP4 Molecular Graphics [48] . The model was validated using MolProbity [49] . All residues in the final model lie within allowed regions of a Ramachandran plot and 99 . 4% lie within the Ramachandran favored region . The crystal structure and structure factors have been deposited in the Protein Data Bank ( PDB entry 3RQ9 ) [50] . Tse2 and Tsi2 derivatives were cloned into pBRGP–ω and pACTR–V–zif plasmids [25] , [51] . The pBRGP::tsi2–ω and pBRGP::tse2NT–ω derivatives direct the synthesis of Tsi2 or Tse2 wild-type and mutant alleles as N-terminal fusions to the ω subunit of E coli RNAP . Plasmid pACTR::tsi2– V–zif directs the synthesis of the Tsi2-VSV-G fusion to the N-terminus of the zinc finger DNA-binding domain of murine Zif268 ( Zif ) . The tsi2 gene was mutagenized randomly by PCR with Taq DNA polymerase . A pool of plasmids encoding the resulting tsi2 mutants were ligated into the pACTR– V–zif plasmid and transformed into DH5α-F′IQ cells . All resulting transformants were pooled for plasmid isolation . Pooled plasmids were co-transformed with pBRGP::tse2NT–ω into KDZif1ΔZ competent cells and plated onto LB plates containing 12 . 5 µg/ml tetracycline , 150 µg/ml carbenicillin , 50 µg/ml kanamycin , 40 µg/ml X-gal , and 500 µM Phenylethyl-β-D-galactosidase ( tPEG ) . LacZ negative ( white ) colonies were selected for inoculation into 96 well plates , subcultured three times to cure plasmid pBRGP::tse2NT–ω and pooled for plasmid isolation yielding pACTR::tsi2*–V–zif . Purified plasmids were digested by ScaI and T7 exonuclease for removal of pBRGP::tse2 NT–ω . After purification , the mutated pACT::tsi2*–V–zif plasmids were co-transformed with pBRGP::tsi2–ω into KDZif1ΔZ competent cells and transformants were plated onto LB plates containing 12 . 5 µg/ml tetracycline , 150 µg/ml carbenicillin , 50 µg/ml kanamycin , 40 µg/ml X-gal and 500 µM tPEG . LacZ positive ( blue ) colonies were subcultured and subjected to plasmid isolation and subsequent sequencing analysis . E . coli KDZif1ΔZ cells were grown to an OD600 of 1 . 0 , permeabilized with 10% CHCl3 , and β-galactosidase activity was quantitatively assayed using a Galacto-Light Plus kit as previously described [52] . Assays were performed with at least two individual experiments in triplicate . Representative data sets are shown and values consist of averages based on three independent measurements from one experiment . Cell associated and supernatant protein samples were prepared as previously described [7] . Western blotting was performed as described previously using α-VSV-G , α-Tse2 and α-RNA-polymerase , with the modification that α-VSV-G antibody probing was performed in 5% BSA in Tris-buffered saline containing 0 . 05% v/v Tween 20 [14] . HisProbe-HRP Kit was used for direct detection of recombinant His-tagged proteins according to the manufacturer's instructions ( Thermo Scientific ) . For growth curves , E . coli BL21 pLysS cells harboring pET29b+ expressing Tse2 and Tsi2 derivatives were grown overnight in liquid LB broth at 37°C and back-diluted into LB broth ( 1∶1000 ) supplemented with 50 µg/ml kanamycin and 12 . 5 µg/ml chloramphenicol . Cultures were grown to an OD600 of 0 . 1–0 . 2 and induced with 0 . 2 mM IPTG . OD600 measurements were determined for E . coli strains in LB broth using an automated BioScreen C Microbiology plate reader with agitation at 37°C . Three independent measurements were performed in triplicate for each strain . A VSV-G epitope sequence was fused to tsi2 to allow for analyzing Tsi2 expression . For growth on solid medium , E . coli BL21 pLysS cells expressing Tse2 and Tsi2 derivatives were grown on LB agar plates with or without IPTG induction . CD spectra were recorded on a Jasco J810 Circular Dichroism Spectrometer using a 1 mm path-length quartz cuvette ( Starna ) . Tsi2 proteins were measured in triplicate at 195–260 nm in 1× PBS buffer , pH 7 . 5 at 25°C . A total of three scans were recorded and averaged for each spectrum . Purified proteins were exchanged into a 1× PBS buffer containing 50 mM NaCl , pH 7 . 5 . Biotin-maleimide was solubilized in Dimethyl sulfoxide ( DMSO ) and added to the protein samples at a final concentration of 10 µM . Protein samples ( 10 µM ) were incubated with biotin-maleimide at room temperature and reactions were quenched at indicated time points by the addition of a final concentration of 0 . 1 mM Tris , pH 8 . 0 . Western blots were used to detect biotin-maleimide labeled Tsi2 with NeutrAvidin and to detect His-tagged Tsi2 with HisProbe-HRP Kit . Purified Tsi2 was loaded onto a Superdex-200 10/300GL HR10/30 column ( GE Healthcare ) equilibrated with a buffer containing 50 mM Tris pH 7 , 500 mM NaCl , and 5% glycerol . Protein standards included ribonuclease A ( 13700 Da ) , carbonic anhydrase ( 29000 Da ) , ovalbumin ( 43000 Da ) , conalbumin ( 75000 Da ) , and aldolase ( 158000 Da ) . P . aeruginosa Δtse2Δtsi2 strains harboring pPSV35::tse2NT–V , pPSV35::tse2–V tsi2–V or pPSV35::tse2NT–V tsi2–V were grown at 37°C with aeration in LB broth containing 30 µg/ml gentamycin . Overnight cultures were back-diluted 1∶500 into LB containing 30 µg/ml gentamycin and 0 . 5 mM IPTG . After P . aeruginosa cells were grown at 37°C for 5 hrs . , protein synthesis was inhibited with the addition of 250 µg/ml tetracycline . Samples were taken at indicated time points and analyzed by Western blot . | Bacterial species have been at war with each other for over a billion years . During this period they have evolved many pathways for besting the competition; one of the most recent of these to be described is the type VI secretion system ( T6SS ) . The T6SS of Pseudomonas aeruginosa is a complex machine that the bacterium uses to intoxicate neighboring cells . Among the toxins this system delivers is type VI secretion exported 2 ( Tse2 ) . In addition to acting on competing organisms , this toxin can act on P . aeruginosa; thus , the organism synthesizes a protein , type VI secretion immunity 2 ( Tsi2 ) , which neutralizes the toxin . In this paper we dissect the function and structure of Tsi2 . We show that although Tsi2 interacts with and stabilizes Tse2 inside the bacterium , the toxin does not require the immunity protein to reach the secretion apparatus . Our structure of Tsi2 shows that the protein adopts a dimeric configuration; however , we find that its dimerization is not required for Tse2 interaction . Instead , our findings indicate that Tse2 interacts with an acidic surface of Tsi2 that is opposite the homodimer interface . Our results provide key molecular insights into the process of T6 toxin secretion and immunity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"biochemistry",
"gram",
"negative",
"protein",
"interactions",
"proteins",
"microbial",
"evolution",
"chaperone",
"proteins",
"protein",
"structure",
"microbial",
"pathogens",
"biology",
"microbiology",
"bacterial",
"pathogens",
"microbial",
"ecology"
]
| 2012 | Structural Basis for Type VI Secretion Effector Recognition by a Cognate Immunity Protein |
The Americas were the last continent colonized by humans carrying malaria parasites . Plasmodium falciparum from the New World shows very little genetic diversity and greater linkage disequilibrium , compared with its African counterparts , and is clearly subdivided into local , highly divergent populations . However , limited available data have revealed extensive genetic diversity in American populations of another major human malaria parasite , P . vivax . We used an improved sample preparation strategy and next-generation sequencing to characterize 9 high-quality P . vivax genome sequences from northwestern Brazil . These new data were compared with publicly available sequences from recently sampled clinical P . vivax isolates from Brazil ( BRA , total n = 11 sequences ) , Peru ( PER , n = 23 ) , Colombia ( COL , n = 31 ) , and Mexico ( MEX , n = 19 ) . We found that New World populations of P . vivax are as diverse ( nucleotide diversity π between 5 . 2 × 10−4 and 6 . 2 × 10−4 ) as P . vivax populations from Southeast Asia , where malaria transmission is substantially more intense . They display several non-synonymous nucleotide substitutions ( some of them previously undescribed ) in genes known or suspected to be involved in antimalarial drug resistance , such as dhfr , dhps , mdr1 , mrp1 , and mrp-2 , but not in the chloroquine resistance transporter ortholog ( crt-o ) gene . Moreover , P . vivax in the Americas is much less geographically substructured than local P . falciparum populations , with relatively little between-population genome-wide differentiation ( pairwise FST values ranging between 0 . 025 and 0 . 092 ) . Finally , P . vivax populations show a rapid decline in linkage disequilibrium with increasing distance between pairs of polymorphic sites , consistent with very frequent outcrossing . We hypothesize that the high diversity of present-day P . vivax lineages in the Americas originated from successive migratory waves and subsequent admixture between parasite lineages from geographically diverse sites . Further genome-wide analyses are required to test the demographic scenario suggested by our data .
Plasmodium vivax is the human malaria parasite with the widest global distribution and accounts for nearly half of the combined malaria burden in South and Southeast Asia , Oceania , and Central and South America . Over one-third of the world's population is currently at risk of infection with this species , with 16 million clinical cases recorded each year [1] . Although P . vivax has most likely evolved from parasites that infect chimpanzees and gorillas in sub-Saharan Africa [2 , 3] , it is nowadays rare in most of this continent , where human populations lack a key erythrocyte receptor for host cell invasion by blood-stage parasites , the Duffy antigen/receptor for chemokines ( DARC ) [4] . Where both species coexist , P . vivax typically causes less severe cases and fewer deaths than P . falciparum , the most virulent human malaria parasite , but represents a major challenge for ongoing malaria elimination efforts worldwide [1] . The Americas were the last continent colonized by humans carrying malaria parasites , but the dates and routes of migration of P . vivax to the New World are still debated [5–7] . Archaeological evidence for infection with this parasite in indigenous , pre-Columbian populations is currently limited to a single report of P . vivax antigens being visualized by immunohistochemistry in the liver and spleen of South American mummies dating from 3 , 000 to 600 years ago [8] . Interestingly , specific antibodies failed to detect P . falciparum antigens in these same samples [8] . These findings are consistent with the hypothesis that P . vivax , but not P . falciparum , was brought to the New World by early human migrations from East Asia or the Western Pacific [5] , but more specific molecular techniques are required to confirm them [9] . Nevertheless , present-day New World populations of P . vivax appear to be more closely related to extant African and South Asian parasites and now extinct European lineages than to East Asian and Melanesian strains [10–12] , consistent with much more recent parasite migrations with European conquerors and African slaves during the colonial era [7 , 13] . Clinical isolates from Brazil are underrepresented in global genomic analyses of P . vivax [11 , 14] . However , this country contributes 37% of the malaria burden in the Americas , a region with over 20 million people at high risk of infection [1] . Obtaining large amounts of host-free P . vivax DNA from clinical samples from Brazil has been a major challenge for genome sequencing projects , because ( a ) blood-stage parasite densities are typically very low [15] , ( b ) clinical blood samples are heavily contaminated with human DNA from leukocytes , and ( c ) methods for long-term in vitro propagation of P . vivax are neither practical nor widely reproducible [16–18] . Here , we combined an improved sample preparation strategy , for reducing human DNA contamination and increasing target parasite's DNA yield , with next-generation genome re-sequencing to examine a local population of P . vivax from the Amazon Basin of Brazil . Our nine high-quality genome sequences were compared to those previously obtained from four countries ( Brazil [BRA] , Peru [PER] , Colombia [COL] , and Mexico [MEX] ) [11 , 14] to reveal local and regional patterns of diversity and differentiation in extant P . vivax populations from the New World .
Study protocols were approved by the Institutional Review Board of the Institute of Biomedical Sciences , University of São Paulo , Brazil ( 936/CEP , 2010 and 1183/CEPSH , 2014 ) . Written informed consent was obtained from all patients . Parasite samples were collected between November 2012 and June 2013 in eastern Acre and southern Amazonas , Amazon Basin of Brazil , close to the borders with Peru and Bolivia . Malaria epidemiology in the study sites has been characterized in detail elsewhere [15 , 19] . Venous blood samples ( 10 ml ) were collected from eight adult patients attending malaria clinics in the town of Acrelândia , Acre ( 9°43' S , 66°53' W ) , and one adult patient living in the farming settlement of Remansinho , Amazonas ( 9°40' S-9°43' S , 66°52' W -66°59' W ) , 120 km east of Acrelândia ( Fig 1 ) . P . vivax infection was diagnosed by on-site microscopy and later confirmed by qPCR as described below . We adapted BioR 01 Plus leukocyte-depletion filters ( Fresenius Kabi , Bad Homburg , Germany; S1A Fig ) to process 10 to 50-ml volumes of venous blood in our field laboratory in the Amazon [20] . The per-unit cost of these filters in Brazil is around US$ 25 . We first cut off under sterile conditions , with a scissor , the tubing that connects the filtering device to the 400-ml blood storage bag and to the adapter ( S1B and S1C Fig ) . S1D Fig shows how the filtering device was used in a laminar flow hood; briefly , a 10-ml syringe was used to apply blood treated with acid citrate dextrose anticoagulant , while a second 10-ml syringe was adapted to the end of the remaining tubing to recover the filtered , leucocyte-depleted material , which was transferred to 50-ml sterile centrifuge tubes . No priming with saline was required . After the filtration process , the leukocyte depletion device was washed through with at least twice the volume of RPMI medium as the original blood sample to recover red blood cells ( RBCs ) that had been retained in the filter and tubing . Next , the mixture of filtered blood and RPMI medium recovered in 50-ml tubes was centrifuged at 800 g for 5 minutes and the supernatant ( plasma plus RPMI medium ) was removed with a sterile Pasteur pipette . For cryopreservation , the RBC pellet was resuspended dropwise in Glycerolyte-57 solution ( Fenwall , Fresenius Kabi ) , at the proportion of 1 . 66 ml of Glycerolyte for each 1 ml of cell pellet , under gentle agitation . One ml aliquots of the RBC-Glycerolyte mixture were transferred to screw-capped cryovials and placed in a Nalgene Mr . Frosty freezing container ( ThermoFisher Scientific , Waltham , MA ) that was kept at -80°C for 24 hours . The next day , cryovials were plunged in liquid nitrogen for long-term storage . Samples were shipped to São Paulo in liquid nitrogen , for subsequent schizont maturation . For sample thawing , cryovials were removed from the liquid nitrogen and maintained for 1 min at room temperature ( 20–25°C ) and 1 min at 37°C . The 1-ml samples were then transferred to 50-ml centrifuge tubes , gently mixed with 200 μl of 12% NaCl solution , and let stand for 1 minute . Next , 10 ml of 1 . 6% NaCl were gently added , the mixture was gently agitated , and centrifuged at 180 g for 8 min [21] . After removing the supernatant , the RBC pellet was washed twice in incomplete McCoy's 5A medium supplemented with glucose ( 0 . 5% w/v ) , HEPES ( 25 mM ) , and hypoxantine ( 0 . 005% w/v ) , and resuspended in complete McCoy medium ( as above but supplemented with 25% AB+ heat-inactivated human serum ) to a final hematocrit of 2% . Short-term culture in vitro was carried in flat-bottomed dishes placed in a gas chamber with controlled O2 and CO2 levels that was kept at 37°C for up to 48 hours . Parasite growth and maturation were monitored as described [17] . Chloroquine resistance ( CQR ) was evaluated using an ex-vivo schizont maturation assay in selected P . vivax samples with > 1 , 000 parasites/μl of blood and > 50% ring stages at the time of thawing [22] . There is no consensus regarding the 50% inhibitory concentration ( IC50 ) indicative of CQR in P . vivax; suggested cut off values range between 100 nM [23] and 220 nM [24] , but all IC50 values in our samples were < 50 nM ( Table 1 ) . DNA templates were isolated from 200-μl aliquots of either whole venous blood ( before leukocyte removal ) or RBC pellet ( after leukocyte removal ) using QIAamp DNA blood kits ( Qiagen , Hilden , Germany ) . To estimate the relative proportion of human and parasite DNA , we used SYBR Green qPCR targeting single-copy genes coding for human topoisomerase III and P . vivax aldolase . Each 15 μl qPCR mixture contained 2 μl of template DNA , 7 . 5 μl of 2× Maxima SYBR Green qPCR master mixture ( Fermentas , Burlington , Canada ) and 0 . 3 μM of each of the primer pairs , Pvaldo-F ( GAC AGT GCC ACC ATC CTT ACC ) plus Pvaldo-R ( CCT TCT CAA CAT TCT CCT TCT TTC C ) and Top3-F ( CAT GTT TGA GCT GAG CCT GA ) plus Top3-R ( CCA CAC CAC ACC CCT AAC TT ) . Standard curves were prepared with serial tenfold dilutions of a plasmid containing both target sequences to allow for copy number quantitation ( number of amplicons/μl of blood ) . We used a Step One Plus Real-Time PCR System ( Applied Biosystems , Foster City , CA ) 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 60°C , with fluorescence acquisition at the end of each extension step . Amplification was followed by a melting program consisting of 15 sec at 95°C , 15 sec at 60°C , and a stepwise temperature increase of 0 . 2°C/sec until 95°C , with fluorescence acquisition at each temperature transition . All reactions were made in triplicate . We measured parasite DNA enrichment as the parasite:human copy-number ratio after filtering divided by parasite:human copy-number ratio before filtering . To estimate the proportion of host and parasite DNA in each sample , we considered the amplicon copy numbers and the genome size of humans ( 3 . 2 Gb ) and P . vivax ( 26 . 8 Mb; Sal-I assembly ) ; the DNA content in each copy of the human genome corresponds approximately to that of 119 copies of the P . vivax genome . Parasite DNA templates were quantified by fluorometry using a Qubit 3 . 0 fluorometer ( Invitrogen , Carlsbad , CA ) and sequenced using Ion Torrent Personal Genome Machine ( PGM ) and Ion Proton platforms ( Life Technologies , Foster City , CA ) at the Unit of Computational Genomics , Laboratory of Bioinformatics , National Laboratory of Scientific Computation , Brazil . Separate libraries were prepared for each sequencing platform , using 1 μg of template DNA per isolate . DNA samples were sheared using the Bioruptor UCD-200 TS ( Diagenode , Liege , Belgium ) sonication system until fragment sizes of 200 bp ( for Ion PGM libraries ) or 150 bp ( for Ion Proton libraries ) were obtained . Libraries were prepared using the Ion Xpress Plus Fragment Library kit , with Ion Xpress Barcode adapters according to the Ion Xpress Plus gDNA Fragment Library Preparation protocol ( Life Technologies ) . Size selection was performed on E-Gel SizeSelect 2% agarose gels using the E-Gel iBase Power System ( ThermoFisher Scientific ) . Emulsion PCR was done on the Ion OneTouch 2 system ( Life Technologies ) with the Ion PGM Template OT2 200 kit or the Ion PI Template OT2 200 kit version 2 for Ion PGM and Ion Proton , respectively , following the manufacturer's instructions ( Life Technologies ) . Ion PGM libraries were loaded on Ion318 chips v2 and sequenced using the Ion PGM Sequencing 200 kit v2; Ion Proton libraries were loaded on Ion PI chips v2 and sequenced using the Ion PI Sequencing 200 kit v2 ( Life Technologies ) . All samples were sequenced on both platforms; sequence reads ( 150–200 bp ) from two runs in each platform were merged into a single fastq file per sample . To place our genomic data in a regional context , we reanalyzed raw paired-end Illumina reads from 107 additional P . vivax clinical isolates from the Americas . Fastq files were downloaded from the Sequence Read Archive ( SRA ) of the National Center for Biotechnology Information , United States , and processed in the same way as our newly obtained sequences . Three clinical isolates from Brazil had sequence data generated on an Illumina Genome Analyzer II platform at the Welcome Trust Sanger Institute ( Hinxton , Cambridge , UK ) , as part of the P . vivax genome variation project coordinated by the MalariaGEN network [14] . Other isolates—20 from Brazil , 34 from Peru , 31 from Colombia , and 19 from Mexico—had whole-genome sequence data generated on an Illumina HiSeq 2000 platform at the Broad Institute of MIT and Harvard ( Cambridge , MA , USA ) , as part of the International Centers of Excellence for Malaria Research ( ICEMR ) program [11] . All isolates from Brazil sequenced in these two previous studies were collected in endemic areas surrounding Acrelândia , Acre State , between 2008 and 2011 . Fastq files were first filtered for quality; 8 SRA samples were excluded from further analysis because of mean quality scores ≤ 30 ( expected base call accuracy ≤ 99 . 9% ) . We next mapped the high quality reads onto the PlasmoDB version 10 . 0 of the Sal-1 reference ( http://plasmodb . org/common/downloads/release-10 . 0/PvivaxSal1/fasta/data/PlasmoDB-10 . 0_PvivaxSal1_Genome . fasta ) [25] , using Bowtie2 version 2 . 2 . 6 [26] with the “very sensitive” preset , allowing one mismatch per seed region; 11 SRA samples were excluded at this stage because of < 60% mapping over the reference . The resulting alignments were merged into BAM files with SAMtools [27] , duplicate reads were identified and marked using the Picard version 2 . 0 . 1 MarkDuplicates tool , and files were indexed with SAMtools . We used GATK version 2 . 0 [28] for SNP calling following the GATK Best Practices ( https://software . broadinstitute . org/gatk/best-practices/ ) . GATK UnifiedGenotyper with the BaseAlignmentQuality option was used to obtain high-confidence SNPs by applying stringent VariantFiltration criteria: ( a ) coverage > 20× , ( b ) mapping quality > 30 , ( c ) base quality > 30 , read depth and allelic fraction by sample ≥ 1 , and ( d ) haplotype score ≤ 3 . We removed all SNPs with > 2 alleles and those with minor allele frequency < 0 . 01 ( reads counted across all sequenced samples ) . SnpEff [29] was used to identify SNPs mapping to coding sequences ( further classified as synonymous or nonsynonymous ) , introns , and intergenic regions of the Sal-I reference genome . The resulting catalogue of 94 , 122 high-confidence SNPs was used to genotype each individual sample using GATK UnifiedGenotyper with default parameters except for the minimum phred-scaled confidence threshold , with a calling variant = 50 and emitting variant = 10 . Heterozygote calls were converted to the majority allele if ≥75% of the reads in that sample were the majority allele; otherwise , the allele was undetermined . Sites with < 5× coverage in a given sample were filtered out at this stage . The final data set of P . vivax nuclear genome sequences from the New World comprised our 9 newly sequenced samples from BRA and 75 high-quality SRA samples ( 2 from BRA , 23 from PER , 31 from COL , and 19 from MEX ) ; 13 SRA samples were removed during the genotyping process because the number of SNPs identified ( range: 13 to 2 , 889 ) was below the predefined minimum of 3 , 000 nucleotide differences compared to the Sal-I reference , consistent with poor sequencing coverage . Overall , we excluded 21 ( out of 23 ) SRA samples from BRA and 11 ( out of 34 ) SRA samples from PER at different stages of this analysis . Isolate codes and SRA accession numbers of samples used in this analysis are given in S1 Table . The average pairwise nucleotide diversity ( π , average number of nucleotide differences per site between pairs of DNA sequences ) was calculated within each geographic population using VCFtools [30] . Values were plotted , using R version 3 . 3 . 0 , as moving averages within 1-kb sliding windows across each chromosome . We recalculated π after masking out subtelomeric regions and three hypervariable internal chromosome regions ( containing sera , msp-3 , and msp-7 gene families ) that were more prone to sequence misalignments and poor read mapping in a previous analysis [14] . The coordinates of these regions are given in the Supplementary Table 2 of Pearson et al . [14] . Similarly , Tajima's D values [31] were calculated using VCFtools; mean Tajima’s D across 1-kb windows were plotted for each population . Frequency distributions of π and Tajima’s D values within 1-kb windows were plotted for each population . We defined windows with the top 50 π values within each population as highly variable genomic regions . We defined as outliers the 1-kb windows with the 50 highest and 50 lowest mean Tajima's D values within each population . We also examined the minor allele frequency ( MAF ) spectrum separately in each population . We next estimated , for each population , the rate at which pairwise linkage disequilibrium ( LD ) decreased with increasing physical distance between SNPs due to meiotic recombination . The squared correlation coefficient r2 between pairs of SNPs of varying distance across the same chromosome was measured using VCFtools; r2 values were binned by distance ( 50-bp windows ) and medians within each window were plotted against physical distance between SNPs . The level of background LD between unlinked markers within each population was estimated by calculating median r2 between all pairs of SNPs on different chromosomes . The Wright's fixation index FST , a measure of population differentiation due to genetic structure [32] , was calculated with VCFtools for each SNP in every pairwise comparison of populations . Values were averaged across all SNPs to estimate overall pairwise differentiation between populations . The 100 SNPs with the highest average FST values across all populations were further characterized . To assess population structure , we first used the PLINK software ( https://www . cog-genomics . org/plink2 [33] ) to carry out principal component analysis ( PCA ) ; up to 10 components were analyzed . For phylogenetic analysis , a neighbor-joining tree was constructed via the maximum composite likelihood substitution model with 1 , 000 bootstrap pseudoreplicates using MEGA 7 . 0 ( http://www . megasoftware . net/ ) . To estimate the ancestry shared between individual isolates , we used the ADMIXTURE software package [34] with either all 94 , 122 high-quality SNPs or a curtailed set of 12 , 762 SNPs that are not linked . To this end , we removed each SNP that had an r2 value > 0 . 1 with any other SNP within a 60-SNP sliding window advanced by 10 SNPs each time . The optimal number of clusters ( K ) was determined by performing multiple runs of the software under different K values ( 2–10 ) and selecting the K values ( K = 2 and K = 3 ) associated with the lowest cross-validation error compared to other K values ( S2 Fig ) . The sequence data supporting the conclusions of this article are available in the Sequence Read Archive of the National Center for Biotechnology Information , United States; accession numbers are provided in S1 Table .
We designed a single-step procedure to reduce human DNA content in P . vivax-infected blood using commercially available leukocyte-depletion filters ( S1 Fig ) . Leucocyte depletion in 17 clinical samples , with initial parasitemias ranging between 854 and 43 , 177 ( median: 7 , 566 ) parasites/μl , decreased the percent human DNA content from a median of 99 . 2% ( range: 82 . 7–99 . 9% ) to 23 . 3% ( range: 0–97 . 6% ) . No human DNA could be detected by quantitative real-time polymerase chain reaction ( qPCR ) in three filtered samples ( i . e . , 0% host DNA contamination ) . The residual human DNA content in our leukocyte-depleted samples was similar to the median of 33 . 9% ( range: 1 . 6–68 . 6% ) found in blood samples from Indonesia after double filtration through CF-11 cellulose columns [17] . The increase in parasite:human DNA ratio after leukocyte depletion ranged between 1 . 7 and 3 , 060-fold ( median: 227-fold ) and appeared to be inversely proportional to the initial , pre-treatment parasite:human DNA ratio ( Spearman correlation coefficient rs = -0 . 512 , P = 0 . 061; S3 Fig ) , but did not correlate with the initial parasitemia ( rs = 0 . 165 , P = 0 . 412 ) . The next challenge consisted in selectively increasing parasite DNA yield for genome sequencing . To this end , leukocyte-depleted P . vivax samples were cultured in vitro for up to 44 h to allow uninucleate trophozoites to mature to multinucleate blood-stage schizonts , increasing parasite DNA content by ≥4-fold [17] . We obtained enough template DNA for library preparation and sequencing from 9 cultured samples , achieving 49 . 8–78 . 6× average genome sequencing depth; 91 . 8–94 . 7% of the reads mapped to reference Sal-I genome ( Table 1 ) . These results compared favorably with the coverage and percentage of reads mapping onto the reference P . vivax genome obtained with other different methods for clinical sample preparation: ( a ) single CF-11 column filtration ( 0 . 8–42 . 1× depth and 0 . 4–27 . 8% mapping in 8 isolates from Colombia [35] and 4 . 2–28 . 1× depth and 18 . 3–55 . 4% mapping in 11 isolates from Peru [36 , 37] ) , ( b ) double CF-11 column filtration ( 70–407× depth and 15 . 0–46 . 2% mapping in 5 isolates from Cambodia and Madagascar [38] ) , ( c ) double CF-11 column filtration followed by in vitro schizont maturation ( 18–116× depth and 11 . 0–89 . 1% mapping in 22 isolates from Thailand and returning travelers [17] ) , and ( d ) in-situ hybridization for P . vivax whole-genome capture in unfiltered samples ( 21 . 9–160 . 4× depth and 20 . 2–80 . 1% mapping in 5 isolates from Peru [16] and 34 . 8–118 . 2× depth and 19 . 6–39 . 7% mapping in 3 isolates from East Asia [39] ) . Therefore , the sample preparation strategy described here allowed for high sequence coverage and depth , but further comparisons are limited by the use of different next-generation sequencing platforms across studies ( Ion PGM and Ion Proton here and Illumina in other studies ) . We used the Ion PGM and Ion Proton platforms to generate between 17 . 8 and 23 . 9 million sequence reads from 9 clinical samples of P . vivax from Brazil ( Table 1 ) . To explore local levels of genomic diversity , our sequence data were combined with those from two P . vivax clinical samples from Brazil that were previously obtained with Illumina platforms [11 , 14] . Sequence reads from these additional samples ( PV4 and Brazil32 ) covered 86 . 2% and 95 . 0% of the reference genome , respectively ( S1 Table ) . Since all isolates from Brazil ( n = 11 ) were collected from sites within a radius of 120 km in the Amazon Basin ( Fig 1 ) , we define BRA as a local , nearly sympatric P . vivax population . After applying stringent quality control filters to raw sequence reads ( see Methods ) , we uncovered 27 , 360 high-confidence single-nucleotide polymorphisms ( SNPs ) in the BRA population . The overall nucleotide diversity π was estimated at 5 . 6 x 10−4 ) . We next compared BRA sequence data with those from three other New World populations of P . vivax: PER ( n = 23 ) , COL ( n = 31 ) , and MEX ( n = 19 ) [11] , from clinical isolates sampled in sites shown in Fig 1 . Two populations were geographically heterogeneous: PER samples were collected in three departments ( Loreto and Madre de Dios , both in the Amazon Basin , and Piura , on the northwestern Pacific coast ) , while COL samples came from four departments ( Nariño , Valle del Cauca , and Chocó along the Pacific Coast , and Córdoba , on the Caribbean coast ) . MEX samples were from five different sites , but all in the southern state of Chiapas ( S1 Table ) . We characterized 94 , 122 biallelic SNPs passing our high-quality filters in 84 samples; 55 . 0% of them are located in intergenic regions , 8 . 8% in introns , and 36 . 2% in coding regions . Most ( 61 . 2% ) coding SNPs were non-synonymous ( nsSNPs ) , as previously found in other regional P . vivax populations [14 , 38] . Unsurprisingly , the number of SNPs found in each population was directly proportional to sample size , being lowest in BRA ( n = 27 , 360 ) and highest in COL ( n = 57 , 262; Fig 2 ) . Overall , 49 , 598 ( 52 . 3% ) SNPs were unique to a population ( i . e . , private ) and 8 , 529 ( 9 . 1% ) were shared by all populations ( Fig 2 ) . Of the 6 , 891 private SNPs found in BRA , 40 . 0% mapped to coding sequences; 65 . 2% of the coding SNPs were non-synonymous . PER and COL shared the highest number of SNPs ( n = 28 , 667 ) , followed by COL and MEX ( n = 24 , 107 ) ( Fig 2 ) . Nucleotide diversity in PER ( π = 5 . 2 x 10−4 ) , COL ( π = 5 . 5 x 10−4 ) , and MEX ( π = 6 . 2 x 10−4 ) was similar to that in BRA ( π = 5 . 6 x 10−4 ) , showing that the local BRA population was as diverse as geographically heterogeneous sample sets from other American countries . Similar levels of genome-wide nucleotide diversity were recently reported in P . vivax populations from Colombia ( π = 6 . 8 x 10−4; n = 8 ) [35] , Thailand ( π = 5 . 3 x 10−4; n = 88 ) , Cambodia ( π = 5 . 0 x 10−4; n = 19 ) , and Indonesia ( π = 5 . 0 x 10−4; n = 41 ) [14] , but the differences across studies in genome sequencing depth and criteria for defining high-quality SNPs limit such comparisons . The frequency distribution of π values within 1-kb sequence windows across the genome was quite similar in all countries ( S4 Fig ) ; the interquartile ranges ( IQR ) were 1 . 7 x 10−4 to 6 . 9 x 10−4 in BRA , 1 . 7 x 10−4 to 6 . 5 x 10−4 in PER , 1 . 3 x 10−4 to 6 . 6 x 10−4 in COL , and 1 . 9 x 10−4 to 7 . 5 x 10−4 in MEX . All distributions were right-skewed , with asymmetry coefficients ranging between 3 . 37 ( BRA ) and 3 . 84 ( MEX ) . We defined domains with the top 50 π values in a population as highly variable genomic regions . They comprised , in addition to numerous sequences coding for hypothetical proteins , gene families such as pst-a , fam-b , fam-d , and fam-e [40] and those coding for major parasite antigens , such as the vir family ( >300 genes , mostly in subtelomeric domains , on several chromosomes ) , the serine repeat antigen ( sera ) family ( 13 genes on chromosome 4 ) , the merozoite surface protein ( msp ) -7 family ( 11 genes on chromosome 12 ) , and the msp-3 family ( 11 genes on chromosome 10 ) ( S2 Table ) ( see also [11 , 14 , 36] ) . These findings are not unexpected , since natural selection favors increased diversity in antigen-coding genes to evade host immunity , but must be interpreted with caution because misalignments of paralogous sequences may have inflated nucleotide diversity estimates in gene families . We thus recalculated genome-wide π values after masking out subtelomeric domains and the internal chromosomal regions comprising the sera , msp-3 and msp-7 gene families [14] , but this procedure affected our overall estimates very little; recalculated values were: BRA ( π = 5 . 5 × 10−4 ) , PER ( π = 5 . 2 × 10−4 ) , COL ( π = 4 . 7 × 10−4 ) , and MEX ( π = 5 . 2 × 10−4 ) . The single-copy msp-1 gene [41] also mapped to a highly variable genomic region in BRA . However , our nucleotide diversity estimates for msp-1 may have been affected by likely sequence misalignments in the numerous repetitive domains across this locus [42 , 43] . The expected value for Tajima’s D is zero under a neutral model that assumes random mating , no recombination , mutation-drift equilibrium , infinite sites , and constant population size . High Tajima’s D values are usually due to balancing selection or recent population size reduction , while negative D values are consistent with population size expansion or purifying selection [44] . Although the P . vivax genomes from COL and PER were not part of a single population ( while BRA and MEX genomes are ) , the Tajima’s D distribution provides information on how the pattern of mutations changes across the genome . All distributions of Tajima’s D values in our populations were right-skewed , with asymmetry coefficients of 0 . 477 ( BRA ) , 0 . 841 ( PER ) , 0 . 902 ( COL ) , and 0 . 412 ( MEX ) ( Fig 3 ) . Negative D values predominated in PER ( median: -0 . 439; IQR: -0 . 959 to -0 . 003 ) and COL ( median: -0 . 466; IQR: -0 . 959 to 0 . 039 ) , but not in BRA ( median: 0 . 026; IQR: -0 . 571 to 0 . 595 ) and MEX ( median: 0 . 039; IQR: -0 . 724 to 0 . 748 ) . Interestingly , the genomic regions with the 50 lowest Tajima’s D values within each population ( threshold D values: -1 . 096 in BRA , -1 . 622 in PER , -1 . 724 in COL , and -1 . 376 in MEX ) comprised several hypothetical and housekeeping genes , but also members of gene families such as vir , sera , msp-3 , pst-a , and fam-a ( S3 Table ) . Only two regions with low Tajima's D values were shared by two populations; one had no annotated gene and the other had a gene coding for a hypothetical protein ( S3 Table ) . Similarly , the genomic regions with the 50 highest Tajima’s D values within each population ( cut-off values: 2 . 024 in BRA , 1 . 890 in PER , 2 . 078 in COL , and 2 . 407 in MEX ) also comprised several hypothetical and housekeeping genes and a few surface antigen genes ( sera , circumsporozoite protein [csp] , and msp-3 ) that may be under balancing selection ( S4 Table ) . However , different genomic regions giving top 50 Tajima’s D values were typically found in each population and only 11 of them ( none comprising antigen-coding genes ) were shared by two or more populations . Whether the high Tajima’s D values found in certain domains are due to random effects of the parasite's demographic history or to balancing selection on specific genes remains to be further examined in larger population samples . MAF distributions were L-shaped in all New World P . vivax populations ( S5 Fig ) , similar to patterns described for P . falciparum populations from sub-Saharan Africa [45 , 46] . The proportion of SNPs with allele frequencies ≤ 0 . 1 were 51 . 6% in BRA , 66 . 3% in PER , 71 . 5% in COL , and 55 . 1% in MEX . Despite the relatively small sample sizes , we interpret the clear predominance of negative Tajima’s D values in PER and COL and of rare alleles in all populations , but mainly in PER and COL , as suggestive of a recent P . vivax population expansion in the Americas . Data from an extensive mitochondrial genome analysis of local parasites are also consistent with the demographic expansion hypothesis [47] . P . vivax genomic sequences from the New World were previously shown to cluster mostly according to their geographic origins , with the majority of MEX samples ( that show more extensive evidence of identity by descent ) clustering together in PCA plots [11] . These findings are particularly relevant to malaria-eliminating countries in the continent; if parasites can be assigned by molecular genotyping to their countries of origin , locally acquired and imported infections can theoretically be easily distinguished . Moreover , genetic analysis can also help to determine the likely origin of imported infections [10 , 48] . Accordingly , our PCA data revealed a clear clustering of BRA samples , which were separated from other populations by principal components C1 and C3 when the complete SNP data set was used . MEX was separated from other populations mostly by C1 and C2 , while PER and COL were less clearly differentiated ( Fig 4A ) . Moreover , a rather similar sample clustering pattern was revealed by a standard , neighbor-joining phylogenetic analysis ( Fig 5 ) . All BRA samples and most ( 16 of 19 ) MEX samples formed well-supported clades , while PER and COL isolates spread across single- and multiple-country clades . Similar to PCA and phylogenetic analysis , the model-based clustering approach implemented in ADMIXTURE software [33] also defined BRA and MEX as distinct populations ( S6A Fig , number of clusters K = 3 ) based on the whole genomic data set . PER and COL were nearly indistinguishable . A similar analysis with 2 clusters did not differentiate MEX from COL ( S6A Fig , number of clusters K = 2 ) . ADMIXTURE analysis using a curtailed data set ( 12 , 762 unlinked SNPs ) allowed a poor differentiation between populations ( S6B Fig , number of clusters K = 2 or 3 ) . Not surprisingly , the most geographically homogeneous samples ( BRA and MEX ) formed well-defined clusters . We next tested whether a recently described P . vivax SNP barcode [49] would be able to correctly assign New World isolates to their countries of origin . Of the 42 SNPs originally included in the barcode , five did not segregate in our populations . PCA with the remaining 37 SNPs was unable to cluster our parasite populations by their countries of origin ( Fig 4B ) . The overall pairwise differentiation between populations , estimated using FST averaged across the entire genome , was directly proportional to the geographic distance between sites , being lowest between COL and PER and highest between BRA and MEX ( Table 2 ) . Among the 100 SNPs with the highest average pairwise FST estimates , only two mapped to antigen-coding genes ( both to sera gene paralogs on chromosome 4 ) ; all others were either noncoding SNPs or mapped to genes encoding hypothetical or housekeeping proteins ( S5 Table ) . We next tested whether these 100 SNPs with the top FST values could separate New World populations more efficiently than the complete genome-wide SNP set ( compare Fig 4A and 4C ) and the currently available 42-SNP P . vivax barcode [49] ( compare Fig 4B and 4C ) . These results suggest that regional SNP barcodes could be further explored to track the geographic origin of P . vivax samples in the Americas . Our genomic sequence data enabled the identification of SNPs mapping to several P . vivax genes known or suspected to be involved in antimalarial drug resistance ( Table 3 ) . We found nonsynonymous mutations at the dihydrofolate reductase-thymidylate synthase ( dhfr ) locus previously associated with pyrimethamine resistance ( S58R and S117N ) almost exclusively in COL ( see also [35] ) and PER ( see also [37] ) , whereas antifolates are no longer recommended in these countries; two previously undescribed SNPs ( H99N and H99R ) were also found in COL . The dihydropteroate synthase ( dhps ) SNP A383G , associated with sulfadoxine resistance , was also observed in COL and PER ( see also [35] ) , with the M205I change being found only in PER ( see also [36] ) . Little or no polymorphism in dhfr and dhps genes was observed in BRA and MEX , likely reflecting differences in sulfadoxine-pyrimethamine use across countries in the region . Although pyrimethamine and sulfadoxine has never been recommended as a treatment for P . vivax infection , they may have been used to treat patients co-infected with P . falciparum and P . vivax , exposing the latter parasite to these drugs and likely selecting for resistant phenotypes . Moreover , the frequent use of sulfonamides to treat co-occurring , unrelated ( mostly bacterial ) infections may have selected sulfa-resistant strains in malaria parasite carriers from some of these countries . Non-synonymous substitutions were also characterized in the multidrug resistance 1 protein ( mdr1 ) gene , including two previously undescribed ( L186W and F1070L ) and six previously reported SNPs ( V221L , M908L , T958M , Y976F , and F1076L [50] ) . The Y976F change , originally believed to confer chloroquine resistance ( CQR ) in Southeast Asia and Melanesia [24] , occurred in a single BRA sample whose CQR phenotype could not be determined ( Table 2 ) . Interestingly , we found no coding changes in the chloroquine resistance transporter ortholog ( crt-o ) gene ( PVX 087980 ) , whose ortholog in P . falciparum displays a key non-synonymous substitution leading to CQR [51] . Five of 12 mutations found at the multidrug resistance associated protein 1 ( mrp1 [PVX 097025] ) locus , which codes for an ATP-cassette binding ( ABC ) transporter putatively involved in antimalarial drug efflux [52] , had been described in a clinical sample from Peru ( L1282I , Y1393D , G1419A , V1478I , and H1586I [36] ) . Similarly , six of 11 had been previously described in Colombia ( Q1419E , P1196S , A1106S , V1022M , S681I , R294M [35] ) and six in Peru ( S1701L , T1698K , Q1419E , A1106S , V1022M , and R294M [37] ) , in multidrug resistance associated protein 2 gene ( PVX_124085 ) . The antimalarial drug-resistance phenotype ( if any ) associated with these mutations remains undetermined . Here we found a sharp LD decline within 100 bp of distance between pairs of SNPs in BRA and MEX , with very low LD in COL and negligible LD in PER ( Fig 6 ) . Interestingly , the LD decay in BRA and MEX populations , originating from the sites with the lowest malaria transmission [53] , was even faster than that described for P . vivax populations from Southeast Asia , where P . vivax endemicity is substantially higher [14] , and that described for P . falciparum populations from areas of much higher endemicity in Sub-Saharan Africa [54 , 55 , 56] and Southeast Asia [56] . These results are consistent with a very high outcrossing rate in P . vivax populations and imply that SNP array-based genome-wide association studies ( GWAS ) would require a very high marker density ( inter-marker distance < 100 bp ) to help find genetic loci that determine phenotypes of interest , such as drug resistance and virulence , in BRA and MEX populations . SNP-based association studies are unlikely to be successful in PER and COL populations , given the negligible LD . In contrast , P . falciparum populations from the Pacific Coast of Colombia showed a much more gradual LD decline , reaching background LD levels only at an inter-marker distance of 240 kb [57] . We found high genome-wide diversity and relatively little geographic structure in P . vivax populations from areas of relatively low malaria transmission in the Americas . These findings contrast with the low genetic diversity and clear subdivision into local , highly divergent populations that is typical of New World populations of P . falciparum [57–61] . Interestingly , local P . vivax populations are nearly as polymorphic as their P . falciparum counterparts from Africa [45 , 46 , 54] and their P . vivax counterparts from Southeast Asia [11 , 14] . Moreover , the low levels of between-population differentiation in P . vivax from the Americas are reminiscent of those in P . falciparum populations from hyperendemic sub-Saharan Africa [45 , 46 , 59 , 62] . Demographic history might explain , at least in part , the observed differences between P . vivax and P . falciparum populations from the New World . The original diversity of the ancestral P . falciparum population may have been partially lost during ( a ) its migration to the Americas in post-Columbian times [63] followed by the adaptation to entirely different local vectors [64] , ( b ) recent selective sweeps driven by antimalarial drugs , such as chloroquine [65] and pyrimethamine [66] , and ( c ) local extinctions and clonal expansions following eradication attempts in the Americas between the mid-1950s and late 1960s [53 , 60] . These events would have resulted in small and fragmented , mostly inbred P . falciparum populations being scattered throughout the region . The significantly greater genome-wide diversity currently found in P . vivax populations worldwide , compared with P . falciparum [11 , 67] , suggests that ( a ) this species retained more ancestral diversity than P . falciparum when transferred from apes to humans and following subsequent spread out of Africa and/or ( b ) P . vivax is simply older than P . falciparum and therefore has accumulated more mutations over time . Moreover , these species might also differ in the ways they colonized the New World . The high diversity of present-day P . vivax lineages across this continent is consistent with an admixture of parasite lineages originating from geographically diverse regions [5 , 7 , 47] , including now extinct European lineages [12] . The success of colonization events and the recombination between these introduced strains may have been favored by the ability of P . vivax to stay dormant in the liver as hypnozoites , prolonging the duration of parasite carriage in human hosts and increasing the probability of superinfections leading to the co-occurrence of genetically diverse clones that may recombine once transmitted to mosquito vectors . Moreover , genome-wide P . vivax diversity in the region appears to have been little affected by recent selective sweeps driven by pyrimethamine [35 , 37] in Brazil and Colombia . Interestingly , P . vivax resistance to chloroquine remains relatively infrequent in the New World [68] and is unlikely to have induced a major selective sweep in local parasites . Improved methods for clinical sample preparation and next-generation sequencing now enable further genome-wide analyses of additional P . vivax isolates from the New World , to test the demographic scenarios suggested by our data . | Plasmodium vivax is the most common human malaria parasite in the Americas , but how and when this species arrived in the New World remains unclear . Here we describe high-quality whole-genome sequence data for nine P . vivax isolates from Brazil , a country that accounts for 37% of the malaria burden in this continent , and compare these data with additional publicly available P . vivax genomes from Brazil , Peru , Colombia , and Mexico . P . vivax populations from the New World were found to be as diverse as their counterparts from areas with substantially higher malaria transmission , such as Southeast Asia , and to carry several non-synonymous substitutions in candidate drug-resistance genes . Moreover , genome-wide patterns of linkage disequilibrium between pairs of polymorphic sites are consistent with very frequent outcrossing in these populations . Interestingly , local P . vivax is more polymorphic , with less between-population differentiation , than sympatric populations of P . falciparum , possibly as a result of different demographic histories of these two species in the Americas . We hypothesize that local P . vivax lineages originated from successive migratory waves and subsequent admixture between parasites from geographically diverse sites . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"and",
"discussion"
]
| [
"parasite",
"groups",
"medicine",
"and",
"health",
"sciences",
"plasmodium",
"population",
"genetics",
"geographical",
"locations",
"microbiology",
"tropical",
"diseases",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"parasitology",
"peru",
"apicomplexa",
"protozoans",
"pharmacology",
"population",
"biology",
"antimicrobial",
"resistance",
"malarial",
"parasites",
"south",
"america",
"brazil",
"people",
"and",
"places",
"colombia",
"genetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"malaria",
"evolutionary",
"biology",
"organisms"
]
| 2017 | Genome-wide diversity and differentiation in New World populations of the human malaria parasite Plasmodium vivax |
The asymmetrically dividing yeast S . cerevisiae assembles a bipolar spindle well after establishing the future site of cell division ( i . e . , the bud neck ) and the division axis ( i . e . , the mother-bud axis ) . A surveillance mechanism called spindle position checkpoint ( SPOC ) delays mitotic exit and cytokinesis until the spindle is properly positioned relative to the mother-bud axis , thereby ensuring the correct ploidy of the progeny . SPOC relies on the heterodimeric GTPase-activating protein Bub2/Bfa1 that inhibits the small GTPase Tem1 , in turn essential for activating the mitotic exit network ( MEN ) kinase cascade and cytokinesis . The Bub2/Bfa1 GAP and the Tem1 GTPase form a complex at spindle poles that undergoes a remarkable asymmetry during mitosis when the spindle is properly positioned , with the complex accumulating on the bud-directed old spindle pole . In contrast , the complex remains symmetrically localized on both poles of misaligned spindles . The mechanism driving asymmetry of Bub2/Bfa1/Tem1 in mitosis is unclear . Furthermore , whether asymmetry is involved in timely mitotic exit is controversial . We investigated the mechanism by which the GAP Bub2/Bfa1 controls GTP hydrolysis on Tem1 and generated a series of mutants leading to constitutive Tem1 activation . These mutants are SPOC-defective and invariably lead to symmetrical localization of Bub2/Bfa1/Tem1 at spindle poles , indicating that GTP hydrolysis is essential for asymmetry . Constitutive tethering of Bub2 or Bfa1 to both spindle poles impairs SPOC response but does not impair mitotic exit . Rather , it facilitates mitotic exit of MEN mutants , likely by increasing the residence time of Tem1 at spindle poles where it gets active . Surprisingly , all mutant or chimeric proteins leading to symmetrical localization of Bub2/Bfa1/Tem1 lead to increased symmetry at spindle poles of the Kar9 protein that mediates spindle positioning and cause spindle misalignment . Thus , asymmetry of the Bub2/Bfa1/Tem1 complex is crucial to control Kar9 distribution and spindle positioning during mitosis .
Asymmetric cell division generates two daughter cells genetically identical but that differ in fate and/or in size and cytoplasmic material . During asymmetric cell division , polarity factors are first concentrated to specific locations to define the poles of cell division . Afterwards the spindle orients according to these polarity cues to segregate one set of chromosomes towards a given polarity determinant and the other away from it , thereby generating two unequal daughter cells ( reviewed in [1–3] ) . Correct spindle positioning is therefore critical to preserve the right lineage of asymmetrically dividing cells . Accordingly , spindle mispositioning in asymmetrically dividing stem cells , which normally generate one daughter stem cell with self-renewal potential and one cell destined to differentiation , steers tumourigenesis by increasing the pool of undifferentiated stem cells [4 , 5] . Surveillance mechanisms , or checkpoints , must therefore respond to spindle positioning errors and delay cell cycle progression until the mitotic spindle is properly oriented with respect to the cell polarity axis [6 , 7] . The budding yeast Saccharomyces cerevisiae is a widely recognized model system to study asymmetric cell division . Spindle positioning in budding yeast requires either one of two redundant pathways , one that depends on the APC ( Adenomatous Polyposis Coli ) -related protein Kar9 , and the other on dynein ( reviewed in [8] ) . Spindle positioning errors are monitored by a surveillance mechanism , referred to as spindle position checkpoint ( SPOC ) , that delays mitotic exit and cytokinesis to provide the time for proper spindle realignment ( reviewed in [6 , 9] ) . The target of the SPOC is a small GTPase called Tem1 , which acts as molecular switch for the activation of a kinase cascade related to the Hippo pathway and named Mitotic Exit Network ( MEN ) . In the fission yeast S . pombe a kinase cascade similar to MEN and referred to as Septation Initiation Network ( SIN ) triggers cytokinesis [10] . The MEN effector of Tem1 is the kinase Cdc15 , which in turn promotes the activation of the downstream Mob1/Dbf2 kinase complex that ultimately leads to activation of the Cdc14 phosphatase [11] . Cdc14 is the main phosphatase that in budding yeast counteracts the activity of cyclin-dependent kinases ( CDKs ) , and it is essential for mitotic exit and cytokinesis by dephosphorylating CDK substrates , as well as by triggering inactivation of mitotic CDKs [12] . Cdc14 is sequestered in the nucleolus in an inactive form throughout most of the cell cycle , until it is released and activated . Although the MEN is necessary for the full release of Cdc14 into the cytoplasm to promote mitotic exit [13 , 14] , another pathway called FEAR ( Cdc Fourteen Early Anaphase Release ) causes a partial release of Cdc14 from the nucleolus into the nucleus at the metaphase-to-anaphase transition [15] . The FEAR pathway involves the polo kinase Cdc5 , the redundant Spo12 and Bns1 proteins , and the separase Esp1 , which by inhibiting the phosphatase PP2ACdc55 allows the dissociation of Cdc14 from its nucleolar inhibitor Net1 [15 , 16] . The FEAR-mediated activation of Cdc14 in anaphase is thought to regulate spindle dynamics and to contribute to timely activation of the MEN ( reviewed in [17] ) . Recent data have shown that MEN is not only important for triggering mitotic exit in telophase , but also has an earlier function in metaphase to promote correct spindle positioning along the polarity axis [18] . In most MEN mutants , except for cdc14 , spindles are indeed misoriented relative to the cell division plane . Notably , the Mob1/Dbf2 kinase was found to phosphorylate the spindle positioning Kar9 protein , thereby favouring its concentration on astral microtubules emanating from only one of the two spindle poles [18] . Asymmetric distribution of Kar9 at spindle poles in metaphase is in turn crucial for proper spindle positioning because it targets the Kar9-decorated aster to the bud , due to Kar9 interaction with the type V myosin Myo2 [19] . The yeast centrosomes , named spindle pole bodies ( SPBs ) , play an important role in the regulation of mitotic exit , as they act as a scaffold for MEN components , such as Tem1 and its downstream kinases ( reviewed in [20] ) . The constitutive SPB component Nud1 recruits MEN proteins to SPBs and is essential for mitotic exit [21] , suggesting that binding of one or several MEN factors to SPBs is required for mitotic exit . Consistently , Tem1 association to SPBs is critical for MEN activation [22] . Like all GTPases , Tem1 is active when bound to GTP and inactive in its GDP-bound form . The common element of the GTPase superfamily is the 160–180 residue G domain involved in nucleotide binding [23] . Within the G domain two flexible “switch regions” , referred to as Switch I and II undergo the most dramatic structural rearrangement upon GTP hydrolysis and therefore define the major conformational changes conferred by GTP versus GDP binding [24] . On the basis of sequence alignment with human Ras , Switch I and II in Tem1 correspond to residues 50–55 and 77–84 , respectively . Upon spindle misalignment the two-component GTPase-activating protein ( GAP ) Bub2/Bfa1 inactivates Tem1 by stimulating GTP hydrolysis [25 , 26] . GTPase-activating proteins accelerate GTP hydrolysis , promoting the GDP-bound inactive form of GTPases [27 , 28] . The GAP activity of the Bub2/Bfa1 complex resides on Bub2 , which carries a TBC domain ( Tre-2 , Bub2 and Cdc16; [29] ) , whereas Bfa1 mediates Bub2 interaction with Tem1 and prevents Tem1 dissociation from guanine nucleotides , thereby acting as guanine dissociation inhibitor ( GDI ) [25 , 26 , 30 , 31] . Often , the release of GDP from GTPases is a slow and thermodynamically unfavourable reaction . This is why GTPase activation requires in most cases the intervention of nucleotide exchange factors ( GEFs ) that catalyse the release of GDP , promoting its replacement by GTP [27] . The identity of the GEF ( s ) for Tem1 , if any , remains elusive . The early proposal based on genetic data that the Lte1 protein might be the GEF for Tem1 has not been confirmed by biochemical assays [30] . Therefore , if inactivation of a GEF for Tem1 could play any role in the SPOC , besides Tem1 inhibition by the GAP Bub2/Bfa1 , remains to be established . The Kin4 protein kinase is a key component of the SPOC ( reviewed in [6] ) . During spindle misalignment it phosphorylates Bfa1 , thereby preventing the inhibitory phosphorylation of the GAP Bub2/Bfa1 by the polo kinase [32] . During the unperturbed cell cycle Kin4 is strategically restricted to the mother cell compartment , where it is thought to sense the anomalous persistence of both SPBs in anaphase . In addition , it is present on both SPBs of misaligned spindles [32–35] . Localization of MEN and SPOC proteins to the SPBs changes during the cell cycle , in that some proteins ( like Cdc15 , Mob1 and Dbf2 ) are loaded on both SPBs at the anaphase onset , whereas other proteins ( like Tem1 , Bfa1 and Bub2 ) are present on both SPBs already in metaphase and their localization becomes much more asymmetric in anaphase , when they preferentially accumulate on the old , bud-directed SPB [36–44] . Moreover , the position of the spindle seems to play a role in controlling the asymmetric localization of Tem1 , Bub2 and Bfa1 in anaphase . Indeed , these proteins localize less strongly but more symmetrically on the two SPBs when a misoriented spindle elongates within the mother cell and the SPOC turns on [36 , 38 , 40] . Whether the asymmetric localization of Tem1 or its GAP is important for triggering MEN signalling remains to be elucidated . Remarkably , the SIN counterparts of several MEN components also localize asymmetrically on SPBs during anaphase , with the homologs of the GAP components Bub2 ( Cdc16 ) and Bfa1 ( Byr4 ) occupying one SPB , while the GTP-bound form of the GTPase Spg1 and its effector kinase Cdc7 occupy the other [45 , 46] . Thus , in spite of the different modes of cell division in the two yeasts ( asymmetric in S . cerevisiae and symmetric in S . pombe ) MEN and SIN signalling is asymmetric in both . SIN asymmetry has been proposed to be crucial for timely cytokinesis , as cdc16 and byr4 mutants where Spg1 and Cdc7 are symmetric in anaphase undergo multiple rounds of septation [46 , 47] . In S . cerevisiae , Kin4 was found to increase the turnover of the Bub2/Bfa1 at SPBs . However , Kin4 does not promote asymmetric localization of Bub2/Bfa1 on properly oriented spindles [36] . Chimeric proteins obtained by fusing Bub2 or Bfa1 to the structural SPB component Cnm67 cause unscheduled mitotic exit in the presence of mispositioned spindles , which led to the proposal that high turnover of Bub2/Bfa1 might be important to inhibit Tem1 in the cytoplasm during SPOC activation [36] . Conversely , a modified version of Bub2 carrying 9 myc epitopes at the C terminus localizes the GAP and Tem1 rather symmetrically at SPBs and prevents mitotic exit in some sensitized MEN mutant backgrounds [25] . In order to shed light onto the relationship between Bub2/Bfa1 symmetry and SPOC response , we report the characterization of a series of mutants altering either the Bub2/Bfa1 subunits or Tem1 and causing symmetric localization of Tem1 and its GAP during properly oriented anaphase . Remarkably , these mutant proteins as a whole tend to activate , rather than inhibit , the MEN . In addition , they lead to more symmetric distribution of Kar9 on spindle poles and to spindle positioning defects , indicating that a delicate balance between MEN activation and inactivation is required for proper spindle alignment .
The catalytic mechanism of GTPase-activating proteins ( GAPs ) requires an ‘arginine-finger’ , where the lateral chain of a conserved arginine ( R85 for Bub2 , [25 , 29] ) , interacts with the nucleotide-binding site of a G protein , thus stimulating hydrolysis of the γ−phosphate . A new catalytic mechanism , called “dual finger” , was proposed for the family of GAPs with TBC ( Tre-2 , Bub2 and Cdc16 ) domain . According to the dual finger mechanism a conserved glutamine residue contributes to stimulate GTP hydrolysis together with the canonical catalytic arginine [48] . To investigate if Bub2 acts indeed via a dual finger mechanism , we generated a mutant Bub2 variant , Bub2-Q132L , where we replaced by leucine the conserved glutamine at position 132 that identifies the glutamine finger on the basis of sequence alignment [48] . Bacterially purified His-tagged Tem1 , Maltose Binding Protein ( MBP ) -tagged Bfa1 and glutathione-S transferase ( GST ) -tagged Bub2 or Bub2-Q132L proteins were used in in vitro GTPase assays , as previously described [25 , 26] . The rate of GTP hydrolysis and dissociation was measured using Tem1 bound to γ[32P]-GTP , whereas the rate of GTP dissociation alone was measured using Tem1 bound to the non-hydrolysable GTP analogue γ[35S]-GTP ( Fig . 1A ) . As shown in Fig . 1A , the kinetics of radioactivity loss from wild type Tem1 loaded with either γ[32P]-GTP or γ[35S]-GTP were very similar , suggesting that Tem1 on its own mostly dissociates GTP without hydrolysing it . The presence of Bfa1 stabilized Tem1 in the GTP-bound form ( Fig . 1B ) , whereas Bub2 stimulated Tem1 GTPase activity in the presence of Bfa1 ( Fig . 1C ) , but not GTP dissociation ( Fig . 1B ) . We then compared the GAP activity of purified GST-Bub2 and GST-Bub2-Q132L . Interestingly , Bub2-Q132L did not display any GAP activity towards GTP-bound Tem1 ( Fig . 1C ) , behaving as the GAP–dead mutant Bub2-R85A previously characterized [25] . Furthermore , it did not stimulate GTP dissociation , exactly like wild type Bub2 ( Fig . 1B ) . In vivo , the Q132L substitution completely abolished the checkpoint function of Bub2 . Indeed , similar to bub2Δ cells , bub2-Q132L cells escaped the mitotic arrest upon nocodazole treatment , as indicated by their ability to re-replicate their chromosomes ( Fig . 1D ) . Checkpoint response to spindle misalignment was also impaired in bub2-Q132L cells . Indeed , when spindle mispositioning was induced by DYN1 or KAR9 deletion [49 , 50] bub2-Q132L cells did not arrest in mitosis as large budded cells but re-budded , similar to bub2Δ cells ( Fig . 1E ) . Thus , consistent with the proposed model [48] , Bub2 GAP activity , and thereby its role in the SPOC , relies on a dual finger mechanism involving two catalytic residues , R85 and Q132 . The bub2-Q132L allele did not accelerate mitotic exit during the unperturbed cell cycle . Indeed , synchronized bub2-Q132L cells could divide and disassemble bipolar spindles with wild type kinetics ( Fig . 1F–G ) . Furthermore , kinetics of degradation of the main mitotic cyclin Clb2 were very similar in wild type and bub2-Q132L cells ( Fig . 1H ) . Interestingly , we found that cell cycle-dependent phosphorylation of Bfa1 , which promotes mitotic exit [51] , was abolished in bub2-Q132L cells , in agreement with the recent proposal that it requires Bub2 activity [52] . We then asked if Bub2-Q132L could still interact efficiently with Bfa1 and Tem1 . Immunoprecipitations of Bub2 or Bub2-Q132L tagged with three HA epitopes showed that both proteins pulled down roughly the same amounts of GFP-tagged Bfa1 . In contrast , Bub2-Q132L precipitated a higher amount of GFP-tagged Tem1 than wild type Bub2 ( Fig . 1I ) , suggesting that abolishing the GAP catalytic activity of the Bub2/Bfa1 complex stabilizes the interaction between Tem1 and its GAP . Because lack of Bub2 GAP activity through the bub2-R85A allele leads to increased symmetric localization of Bub2 to SPBs in anaphase [25] , we analyzed the subcellular distribution of eGFP-tagged Bub2-Q132L . In contrast to wild type Bub2 , which was almost exclusively present on the bud-directed SPB in 84% of anaphase cells , Bub2-Q132L-HA3 was found on both SPBs in 97% of cells in anaphase . In addition , Bfa1 and Tem1 were also more symmetrically localized on the SPBs of bub2-Q132L cells than they were in wild type cells during the same cell cycle stage ( Fig . 1J ) . We therefore conclude that , consistent with previous data , interfering with Bub2 GAP activity affects the asymmetry of the Tem1/Bub2/Bfa1 complex on anaphase spindle poles . Finally , we analysed the localization of the Tem1 effector kinase Cdc15 in bub2-Q132L cells . We found GFP-tagged Cdc15 on the SPBs of metaphase spindles in 55% of the cells . Deletion of BUB2 or its replacement with the bub2-Q132L allele increased both the total percentage of cells with Cdc15 at SPBs ( 83% and 80% , respectively , Fig . 1K ) and the percentage of cells with symmetrically localized Cdc15 in metaphase ( 36% in bub2Δ and bub2-Q132L cells versus 11% of wild type cells ) . Thus , lack of Bub2 GAP activity leads to more efficient recruitment of Cdc15 at spindle poles . Since spindle misalignment leads to persistent residence of Bub2/Bfa1 on both SPBs [40] , we and others proposed that symmetric distribution of the GAP complex might lead to inhibition of Tem1 [25 , 40] . This idea was further supported by our previous finding that a myc-tagged variant of Bub2 ( Bub2-myc9 ) localizing mostly symmetrically on SPBs was lethal and prevented mitotic exit in sensitized backgrounds [25] . On the other hand , the Bub2/Bfa1 complex is required throughout most of the cell cycle for Tem1 association with SPB , which in turn triggers its activation [22 , 40] . To assess the importance of Bub2/Bfa1 asymmetry at SPBs , we tethered Bfa1 or Bub2 to both SPBs by fusing them to the structural SPB component Spc72 . We confirmed that in about 90% of the cells the Spc72-Bfa1 chimeric protein localised constitutively to the SPBs throughout the cell cycle ( Fig . 2A ) and was able to recruit Tem1 to both SPBs in 74% of anaphase cells , as opposed to 35% of wild type cells ( Fig . 2B ) . Both Spc72-Bfa1 and Spc72-Bub2 chimeric proteins were functional based on their ability to complement lack of endogenous BFA1 or BUB2 , respectively , for what concerns the checkpoint response to microtubule depolymerisation . Indeed , in the presence of nocodazole , SPC72-BFA1 and SPC72-BUB2 cells arrested in mitosis with 2C DNA contents as well as wild type cells , whereas bub2Δ and bfa1Δ cells re-replicated their genome in the same conditions ( Fig . 2C-D ) . Thus , the Spc72-Bfa1 and-Bub2 chimera are likely functional in that they retain their inhibitory properties towards Tem1 . Previously characterized Bub2 and Bfa1 chimeric proteins constitutively anchored to SPBs are SPOC-defective [36] . Similarly , our Spc72-Bfa1 and Spc72-Bub2 failed to activate the SPOC upon spindle mispositioning caused by DYN1 deletion ( Fig . 2E ) . Indeed , dyn1Δ SPC72-BFA1 bfa1Δcells undergoing anaphase in the mother cell , which is symptomatic of spindle mispositioning , exited the cell cycle and re-budded , in contrast to dyn1Δ cells that arrested in mitosis as large budded cells ( Fig . 2E ) . The SPOC failure of SPC72-BFA1 bfa1Δ cells was not worsened by deletion of BUB2 or KIN4 or both , consistent with the notion that Kin4 and Bub2/Bfa1 act in concert to inhibit Tem1 . Similar results were obtained with the Spc72-Bub2chimera ( S1 Fig . ) . Thus , constitutive targeting to both SPBs of the GAP Bub2/Bfa1 , and of Tem1 as a consequence , leads to unscheduled Tem1 activation , consistent with a previous proposal [22] . We then asked if symmetric localization of Tem1 driven by the Spc72-Bfa1 chimera leads to more efficient recruitment of Cdc15 to SPBs in metaphase . This was indeed the case . Whereas GFP-tagged Cdc15 was present at the SPBs of 55% wild type metaphase cells , 90% of metaphase cells expressing the fusion SPC72-BFA1 displayed SPB-bound Cdc15 ( Fig . 2F ) . Furthermore , Cdc15 was significantly more symmetric in SPC72-BFA1 than in wild type cells . Thus , stable tethering of Tem1 to SPBs by fusion to an SPB component [22] or by SPB recruitment via its inhibitory GAP ( [36] and our data ) leads in both cases to premature Tem1 activation . We then asked if expression of the Spc72-Bfa1 chimera could have any phenotypic consequence for conditional mutants affecting the MEN . Remarkably , SPC72-BFA1 as the only source of Bfa1 in the cells suppressed the growth defects of several MEN mutants . In particular , it could partially rescue the temperature sensitivity of tem1–3 and mob1–77 ( Fig . 2G ) , as well as the cold-sensitivity of cdc15–2 and dbf2–2 mutant cells ( S2A Fig . ) . A slight suppression , if any , was observed for the temperature-sensitivity of cdc5–2 cells , whereas the temperature-sensitivity of cdc14–3 cells was not suppressed at all ( S2B-C Fig . ) . Suppression of tem1–3 was recessive , as it was not observed when a wild type copy of BFA1 was present concomitant to SPC72-BFA1 in the cells ( Fig . 2G ) . Importantly , suppression was not due to reduced GAP activity , as it could not be recapitulated by deletion of BFA1 ( Fig . 2G ) , or BUB2 or both ( S2D Fig . ) . Since the mitotic exit defects of tem1–3 cells at high temperatures correlate with a loose interaction of the mutant Tem1 protein with SPBs [53] , we conclude that Spc72-Bfa1 suppresses the temperature-sensitivity of tem1–3 cells likely by recruiting Tem1 to the SPBs . Consistent with this notion , the Spc72-Bfa1 and Spc72-Bub2 chimera suppressed the lethality caused by overexpression of the SPOC kinase Kin4 ( Fig . 2H ) , which increases the turnover of Bub2/Bfa1 , and by consequence of Tem1 , at SPBs [36] . Thus , these data , together with those from a previous study [36] , indicate that symmetric persistence of Bub2/Bfa1 at SPBs does not interfere with mitotic exit . Rather , in spite of being part of an inhibitory GAP complex , Bfa1 is a receptor of Tem1 at SPBs , where Tem1 promotes MEN signaling . Stable residence of Bub2/Bfa1 at SPBs causes unscheduled mitotic exit by decreasing Tem1 turnover at SPBs , as previously suggested [22] . The ability of Spc72-Bfa1 and-Bub2 tethers to efficiently prevent mitotic exit upon microtubule depolymerization , but not upon spindle mispositioning , was somewhat puzzling . A major difference between the two conditions lies in the activation of the spindle assembly checkpoint ( SAC ) after nocodazole treatment . Through inhibition of Cdc20/APC , SAC leads to securin stabilization , in turn preventing activation of separase and the FEAR pathway [15] . If inhibition of the FEAR pathway is the only reason for the failure of Spc72-Bfa1 and-Bub2 chimera to promote mitotic exit , premature FEAR activation should allow mitotic exit in cells expressing Spc72-Bfa1 and-Bub2 treated with nocodazole . Conversely , FEAR inactivation should prevent mitotic exit in the same cells undergoing spindle misalignment . To test this hypothesis , we prematurely activated the FEAR pathway and Cdc14 release from the nucleolus by either inactivation of the PP2ACdc55 phosphatase or ESP1 overexpression [16] . Remarkably , PP2ACdc55 inactivation through deletion of CDC55 had a synergistic effect with SPC72-BFA1 and SPC72-BUB2 on the kinetics of mitotic exit upon nocodazole treatment , as judged by the ability of cells to re-replicate their DNA ( Fig . 3A-B ) . Similar data were obtained with ESP1 overexpression from the galactose-inducible GAL1 promoter ( Fig . 3C-D ) . In contrast , reducing the levels of mitotic CDKs through CLB2 deletion did not accelerate mitotic exit in SPC72-BFA1 cells ( Fig . 3E ) . Importantly , FEAR inactivation through deletion of both SPO12 and BNS1 reduced the unscheduled mitotic exit caused by SPC72-BFA1 in dyn1Δ cells ( Fig . 3F ) . Thus , constitutive recruitment of the Bub2/Bfa1 complex to SPBs leads to precocious Tem1 activation . Whether this translates into a premature mitotic exit depends on the activation state of the FEAR pathway . To further investigate the links between Tem1 activity and the establishment of SPB asymmetry of the Tem1/Bub2/Bfa1 complex , we generated a TEM1-Q79L mutant allele , where the catalytic glutamine in the G domain ( Q79 , according to sequence comparison with Rab-like GTPases [27] ) , was replaced by leucine . We first tested the catalytic properties of Tem1-Q79L in in vitro GTPase assays in the presence of Bfa1 and Bub2 . As shown in Fig . 4A , Tem1-Q79L was completely refractory to stimulation of GTP hydrolysis by the GAP Bub2/Bfa1 in vitro , suggesting that in vivo it is preferentially in its active GTP-bound form . Thus , Q79 of Tem1 likely participates directly to GTP hydrolysis along with R85 and Q132 of Bub2 . When expressed in yeast cells as the sole source of Tem1 , Tem1-Q79L did not cause any detectable growth defect at any temperature . Furthermore , TEM1-Q79L mutant cells showed kinetics of cell cycle progression similar to those observed in wild type cells ( Fig . 4B ) . The absence of obvious cell cycle phenotypes in unperturbed conditions was somewhat surprising , as a similar mutation in fission yeast spg1+ , encoding the SIN counterpart of the Tem1 GTPase , leads to premature cytokinesis and formation of multiple septa [54] . We therefore analysed by time-lapse video microscopy the speed of actomyosin ring contraction , as a marker of cytokinesis [55] , in wild type and TEM1-Q79L cells expressing GFP-tagged myosin II ( Myo1 ) . Strikingly , contraction of the actomyosin ring took place on average 2’ faster in TEM1-Q79L relative to wild type cells ( i . e . 6’ and 8’ , respectively , Fig . 4C ) , consistently with previous data on bub2Δ cells [56] . Thus , the TEM1-Q79L allele accelerates at least some aspects of cytokinesis without affecting cell viability . To gain further insights into the factors allowing TEM1-Q79L cells to grow at normal rates , we carried out a synthetic genetic arrays ( SGA ) screen to find deletions of non-essential genes that become synthetically lethal/sick with TEM1-Q79L ( Table 1 ) . This screen uncovered several genes encoding proteins involved in spindle positioning and nuclear migration , such as Kar9 , the dynein light chain ( DYN2 ) and components of the dynactin complex that cooperates with dynein for spindle positioning [57] . In addition , this screen uncovered several genes implicated in microtubule biogenesis , which might indirectly influence spindle positioning . Thus , TEM1-Q79L aggravates the sickness of cells undergoing spindle mispositioning . Interestingly , the same deletion mutants were also identified in other SGA screens as synthetically sick or lethal with BUB2 or BFA1 deletion [58–60] . A number of additional non-essential genes whose deletion displayed synthetic interactions with TEM1-Q79L was also uncovered with this screen and will be described elsewhere , since the significance of these genetic interactions has not been further explored in this context . We then tested the ability of TEM1-Q79L mutant cells to respond to microtubule depolymerization and spindle mispositioning . In the presence of nocodazole , whereas wild type cells arrested in mitosis with 2C DNA contents , TEM1-Q79L cells re-replicated their genome similar to cells lacking Bub2 ( Fig . 4D ) . In addition , TEM1-Q79L dyn1Δ cells exited mitosis and re-budded in face of spindle position defects ( Fig . 4E ) . Thus , the TEM1-Q79L mutant allele affects SPOC response . As expected , the checkpoint defect was dominant , as the TEM1-Q79L allowed mitotic exit and re-replication in the presence of nocodazole even when expressed from an episomal plasmid in cells carrying also the endogenous TEM1 gene ( Fig . 4F ) . Consistent with its constitutive activation , the TEM1-Q79L allele was also able to suppress the lethality associated with overexpression of KIN4 from the galactose-inducible GAL1 promoter ( Fig . 4G ) , which delays mitotic exit by keeping the Bub2/Bfa1 GAP active [34] . The “dual finger” model predicts that GTP hydrolysis is catalysed by the GAP and the so-called catalytic glutamine of the GTPase could stabilize the interaction between the GAP and the GTPase without directly contributing to the catalytic reaction [48] . We formally tested this idea by analysing the interaction between Tem1 or Tem1-Q79L with Bub2 and Bfa1 in co-immunoprecipitation experiments . Remarkably , HA-tagged Tem1-Q79L pulled down a higher , rather than a lower , amount of Bub2 and Bfa1 ( tagged with 3PK epitopes and GFP , respectively ) ( Fig . 5A ) . Thus , constitutive binding to GTP seems to increase the affinity of Tem1 for its GAP . To further corroborate this conclusion we co-expressed Bfa1-GFP and Bub2-GFP with HA-tagged Tem1 or Tem1-Q79L . Immunoprecipitation of Tem1-Q79L-HA3 pulled down higher amounts of both Bfa1-GFP and Bub2-GFP than Tem1-HA3 ( Fig . 5B ) , without affecting the relative proportion of Bfa1-GFP and Bub2-GFP in the immunoprecipitates . We therefore conclude that locking Tem1 in the GTP-bound form enhances its affinity for Bub2/Bfa1 without affecting the stoichiometry of the GAP complex . Since our data suggest that proper regulation of Tem1 GTP hydrolysis is required for asymmetry of the Tem1/Bub2/Bfa1 complex at SPBs , we analysed the subcellular distribution of Tem1-Q79L tagged with eGFP . As expected , Tem1-eGFP was already asymmetric in 53% of metaphase cells , whereas Tem1-Q79L-eGFP was asymmetric in a lower fraction of cells ( 35% ) . In anaphase , whereas Tem1-eGFP localized asymmetrically to the bud-directed SPB in 90% of cells , Tem1-Q79L-eGFP was present symmetrically at SPBs in 70% of the cells ( Fig . 5C ) . Thus , abolishing the GAP-stimulated GTPase activity of Tem1 through different kinds of mutations invariably leads to Tem1 symmetric localization at SPBs . Interestingly , whereas BFA1 deletion markedly affected Tem1-eGFP recruitment to SPBs as previously reported [22 , 40 , 61] , it had a less pronounced effect on the SPB localization of Tem1-Q79L-eGFP ( Fig . 5D ) , suggesting that loading to SPBs of constitutively active Tem1 is partially GAP-independent . Previous data suggested that loading of Bfa1 and Bub2 on SPBs and their asymmetry in anaphase still occur in cells lacking Tem1 [30 , 40] . On the other hand , experimental evidence indicates that an increased residence time of Tem1 at the SPBs or its decreased GTPase activity can influence Bub2/Bfa1 localization [22 , 25 , 62] . Because our results indicate that the Q79L substitution affects Tem1 activity as well as its localization in anaphase , we asked if Tem1-Q79L had any impact on localization of Bfa1 . Both wild type and TEM1-Q79L mutant cells showed a similar partially asymmetrical SPB localization of Bfa1-eGFP in metaphase ( Fig . 5E ) . In contrast , at the onset of anaphase , while Bfa1 drastically dropped to hardly detectable levels on the mother-bound SPB in 95% of wild type cells , it remained completely symmetrical on SPBs in 22% of TEM1-Q79L cells and persisted to low but clearly detectable levels on the mother-bound SPB in 46 . 6% of the cells ( Fig . 5F ) . The symmetric localization of Bfa1 in TEM1-Q79L cells did not depend on premature activation of downstream MEN kinases , as it was not affected by Cdc15 inactivation through the cdc15–2 temperature-sensitive allele ( S3 Fig . ) . Since Kin4 promotes Bfa1 turnover at SPBs upon SPOC activation [36] , we analysed Bfa1 localization in wild type and TEM1-Q79L cells lacking KIN4 ( Fig . 5E-F ) . In agreement with previous data [36] , deletion of KIN4 alone did not affect Bfa1 distribution on SPBs of wild type cells in unperturbed conditions . In stark contrast , it had a synergistic impact with the TEM1-Q79L mutant allele on Bfa1 localization at SPBs specifically in anaphase , making it completely symmetrical in 65% of the cells and partially asymmetric in 30% of the cells ( Fig . 5F ) . Consistently , the ratio in Bfa1-GFP fluorescence intensity at the mother- versus the bud-directed SPB was close to 0 for wild type and kin4Δ cells , while it significantly increased in TEM1-Q79L and TEM1-Q79L kin4Δ cells ( Fig . 5G ) . Thus , these data reveal an unanticipated role of Kin4 in actively dislodging Bfa1 from the mother-bound SPB during anaphase of the unperturbed cell cycle . Furthermore , they indicate that Tem1 GTP hydrolysis is a primary determinant of Bfa1 asymmetry in anaphase . To investigate further if the TEM1-Q79L allele leads to premature MEN activation , we analysed the subcellular localization of downstream MEN components , such as Cdc15 and Mob1 . We observed that 99% of TEM1-Q79L mutant cells recruited Cdc15 to SPBs in metaphase , as opposed to 55% in wild type cells ( Fig . 6A ) . Furthermore , Cdc15 was significantly more symmetric in TEM1-Q79L than in wild type cells . Deletion of KIN4 , either alone or in combination with the TEM1-Q79L allele did not have any impact on Cdc15 distribution ( Fig . 6A ) . In spite of Cdc15 enhanced loading on SPBs , recruitment of Mob1 to SPBs in metaphase was only slightly increased in TEM1-Q79L relative to wild type cells ( Fig . 6B ) , consistent with the notion that mechanisms other than the SPOC restrain MEN activity downstream of Tem1 until late anaphase . Phosphorylation of Cdc15 and Mob1 by cyclin B/CDKs together with Bub2/Bfa1 GAP activity provides a dual inhibition of the MEN [63] . Consistently , we found that combining the TEM1-Q79L allele with deletion of CLB2 , which encodes the main mitotic cyclin B , caused synthetic growth defects at 37°C ( Fig . 6C ) . However , no synthetic lethality or detrimental synthetic defects were induced by the additional deletion of KIN4 or BFA1 , suggesting that the components of the MEN downstream of Cdc15 are likely targets of negative regulators additional to Bub2/Bfa1 and Clb2-associated CDKs . Partial asymmetry of Bub2/Bfa1 at SPBs begins already in metaphase [39] . Since during metaphase MEN induces asymmetric localization of Kar9 at spindle poles , which is in turn required for correct spindle positioning [18] , we checked if symmetrical Bub2/Bfa1 and Tem1 might affect Kar9 localization and spindle positioning . To this end , we analysed the distribution of Kar9 tagged with eGFP on the metaphase spindles of wild type , SPC72-BFA1 bfa1Δ , SPC72-BUB2 bub2Δ , and TEM1-Q79L cells . Whereas 84 . 4% of wild type cells showed strongly asymmetric Kar9 , this value dropped to 43 . 5% , 50 . 5% and 47 . 6% in SPC72-BFA1 bfa1Δ , SPC72-BUB2 bub2Δ and TEM1-Q79L cells , respectively , while the remaining fraction of cells displayed partial or complete symmetry ( Fig . 7A ) . Since our results indicate that the GAP activity of Bub2 and Bfa1 influences the localization of the GAP complex in the cell , we characterized Kar9 distribution in the GAP-dead Bub2 variants , bub2-Q132L and bub2-R85A cells , in cells expressing the symmetric GAP-inactive BUB2-myc9 construct [25] . All strains showed more symmetric localization of Kar9 , with 33 . 9% of complete asymmetry for bub2-Q132L , 27 . 8% for bub2-R85A , 53 . 6% for BUB2-myc9 , as opposed to 80 . 3% for wild type cells ( Fig . 7B ) . Surprisingly , we found increased Kar9 symmetry also in bub2Δ and bfa1Δ mutant cells ( 45 . 7% and 41 . 9% of complete asymmetry , respectively ) , where Tem1 is present at SPBs at low levels [22] , but symmetrically ( Fig . 5C ) . Increased symmetry of Kar9 in the mutants ( with the exception of bub2Δ and bfa1Δ that were not analysed ) was accompanied by increased Bfa1 symmetry ( Fig . 7B ) . Therefore , establishment of Bub2/Bfa1 and Tem1 asymmetry impacts on Kar9 asymmetry . Nonetheless , Bub2/Bfa1 symmetric distribution at SPBs is not sufficient to drive Kar9 symmetry . Indeed , when spindles were misaligned in dyn1Δ mutant cells Bub2 became increasingly more symmetric from metaphase to anaphase , whereas Kar9 symmetry increased only slightly ( Fig . 8A-B ) , in agreement with recently published data [64 , 65] . Thus , once Kar9 asymmetry has been established , it cannot be reversed by spindle misalignment , in contrast to that of Bub2/Bfa1 . We then asked if Kar9 mislocalization in our mutants affects spindle positioning . SPC72-BFA1 and TEM1-Q79L cells expressing Spc42-mCherry were synchronized to collect cells in metaphase and imaged . Measurements of spindle distances from the bud neck ( Fig . 7C ) and spindle angles relative to the mother-bud polarity axis ( Fig . 7D ) indicated that both TEM1-Q79L and SPC72-BFA1 cells significantly affected the position and the orientation of metaphase spindles . Thus , hyperactive Tem1 impaired Kar9-dependent spindle positioning . In conclusion , although constitutive symmetric localization of Tem1 and its GAP Bub2/Bfa1 does not appear to affect mitotic exit , it does compromise asymmetry of Kar9 at spindle poles , thereby causing spindle positioning and orientation defects .
The mechanistic details of how GTPases switch between a GTP- and a GDP-bound state build on initial structural studies on Ras . In Ras a conserved glutamine in the switch II domain of the GTPase and a conserved arginine of the GAP both contribute to GTP hydrolysis [66 , 67] and , consistently , mutations of either residue abolish GTP hydrolysis . Crystal structure of some Rab GTPases in complex with their TBC ( Tre-2 , Bub2 and Cdc16 ) GAP revealed that the conserved switch II glutamine ( Q79 of Tem1 ) does not directly participate in GTP hydrolysis . Rather , catalysis is entirely brought about by a conserved arginine and a conserved glutamine of the TBC GAP through a mechanism referred to as “dual-finger” [48 , 68] . Hence , the switch II glutamine of the GTPase was proposed to stabilize its interaction with the GAP [48] . Recently , however , Rab GTPases have been shown to be more plastic than originally anticipated in their activation/hydrolysis mechanisms . In particular , the contribution of the switch II glutamine in GTP hydrolysis is variable and for some GTPases it contributes , together with a conserved lysine in the P-loop , to activation of the GTPase by stabilizing its GEF-bound nucleotide-free form [69] . Therefore , the outcome of mutations of conserved catalytic residues varies depending on the GTPase , GEF and GAP , and is altogether unpredictable . Here we have addressed the importance of Q79 in the switch II and the dual-finger mechanism in Tem1 GTP hydrolysis . First , we have established that the intrinsic rate of Tem1 GTP hydrolysis is negligible , and loss of GTP is mostly accounted for by nucleotide dissociation . In agreement with previous data [25 , 26 , 30] , Bfa1 prevents nucleotide dissociation and therefore acts as guanine-nucleotide dissociation inhibitor ( GDI ) . Second , we show for the first time that Tem1 Q79 is directly involved in the GAP-induced GTP hydrolysis without impairing its interaction with Bub2 and Bfa1 . Mutation of Q79 into leucine generates a hyperactive , dominant Tem1 that is refractory to its GAP , recruits more efficiently Cdc15 to SPBs and leads to unscheduled mitotic exit in the presence of spindle positioning defects . Finally , we show that the dual-finger mechanism applies also to GTP hydrolysis of the Tem1-Bub2-Bfa1 complex . Indeed , glutamine 132 of Bub2 is involved in GTP hydrolysis , in addition to arginine 85 that we previously showed [25] . Consistent with an important role of Q132 in Tem1 inhibition , bub2-Q132L mutant cells are SPOC-defective and undergo mitotic exit upon microtubule depolymerization . Thus , we have defined Q79 of Tem1 together with R85 and Q132 of Bub2 as a catalytic triad for GAP-induced GTP hydrolysis . The Bub2/Bfa1 complex is required for efficient Tem1 binding to SPBs throughout most of the cell cycle , except in late mitosis [22 , 40 , 61] . In contrast , SPB recruitment of Bub2/Bfa1 does not require Tem1 [30 , 40] . The amount of Tem1 at SPBs depends on the turnover of Bub2/Bfa1 at SPBs , which in turn is accelerated by spindle mispositioning through Kin4-dependent phosphorylation of Bfa1 [36 , 39] . Thus , as recently proposed [61] , the GAP Bub2/Bfa1 is a major Tem1 receptor at SPBs and its regulation is instrumental for establishing Tem1 asymmetry . Critical regulators of Bub2/Bfa1 asymmetry are the polo kinase Cdc5 and the phosphatase PP2ACdc55 . Cdc5 phosphorylates and inactivates the Bub2/Bfa1 complex leading to Tem1 activation [51 , 70] , whereas PP2ACdc55 dephosphorylates Bfa1 [71] . Phosphomimetic mutations in some Cdc5-dependent Bfa1 phosphorylation sites , as well as loss of PP2ACdc55 , are sufficient to induce premature asymmetry of Bub2/Bfa1 at SPBs , whereas Cdc5 inactivation or phospho-ablating mutations in Bfa1 lead to its persistent symmetry [62 , 71] . Similar mechanisms might be operational in fission yeast to establish SIN asymmetry . Indeed , polo kinase has been proposed to phosphorylate the Bfa1 homolog Byr4 and promote its dissociation from SPBs [72] , thereby influencing the distribution of GTP-bound Spg1 and its effector kinase Cdc7 . Furthermore , PP2A regulates Byr4 asymmetry through dephosphorylation of the SIN anchor at SPBs Cdc11 [73 , 74] . Indeed , Byr4 binds more efficiently to dephosphorylated Cdc11 [73] , whereas the Cdc15-like kinase Cdc7 binds preferentially phosphorylated Cdc11 [75] . Thus , S . cerevisiae and S . pombe might adopt common regulatory strategies to establish MEN and SIN asymmetry . We previously proposed that Tem1 GTP hydrolysis promotes asymmetry of both Tem1 and its GAP Bub2/Bfa1 at SPBs in anaphase [25] and results from other studies [22 , 62] support this conclusion . Consistent with this idea , we now show that mutating the second catalytic finger of Bub2 ( Q132 ) or mutating the catalytic Q79 of Tem1 leads in both cases to increased symmetry of Bub2/Bfa1 and Tem1 at SPBs . At a first glance these results appear at odds with the finding that in the complete absence of Tem1 Bub2/Bfa1 is not only recruited to SPBs with normal kinetics , but becomes asymmetric in anaphase exactly like in wild type cells [30] . However , we now show that when GTP hydrolysis is abolished Bub2 and Bfa1 bind more avidly to Tem1 . Thus , the increased symmetry of Bub2/Bfa1 at SPBs in these conditions likely reflects its stronger affinity for GTP-bound Tem1 . The different affinity of Bub2/Bfa1 for GTP- versus GDP-bound Tem1 has important implications for the SPOC , where active Tem1 needs to be quickly inactivated in the presence of a mispositioned spindle . Based on our previous results using a version of Bub2 tagged with nine myc epitopes at the C-terminus ( Bub2-myc9 ) , we proposed that removal of Bub2/Bfa1 from the mother SPB is important for timely mitotic exit [25] . Indeed , Bub2-myc9 is more symmetrically localized at SPBs than wild type Bub2 and is lethal for cdc5–2 and tem1–3 mutants because it prevents mitotic exit in these sensitized backgrounds . Now we further tested this idea by expressing chimeric proteins that constitutively recruit the GAP and Tem1 to both SPBs . Contrary to our predictions , these chimeric proteins partially rescued , instead of aggravating , the temperature-sensitive growth phenotype of several MEN mutants . In particular , Spc72-Bfa1 rescued the temperature-sensitivity of tem1–3 cells likely by suppressing the SPB-binding defects of the mutant Tem1–3 protein at high temperature [53] . Importantly , suppression of MEN mutants by our chimeric proteins is not accounted for by their possible impaired GAP activity , because it could not be recapitulated by BFA1 and/or BUB2 deletion . The chimeric proteins Spc72-Bfa1 and-Bub2 caused also unscheduled mitotic exit in the presence of mispositioned spindles , similar to Bub2- and Bfa1-Cnm67 fusion proteins previously characterized [36] . Therefore , despite different parts of Bub2 and Bfa1 are fused to the SPB anchor ( the N-terminus in our Spc72- fusions and the C-terminus in the-Cnm67 chimera ) , constitutive binding of the Bub2/Bfa1 complex to SPBs invariably leads to SPOC defects . Our data are totally consistent with the notion that Tem1 recruitment to SPBs is necessary for its MEN function [22] . In this scenario , SPB-locked Bub2/Bfa1 activates Tem1 by increasing its symmetry and residence time at SPBs , thereby causing unscheduled mitotic exit in the presence of mispositioned spindles . It is worth noting that symmetry of SIN components , such as the Cdc7 kinase , at SPBs also causes unscheduled cytokinesis and repeated rounds of septation [46 , 47] , suggesting that asymmetry is a common strategy in S . cerevisiae and S . pombe to restrain MEN/SIN signalling . The molecular basis for SPB-driven Tem1 activation is not known . It is possible that the GAP Bub2/Bfa1 at SPBs is constitutively kept inactive by Cdc5-mediated phosphorylation , thereby making Tem1 at SPBs refractory to GAP-mediated inhibition . Upon spindle misalignment , Kin4-mediated dislodgement of Bub2/Bfa1 from SPBs becomes essential for Tem1 inhibition in the cytoplasm and SPOC response [36 , 61] . An interesting non-mutually exclusive hypothesis is that a putative GEF for Tem1 localizes at SPBs [76] . However , as mentioned above the identity of Tem1 GEF ( s ) remains elusive . The finding that the chimeric proteins Spc72-Bfa1 and-Bub2 , similar to Bub2- and Bfa1-Cnm67 [36] and Cnm67-Tem1 [22] , support a mitotic arrest after microtubule depolymerization , but not after spindle mispositioning , was somehow puzzling . One major difference between the SAC-mediated metaphase arrest and the SPOC-mediated anaphase arrest is that in the latter , but not in the former , the PP2ACdc55 phosphatase is inhibited by the FEAR pathway [15 , 16] . We find that , indeed , CDC55 deletion or ESP1 overexpression in cells expressing Spc72-Bub2 or-Bfa1 drives unscheduled mitotic exit in the presence of nocodazole . In contrast , FEAR inhibition by deletion of SPO12 and BNS1 [15] prevents mitotic exit in cells expressing the same chimeric proteins and experiencing spindle position defects . Thus , whether the FEAR is activated or inhibited influences the impact of the chimeric proteins on mitotic exit . PP2ACdc55 has been recently shown to antagonize the Cdc5-dependent phosphorylation of Bfa1 [71] , thereby providing a mechanistic explanation to our data . The FEAR-mediated partial release of Cdc14 from the nucleolus might also contribute to MEN activation by counteracting the CDK-dependent inhibitory phosphorylation of MEN components [63] . In conclusion , persistent symmetric localization of the GAP Bub2/Bfa1 does not interfere with mitotic exit . Most likely , the Bub2-myc9 protein that we described previously prevents timely MEN activation by a different mechanism . Consistently , tagging of Spc72-Bub2 at the C-terminus with nine myc epitopes causes synthetic sickness in combination with the cdc5–2 mutant allele affecting the polo kinase ( S1 Table ) . We show that symmetric localization of the Bub2/Bfa1/Tem1 complex , independently of whether it is driven by chimeric proteins or loss of GTPase activity , interferes with Kar9 asymmetry at spindle poles in metaphase , as well as with spindle positioning and orientation relative to the cell division axis . Although Tem1 inactivation causes similar phenotypes for what concerns Kar9 localization and spindle orientation , it does not affect spindle positioning at the bud neck [18] , indicating that Tem1 hyperactivation and inactivation are not equivalent in this respect . The molecular bases of this difference remain to be established . Similarly , whether Tem1 hyperactivation primarily affects Kar9 localization and , as a consequence , spindle positioning or vice-versa remains to be investigated . Although the phenotypic analyses of our mutants did not reveal any apparent alterations of astral microtubules , at the moment we cannot exclude that subtle defects in microtubule dynamics could account for spindle mispositioning and , in turn , increased Kar9 symmetry . Previous [64] and our data indicate that not all conditions leading to symmetric distribution of the Bub2/Bfa1 complex and Tem1 cause symmetric localization of Kar9 at spindle poles . Indeed , whereas Bub2 , Bfa1 , Tem1 and Kar9 are all asymmetric , to different extents , on metaphase spindles , Bub2 , Bfa1 and Tem1 become increasingly more symmetric upon spindle misalignment , while Kar9 remains strongly asymmetric . These data suggest that establishment of Bub2/Bfa1/Tem1 symmetry on misaligned spindles is an active process and Kar9 asymmetry is so robust that once established it cannot be reversed by bringing the Bub2/Bfa1/Tem1 complex to both spindle poles . In contrast , in our mutants Bub2 , Bfa1 and Tem1 are more symmetric already in metaphase and might therefore interfere with the establishment of Kar9 asymmetry . Furthermore , we speculate that in these mutants the residence of Tem1 and the GAP Bub2/Bfa1 at SPBs is relatively stable , while these proteins turn over very fast at the SPBs of misaligned spindles [36] . An alternative explanation to our data is that Tem1 hyperactivation , rather than symmetry , is responsible for Kar9 mislocalization and spindle misalignment . Indeed , while in our mutants Tem1 is likely in the active GTP-bound form , it is inactivated and GDP-bound when the SPOC is turned on . Factors involved in cell polarity were implicated in the asymmetry of Bub2-Bfa1 at spindle poles [25 , 39] . Thus , it is tempting to speculate that in budding yeast the role of cell polarity in spindle positioning might be partly exerted through asymmetric localization of the Bub2-Bfa1-Tem1 trimeric complex at spindle poles , which in turn influences Kar9 asymmetry . Remarkably , other eukaryotic cells ( i . e . nematodes , flies and mammals ) employ heterotrimeric G proteins for spindle positioning during both symmetric and asymmetric cell division ( reviewed in [77] . A striking parallel can be drawn between the asymmetric enrichment of their GDIs GPR-1/2 , which is controlled by polarity factors and necessary for proper spindle alignment ( reviewed in [78] , and asymmetric localization of Bub2-Bfa1 . Future work will certainly shed new light onto possible additional similarities in the mechanisms adopted by different organisms to achieve correct spindle positioning .
All strains , except those used for Fig . 7B and 8 ( derivatives of S288c ) are derivatives of W303 ( ade2–1 , trp1–1 , leu 2–3 , 112 , his 3–11 , 15 , ura3 , ssd1 ) and listed in S2 Table . Cells were grown in YEP medium ( 1% yeast extract , 2% bactopeptone and 50mg/L adenine ) supplemented with 2% glucose ( YEPD ) or 2% galactose ( YEPG ) . Unless otherwise stated , α-factor , nocodazole and benomyl were used at 2 , 15 and 12 , 5 μg/ml respectively . Synchronization experiments with α-factor were performed at 25°C . Bacterial cells were grown in LD broth ( 1% bactotryptone , 0 , 5% yeast extract and 0 , 5% NaCl pH7 , 25 ) supplemented with 50 µg/ml ampicillin and 34 μg/ml chloramphenicol . The SPC72-BUB2 fusion was generated by triple ligation of a HindIII/XbaI PCR fragment containing the whole ORF of 340 bp of SPC72 promoter , a XbaI/EcoRI PCR fragment containing the ORF of BUB2 spanning codons 2–172 , and the LEU2-based Yiplac128 vector linearized with HindIII and EcoRI . The generated plasmid ( pSP275 ) was linearized with BamHI for integration at the BUB2 locus , thereby generating a gene fusion under the SPC72 promoter and containing the whole ORF of SPC72 fused in frame to the entire ORF of BUB2 , as well as a truncated BUB2 gene lacking the last 134 codons . Single integration of the construct at the BUB2 locus was checked by Southern blot . The SPC72-BFA1 fusion was generated by triple ligation of a HindIII/XbaI PCR fragment containing the whole ORF of 340 bp of SPC72 promoter , a XbaI/BglII PCR fragment containing the entire ORF of BFA1 starting from the 2nd codon and 330 bp of 3’ UTR , and the LEU2-based Yiplac128 vector linearized with HindIII and BamHI . The generated plasmid ( pSP371 ) was linearized with PstI for integration at the SPC72 locus and single integration of the construct at the SPC72 locus was checked by Southern blot . The ORF of TEM1 and about 1000 bp of promoter region was cloned in Yiplac128 ( pSP596 ) . A variant carrying the TEM1-Q79L mutation was generated by site-directed mutagenesis ( pSP597 ) . The TEM1-bearing plasmids have been integrated at the LEU2 locus by BstXI digestion and single integrations have been checked by Southern blot . Gene deletions were generated by one-step gene replacement [79] . One-step tagging techniques [80 , 81] were used to tag Tem1 , Tem1-Q79L , Bfa1 , Bub2 , Spc72-Bfa1 and Kar9 with multiple HA tags or eGFP . E . coli BL21 cells carrying pLysE plasmid ( Novagen ) and 6His-TEM1 , 6xHis-TEM1-Q79L , MBP-BFA1 , GST-BUB2 and GST-BUB2-Q132L expression plasmids were grown in LD broth containing ampicillin and chloramphenicol at 37°C for 3 h , transferred to 14°C for 1 h and induced with 0 , 1 mM isopropyl-1-thio-β-D-galactopyranoside for 15 h . Cells expressing MBP-Bfa1 , 6His-Tem1 and GST-Bub2 fusions were resuspended , respectively , in the following cold lysis buffers: 50 mM Tris-HCl pH7 . 5 , 200 mM NaCl and 2mM DTT supplemented with a cocktail of protease inhibitors ( Complete; Boehringer ) ; 50 mM Tris-HCl pH8 , 300 mM NaCl , 2mM MgCl2 and 10mM imidazole supplemented with a cocktail of protease inhibitors; 50 mM Tris-HCl pH7 . 5 , 200 mM NaCl supplemented with a cocktail of protease inhibitors . Cells were incubated with 1 mg/ml lysozyme in ice for 30 min , placed at 37°C for 5 min and sonicated at 4°C for 10 seconds . The extract was then clarified by centrifugation at 15 , 000 rpm for 30 min at 4°C . Tem1–6xHis and Tem1-Q79L-6xHis were purified by affinity chromatography with Ni-NTA columns ( QUIAGEN ) . The MBP-Bfa1 fusion protein was purified using Amylose resin ( New England Biolabs , Inc . ) , whereas GST-Bub2 and GST-Bub2-Q132L were purified with glutathione-Sepharose ( GE Heathcare ) . After elution , the fusion proteins were dialyzed against 50 mM Tris-HCl pH7 . 5 , 200 mM NaCl and stored at -80°C . For quantification , purified proteins were analyzed by Comassie staining and by Western blot with anti-GST polyclonal antibodies ( Santa Cruz Biotechnology , Inc . ) , anti-MBP mAb ( New England Biolabs , Inc . ) and 6xHis mAb ( CLONTECH Laboratories , Inc . ) . GTPase assay were performed according to [25] . In brief , 240 nM of Tem1–6×His was incubated in 25 μl of loading buffer ( 20 mM Tris-HCl , pH 7 . 5 , 25 mM NaCl , 5 mM MgCl2 , and 0 . 1 mM DTT ) containing 0 . 1 MBq of γ[32P]GTP or 0 . 03 MBq of γ[35S]GTP in the absence or presence of 150 nM of MBP-Bfa1 for 10 min at 30°C . The reaction was then put on ice , and 10 μl of reaction were added to 50 μl of reaction buffer ( 20 mM Tris-HCl , pH 7 . 5 , 2 mM GTP , and 0 . 6 μg/μl BSA ) containing 15 μM of GST-Bub2 . The mixture was incubated at 30°C , and for each time point 10 μl of the reaction was diluted in 990 μl of cold washing buffer ( 20 mM Tris-HCl , pH 7 . 5 , 50 mM NaCl , and 5 mM MgCl2 ) . The samples were filtered through nitrocellulose filters , washed with 12 ml of cold washing buffer , and air dried , and the filter-bound radioactivity nucleotide was determined by scintillation counting . Immunoprecipitations were performed as described in [82] . Bub2-HA3 , Bub2-Q132L-HA3 , Tem1-HA3 and Tem1-Q79L-HA3 were immunoprecipitated from 1 mg of total extract by a HA-affinity resin ( Roche ) . For Western blot analysis , protein extracts were prepared according to [83] . Proteins transferred to Protran membranes ( Schleicher & Schuell ) were probed with anti-PK mouse monoclonal antibodies for PK-tagged Bub2 , with anti-GFP rat monoclonal antibodies for GFP-tagged Tem1 , Bub2 and Bfa1 ( Chromotek ) and with an anti-HA monoclonal antibody ( 12CA5 ) . Secondary antibodies were purchased from GE Healthcare , and proteins were detected by an enhanced chemiluminescence system according to the manufacturer . In situ immunofluorescence was performed according to [84] . Immunostaining of α-tubulin was performed with the YOL34 monoclonal antibody ( Serotec ) followed by indirect immunofluorescence using rhodamine-conjugated anti–rat antibody ( 1:100; Pierce Chemical Co . ) . Cells expressing GFP and mCherry-tagged proteins were grown in minimum complete medium . Digital images of live cells , cells fixed with 3 . 7% formaldehyde or cold EtOH were taken with an oil 63X 1 , 4–1 , 6 HCX Plan-Apochromat objective ( Zeiss ) with a Coolsnap HQ2–1 charge device camera ( Photometrics ) mounted on a ZeissAxioimager Z1/Apotome fluorescence microscope controlled by the MetaMorph imaging system software . Z-stacks of 12 planes at 0 . 3 μm step size were acquired . For analysis of actomyosin ring contraction ( Fig . 4C ) cells were mounted in SD medium on Fluorodishes and filmed at room temperature ( ~21°C ) with a DeltaVision OMX microscope using a 63X 1 . 4 NA oil immersion objective and the softWoRx software ( Applied Precision ) . Z stacks containing 31 planes were acquired every 1’ with a step size of 0 . 2 μm and a binning of 1 . Z-stacks were deconvolved with Huygens ( Scientific Volume Imaging ) and max-projected . For the analyses in Fig . 8 , 120s time-lapse microscopy was performed using an Olympus BX51 microscope controlled by the TILLVision software ( TILLPhotonics ) . For the localization of Kar9-YFP and Bub2-GFP , Z-stacks of four layers ( step size 0 . 35 μm ) and maximum intensity projections were used . Fluorescence microscopy was performed with a monochromator PolychromIV as light source and a CCD camera ( Imago , TillPhotonics ) . Fluorescence intensity measurements of max intensity-projected images were performed using the ImageJ software . The index of Bfa1 symmetric distribution ( σ , Fig . 5G ) was measured using the following equation: σ ( 0<σ<1 ) = I1/I2 , where I1 is the fluorescence intensity of the brightest of the two SPBs , and I2 is the fluorescence intensity of the dimmest . The position and the orientation of the spindle were measured on max intensity-projected images using ImageJ . The position was determined measuring the distance between the bud neck and the nearest SPB . The orientation was determined measuring the smaller of the two angles that the spindle forms intersecting the polarity axis of the cell . Adobe Photoshop and ImageJ were used to mount the images and to produce merged color images . No manipulations other than contrast and brightness adjustments were used . Student’s t-test or chi-square test was used to evaluate statistical significance of differences , depending on whether one ( t-test ) or more parameters ( chi-square test ) were compared for each experimental condition . Nuclear division was scored with a fluorescence microscope on cells stained with propidium iodide ( Sigma Aldrich ) . Flow cytometric DNA quantification was determined according to [84] on a Becton-Dickinson FACScalibur . | In asymmetrically dividing cells , proper positioning of the mitotic spindle relative to polarity determinants is crucial to ensure the unequal fate of daughter cells . In stem cells , derangement of the mechanisms controlling asymmetric cell division , including spindle positioning , affects the developmental fate of daughter cells and can promote tumourigenesis . The budding yeast Saccharomyces cerevisiae is an outstanding model system to study spindle positioning and its links with cell cycle progression . Indeed , budding yeast has redundant mechanisms driving spindle positioning and a “spindle position checkpoint” ( SPOC ) that delays cell division whenever the spindle is not properly aligned . The target of the SPOC is the small GTPase Tem1 that controls both spindle positioning and mitotic exit and whose activity can be inhibited by the GTPase-activating protein Bub2/Bfa1 . Tem1 , Bub2 and Bfa1 form a complex at spindle poles that becomes asymmetric and accumulates on one spindle pole when the spindle is properly aligned , while it remains symmetric in case of spindle mispositioning . Through expression of several mutant or chimeric proteins leading to symmetric distribution of the Bub2/Bfa1/Tem1 complex , we establish that asymmetry of these proteins does not drive mitotic exit but rather it contributes to spindle alignment . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| []
| 2015 | Asymmetry of the Budding Yeast Tem1 GTPase at Spindle Poles Is Required for Spindle Positioning But Not for Mitotic Exit |
The study of the chronological life span of Saccharomyces cerevisiae , which measures the survival of populations of non-dividing yeast , has resulted in the identification of homologous genes and pathways that promote aging in organisms ranging from yeast to mammals . Using a competitive genome-wide approach , we performed a screen of a complete set of approximately 4 , 800 viable deletion mutants to identify genes that either increase or decrease chronological life span . Half of the putative short-/long-lived mutants retested from the primary screen were confirmed , demonstrating the utility of our approach . Deletion of genes involved in vacuolar protein sorting , autophagy , and mitochondrial function shortened life span , confirming that respiration and degradation processes are essential for long-term survival . Among the genes whose deletion significantly extended life span are ACB1 , CKA2 , and TRM9 , implicated in fatty acid transport and biosynthesis , cell signaling , and tRNA methylation , respectively . Deletion of these genes conferred heat-shock resistance , supporting the link between life span extension and cellular protection observed in several model organisms . The high degree of conservation of these novel yeast longevity determinants in other species raises the possibility that their role in senescence might be conserved .
Yeast , worms , and flies have been studied extensively to identify the genetic determinants of aging . Studies conducted in these model organisms have demonstrated a partially conserved life span regulatory role for the nutrient-sensing/insulin/IGF-I-like pathways , which are found in species ranging from yeast to mice [1] , [2] . Two different paradigms have been established to study the life span of yeast . Chronological life span ( CLS ) measures the mean and maximum survival time of populations of non-dividing yeast [3] , while replicative life span ( RLS ) refers to the number of daughter cells generated by an individual mother cell before it ceases to divide [4] , [5] . Several genes similarly affect both CLS and RLS , while others have opposite effects on the two aging paradigms , suggesting that the mechanisms underlying the CLS and RLS are only partially overlapping [6] , [7] . By screening transposon-mutagenized yeast populations ( previously selected for their ability to withstand either oxidative or heat stress ) for mutants with an extended CLS , the serine-threonine protein kinase Sch9 and adenylate cyclase ( Cyr1 ) were identified as negative regulators of longevity [8] . The effect of the Ras/Cyr1/PKA pathway on aging had been previously described based on its role in glucose signaling [9] , [10] . Reducing the activity of Sch9 or Cyr1 and consequently that of the nutrient-sensing pathways they participate in ( TOR/Sch9 and Ras/Cyr1/PKA ) , CLS is extended by up to 3-fold , with a concomitant increase in resistance to cellular stress [8] . Consistent with this observation , inactivation of the G-protein Ras2 , which promotes Cyr1 function , also extends CLS [11] . The two closest metazoan homologues of Sch9 , Akt and S6K , have been implicated in the insulin/IGF-I-like signaling and life span regulation in all the major model organisms [1] , [12] , [13] , [14] . Conversely , the role of the Ras/Cyr1/PKA signaling in aging of higher eukaryotes has been more elusive [15] . However , recently , mice lacking adenylate cyclase 5 ( AC5 ) have been reported to be long-lived and fibroblasts derived from these mice have been shown to be resistant to oxidative stress , consistently with previous observations in yeast cyr1 mutants [16] . Moreover , the disruption of RIIβ , which codes one of the mammalian PKA regulatory subunits , has been shown to promote median and maximum life span extension in male mice [17] . In the last few years several laboratories have turned to the yeast CLS to elucidate how post-mitotic and reversibly arrested cells age in higher eukaryotes . However , some concern over the extensibility of this model has been raised in light of recent observations that acetic acid , which accumulates extracellularly in the culture medium , is a key cause of chronological aging in yeast [18] . The question is if acetic acid-dependent cell death is relevant to aging in metazoans . Previously , we found that ethanol accumulates during chronological aging and promotes death , and that its removal extends CLS [7] . We also found that glycerol replaces ethanol in cultures of long-lived yeast and its synthesis is crucial for longevity extension [19] . Burtner et al . have proposed that ethanol is metabolized to produce acetic acid , to which long-lived mutants are more resistant than wild type yeast [18] . Others have suggested that ethanol removal via the activation of gluconeogenesis mediates longevity extension [20] . Although ethanol and acetic acid at high concentrations may in fact be directly toxic to the cell , for S . cerevisiae they are commonly encountered carbon sources and thus , their removal may extend life span in part by promoting calorie restriction , a non-genetic intervention known to extend the life span of a broad range of species [21] . Further studies are needed to clarify the range of metabolic changes that occur during chronological aging to understand how acetic acid or other acids , ethanol , or glycerol might be relevant to aging of multicellular eukaryotes . While it is plausible that , by analogy with yeast , the composition of the extracellular milieu of multicellular organisms contributes to aging [22] , different metabolites might be implicated in aging of multicellular species . Notably , mutations in the Sch9 and Ras/Cyr1/PKA pathways in yeast extend CLS even after removal of extracellular carbon sources indicating that the release of ethanol and acetic acid into the medium is not a requirement for these genes to exert their effect on longevity [23] . Previously , Powers and coworkers used the yeast diploid homozygous deletion collection , which covers 96% of the yeast genome [24] , [25] , to develop an assay to monitor the CLS of all individual deletion mutants . The principal finding of their screen was the identification of the TOR pathway as a pro-chronological aging pathway . In fact , deletion of either TOR1 or of several other genes controlled by the TOR cascade , e . g . GLN3 ( encoding a transcription factor induced by the amino acid starvation response ) , prolongs CLS [26] . A pro-chronological aging role for the serine/threonine kinase Tor1 has recently been confirmed by others [27] and the down-regulation of the TOR signaling cascade has also been implicated in the CLS extension induced by calorie restriction [23] . In yeast , Sch9 is a direct target of the Tor-containing complex 1 ( TORC1 ) and its inactivation mediates the CLS extension observed in a tor1Δ context [23] , [28] , [29] . A role for TOR in longevity regulation has been confirmed in worms and flies [12] , [30] , [31] and recently , by analogy with yeast [26] , mice and flies treated with rapamycin , an inhibitor of TORC1 , have been reported to live longer than untreated controls [32] , [33] . The conservation of the TOR kinases and of their role in aging across species suggests that rapamycin may represent the first drug that functions to prolong life span of multiple species including mammals . High rates of false positives and negatives are common in genomic screens [34] , accordingly , we decided to use a different methodological approach to screen for gene deletions that affect CLS . We relied on competitive screening of pools of the ∼4800 non- essential deletion mutants in the haploid wild type BY4741 genetic background [35] . Notably , the deletion strategy designed to construct the yeast knock-out collection generates two unique 20bp DNA tags on each deletion mutant ( uptag and downtag ) . These tags allow the monitoring of the changes in representation of each deletion mutant in a pool using a barcode microarray that carries the complement of the tag sequences . Thus , our method differs from that of Powers et al . in that: 1 ) it measures the CLS of pooled , competitive cultures of standard size ( 50 mL ) instead of that of individual micro-cultures ( 0 . 2 mL ) of each deletion mutant , and 2 ) it employs a DNA microarray-based technique to quantify the age-dependent individual strain abundance rather than absorbance measurement of individual cultures .
In order to identify novel genes implicated in life span regulation we measured the CLS of two independent yeast populations obtained by diluting two identical pools of 4×106 frozen cells into 50 mL of synthetic complete medium containing 2% glucose ( SDC ) . After 3 days , the two yeast cultures reached a densitiy of 1 . 5×108/mL . Because in a standard CLS experiment , no further increase of cell density is usually observed after 3 days , the number of colony forming units ( CFUs ) measured at day 3 was defined as 100% survival [3] . The survival curves for each pooled culture are shown in Figure 1A , the actual CFUs data are reported on Table S1 . Interestingly , both mean and maximum survival times were significantly shorter as compared to those of the wild type BY4741 ( Figure 2A ) [7] . This may be due to the fact that numerous deletions reduce survival [36] and/or the survival defects of the corresponding mutants are exacerbated when they grow in the presence of 4800 other deletion strains . Consistent with this hypothesis , we observed a high number of budded cells in pooled cultures ( data not shown ) suggesting that several deletions may cause an increase of the non-quiescent fraction of cells [37] . Notably , post-diauxic and stationary phase cultures of yeast aging chronologically are composed of both quiescent and non-quiescent cells , although cell division within the population grown in SDC medium appears to be minimal and to not affect the measurement of CLS [38] , [39] . Non-quiescent cells differ from quiescent cells in that they do not arrest in G0 properly , are more susceptible to reactive oxygen species and apoptosis , and lose viability more rapidly than G0-arrested quiescent cells [38] , [40] . The survival curves of both pooled cultures showed an increase of CFUs at day 12 and 15 ( Figure 1A ) . This may be caused by specific mutants that can utilize the low nutrient medium for growth ( see next section , paragraph on adaptive regrowth ) [41] . To measure the viable cells corresponding to each individual mutant , aliquots containing 6 . 25×105 cells of each culture were diluted in fresh medium and grown until they reached a cell density of 107/mL . Samples corresponding to approximately 2×107 cells were frozen at day 3 , 9 , 11 , 15 , and 20 . Genomic DNA was extracted from cell pellets as described by Pierce et al . [35] . Aging cultures were not used directly for DNA extraction to avoid any noise that might be contributed by unlysed dead cells . Both uptags and downtags were PCR-amplified and hybridized to Affymetrix TAG4 arrays , which were processed as previously described [35] . For each time point , the log2 intensity ratio was calculated with respect to day 3 ( 100% survival ) and the aging profiles for each individual mutant were extracted ( Table S2 ) . The root squared mean errors ( RSME ) between the two replicates were calculated and mutants with high RSME ( 90th percentile ) were excluded from the analysis ( Table S2 ) . The microarray results were used to approximate a survival curve for each individual deletion strain by multiplying the fold ratio change in the microarray results by the CFUs relative to the pools ( Table S3 ) . K-means clustering analysis ( K = 10 ) was performed on the averaged log2 ratios between the two pools and five clusters corresponding to mutants whose life span trajectories differed from that of the mean of the pool were identified by manual inspection ( Figure 1B ) . While mutants belonging to four clusters , 1-3-6-7 , were classified as short-lived , cluster 2 included long-lived mutants ( Figure 1B , Tables S5 , S6 ) . In parallel , we also used a significance analysis of time course microarray experiments developed for identifying differentially expressed genes in a time course to test for consistency between our replicates ( EDGE analysis , see Materials and Methods , Tables S4 , S5 , S6 ) [42] , [43] . K-means clustering indicated that 594 genes are required for normal life span ( Table S5 ) . Among these , we observed an enrichment of genes belonging to the “mitochondrion” gene ontology group ( GO: 0005739 , 24 . 3% vs 15 . 4% , relative vs background frequency ) , with 6 . 1% and 3 . 4% being part of the “mitochondrial inner membrane” ( GO:0005743 , background frequency 2 . 4% ) and “mitochondrion degradation” ( GO: 0000422 , background frequency 0 . 5% ) GO categories , respectively ( Table S7 ) . Many mitochondrial genes among those whose deletion shortens the CLS were expected because functional mitochondria are important for survival after the diauxic shift when glucose is depleted and yeast switch from fermentative to respiratory metabolism [44] , [45] . The list of genes whose deletion is associated with reduced life span is also enriched in members of the “autophagy” , “macroautophagy” , and “microautophagy” GO biological process categories ( GO: 0006914 , GO:0016236 , GO: 0016237 , respectively ) ( Table S7 ) . This suggests that protein and organelle turnover by vacuolar digestion is required for normal survival and may contribute to prolong yeast life span , consistently with proposals for C . elegans and Drosophila [46] , [47] , [48] . Among the shortest-lived mutants , we identified several mutants carrying deletions of genes implicated in protein targeting to the vacuole ( VPS genes ) . To validate our screening results we measured the life span of mutants lacking individual Vps proteins , namely Vps25 , Vps27 , Vps21 , Vps36 , and Vps8 ( q<0 . 1 , EDGE analysis , Table S5 ) . Four of the five mutants were short-lived ( Figure 2B , see below ) . All the experiments described hereafter in the BY4741 background were performed by switching the cells to water at day 3 after the yeast populations had reached saturation rather than by leaving them in medium . Incubation in water represents a form of starvation/extreme calorie restriction ( CR ) , which , similarly to the reduction of glucose content in the growth medium , promotes life span extension [7] , [39] , [44] , [49] . Previously , we have shown that similar pathways are implicated in both starvation ( water ) - and CR ( 0 . 5% glucose ) -dependent CLS extension [23] . We have also shown that virtually all the mutants that show longevity extension in SDC are long-lived also when different media are used for the survival studies ( e . g . synthetic complete + 0 . 5% glucose , water , or SDC without tryptophan on plates ) ( [7] , [19] , [23] and M . Wei , unpublished results ) . The monitoring of longevity in water is also a useful means to rule out any occurrence of adaptive regrowth , which can confound the interpretation of our survival data . Adaptive regrowth occurs when aging cells acquire mutations that allow them to reenter the cell cycle in conditions than normally do not promote growth [41] . It is usually observed in wild type yeast after a large fraction of the yeast population is inviable , because it depends upon the nutrients released by the dead cells to occur and can be prevented by switching the cells to water and washing them periodically [41] , [44] . The frequency of adaptive regrowth is increased in mutants that are more sensitive to DNA damage , e . g . sod1Δ or sgs1Δ [39] , [41] . Since BY4741 shows a modest response to starvation/extreme CR in comparison with other genetic backgrounds ( Figure 2A ) ( [7] , [39] and P . Fabrizio unpublished results ) , we hypothesized that this may depend in part on a tendency of BY4741 , in contrast with other strains , to resume cell division when a large fraction of cells is still alive . Thus , to obtain more conclusive data relative to the nature of our putative BY4741 short- and long-lived mutants , we performed our survival assays in water . These experiments test the role of the putative life span regulatory genes in starvation/extreme CR-dependent life span extension and do not represent a direct validation of our screen , which did not assay for survival in water . Nevertheless , the individual strain survival assays in water allow us to identify mutations that diminish or prolong life span in the BY4741 background and to avoid mistaking deletions that promote adaptive regrowth for those that extend life span . Deletion of VPS25 and VPS27 causes a dramatic reduction of life span ( average of three independent experiments ) to a level below that of wild type cells in SDC ( Figure 2A and 2B ) ( p<0 . 001 ) . Lack of Vps21 and Vps8 reduced life span under starvation conditions to a level similar to that of wild type cells incubated in SDC ( p<0 . 01 and 0 . 05 , respectively ) ( Figure 2A and 2B ) . In contrast , the vps36 deletion mutant lived as long as the wild type BY4741 . Thus , Vps36 is not required for the starvation/extreme CR-dependent life span extension ( Figure 2B ) . Since the Vps proteins are important for protein degradation , they may contribute to the removal of oxidized/damaged proteins known to accumulate during aging [50] , [51] . Consequently , in their absence yeast might be more sensitive to oxidants . To test this hypothesis , we monitored the resistance to hydrogen peroxide ( 100–200 mM for 30 minutes ) of different vps mutants during chronological aging at day 1 and 3 and found an association between life span and resistance to oxidative stress , with vps25Δ and vps27Δ being the shortest-lived and also the most stress sensitive and vps36Δ having a normal life span and also unaltered stress resistance ( Figure 2C ) . The vps25Δ and vps27Δ mutants were also tested for their resistance to acetic acid by exposing day 3 cultures to 300 mM acetic acid for 3 hours . Both mutants showed an increased sensitivity to acetic acid in comparison with the wild type ( Figure 2D ) . Combined with the increased sensitivity to hydrogen peroxide , this appears to reflect a general susceptibility of these mutants to stress and not the mechanism leading to early cell death , since 300 mM acetic acid is much higher than the level normally encountered/generated by cells ( Figure 2D ) [18] . Together , these results indicate that functional Vps-dependent protein degradation systems are essential for starvation-dependent life span extension . While mutations that shorten life span may not be directly associated with aging but rather may simply cause reduced cellular fitness , mutations that extend life span are , in most cases , indicative of an involvement of the corresponding genes in the aging process . K-means clustering analysis allowed us to identify 42 putative long-lived mutants ( Table S6 ) . To select the strains to be retested for longevity under starvation/extreme CR , after excluding the mutants carrying deletions of dubious ORFs not overlapping any ORF/gene on the complementary strand ( YOR012W , YDR102C ) and the ydr442wΔ and sfl1Δ mutants , which showed a marked flocculation phenotype in synthetic medium , we randomly chose 14/42 mutants ( Table 1 ) . Five of them , acb1Δ , cka2Δ , trm9Δ , ydr417cΔ , and aro7Δ were confirmed as long-lived in the BY4741 genetic background ( Figure 3A–3E , p<0 . 01–0 . 05 ) . The life span of mutants lacking either Cup9 , Apd1 , Zta1 , or Ssn2 , a transcriptional repressor , a protein required for normal localization of actin patches , a quinone reductase , and a subunit of the RNA polymerase II mediator complex , respectively , was not significantly different from that of the wild type ( Figure 3F , Figure S1 , and data not shown ) . While the mutants living significantly longer than the wild type ( acb1Δ , cka2Δ , trm9Δ , ydr417cΔ , and aro7Δ ) were heat resistant ( Figure 4A and 4B ) ( see below ) no major changes in heat-shock resistance were observed in the mutants ( cup9Δ , apd1Δ , zta1Δ , ssn2Δ ) whose life span extension was not significant ( Figure 4A and data not shown ) . The deletion of ACB1 , which encodes a highly conserved acyl-CoA-binding protein implicated in acyl-CoA-ester transport , sphingolipid synthesis , and fatty acid chain elongation [52] , caused a 2 . 2-fold mean life span extension in the genetic background BY4741 ( Figure 3A ) . Lack of Acb1 also increased heat-shock resistance in chronologically aging cells ( Figure 4A ) , a phenotype observed in the great majority of long-lived mutants so far identified [53] . Similarly , resistance to a very high concentration of acetic acid was enhanced by the deletion of ACB1 ( Figure 4D ) . However , in contrast with other long-lived yeast , the acb1Δ mutants did not exhibit any resistance to oxidative stress measured as the ability to maintain viability after 30 minute-treatment with 200–300 mM H2O2 ( data not shown ) . To test the role of Acb1 in life span regulation in non-CR conditions ( incubation in SDC medium ) and in different genetic backgrounds , we deleted ACB1 in W303-1A and DBY746 , which usually undergo adaptive regrowth only in the late phases of chronological survival ( [23] , [44] and P . Fabrizio unpublished results ) . In these backgrounds the acb1Δ mutants showed severe growth defects , were slightly short-lived and heat-shock sensitive ( data not shown ) . Since a yet uncharacterized adaptation that leads to faster growth has been reported to occur at high frequency in acb1Δ cultures [54] , we verified the linkage between ACB1 and our phenotypes of interest in the BY4741 acb1Δ mutant from the deletion collection , which displays only a modest growth defect and might carry suppressor mutations . To do this , the mutant was transformed with a centromeric plasmid containing the ACB1 gene under its own promoter and both heat-shock resistance and CLS were monitored . ACB1 expression complemented both heat-shock resistance and life span extension of the acb1Δ mutant ( Figure 4F and Figure S2 ) indicating that both phenotypes are caused by the deletion of ACB1 . The deletion of CKA2 , which encodes one of the two catalytic subunits of the serine-threonine kinase CK2 , approximately doubled the mean life span of BY4741 under starvation/extreme CR ( Figure 3B ) . CK2 is a tetramer comprised of two catalytic and two regulatory subunits , which regulates cell growth/division ( among other functions ) in all eukaryotes so far investigated [55] , [56] . Analogous to the acb1Δ mutant , yeast lacking Cka2 are heat-shock and acetic acid resistant but not resistant to H2O2 ( Figure 4A and 4D , and data not shown ) . The deletion mutants corresponding to either one of the two regulatory subunits ( Ckb1 and Ckb2 ) were also resistant to heat . Conversely , yeast lacking the catalytic subunit Cka1 were approximately as resistant as wild type cells ( Figure 4C ) . These results suggest that the activity of the holoenzyme and not only of the free catalytic subunits , which are known to have functions independent of the regulatory subunits [57] , are responsible for the phenotypes observed . Furthermore , the involvement of both Ckb1 and Ckb2 in the regulation of stress resistance is in agreement with the requirement of both regulatory subunits for the full CK2 activation [56] . The role of CK2 in life span regulation and heat-shock resistance was confirmed in SDC medium in the W303-1A and DBY746 genetic backgrounds ( Figure 5A and 5B , Figure 6A ) . To support the hypothesis that the holoenzyme activity promotes aging , we deleted CKB2 in DBY746 and monitored life span and stress resistance of the corresponding mutant . Lack of Ckb2 promoted a modest but significant ( p<0 . 05 ) longevity extension and a marked increase of heat resistance in comparison with the wild type ( Figure 5B , Figure 6B ) . Two highly specific CK2 inhibitors , 4 , 5 , 6 , 7-tetrabromo-benzotriazole ( TBBt ) and 4 , 5 , 6 , 7-tetrabromo-benzimidazole ( TBBz ) , have been identified and shown to inhibit the activity of the holoenzyme [57] . More specifically , in yeast TBBz inhibits the CK2 complex selectively and not the free Cka2 catalytic subunits [57] . We tested both inhibitors in our system and found that TBBz ( 10–200 µM ) but not TBBt ( 5–15 µM ) increased substantially the heat resistance of day 3 DBY746 cultures ( Figure 6C , Figure S3A ) . Furthermore , TBBz but not TBBt improved survival at day 5 ( Figure 5C , Figure S3B ) . Together , these results confirm that the activity of the holoenzyme is responsible for the pro-aging effect of Cka2 . TRM9 codes a tRNA methyltransferase that methylates uridine residues at the wobble position in tRNA ( Glu ) and tRNA ( Arg3 ) [58] . Its deletion in BY4741 almost tripled yeast mean CLS under starvation/extreme CR ( Figure 3C ) , increased heat resistance ( Figure 4B ) , but reduced resistance to acetic acid ( Figure 4E ) . Similar results were obtained by testing a trm9Δ mutant generated in the DBY746 background in SDC medium ( Figure 7A and 7D ) . In this background , lack of Trm9 exacerbated the mild growth defect observed in BY4741 as estimated by colony size ( Figure 4B , Figure 7D ) . The deletion of YDR417C also promoted longevity extension and heat-shock resistance but reduced acetic acid resistance ( Figure 3D , Figure 4A and 4D ) . This dubious ORF overlaps widely with the gene coding the ribosomal protein Rpl12b . The life span and resistance to heat of yeast lacking Rpl12b were similar to that of the ydr417cΔ mutant ( Figure 7C and 7E , Figure 4A , Figure 3D ) . Notably , no protein expression corresponding to YDR417C was detected by analyzing strains carrying either the GFP- or TAP-tagged version of this ORF . By contrast , Rpl12b was detected using both tagging systems [59] , [60] . In DBY746 the deletion of YDR417C caused a marked reduction of colony size , almost doubled the mean life span in SDC medium ( p<0 . 005 ) , and increased heat-shock resistance of chronologically aging yeast ( Figure 7B and 7D ) . ARO7 encodes for chorismate mutase , which is required for the biosynthesis of the aromatic amino acids tyrosine and phenylalanine from chorismate . Its deletion lowered fermentative growth rates ( data not shown ) and caused a ∼60% reduction of the total number of cell counted at day 3 . Approximately 5×107cells/mL were alive at day 3 and ∼70% survived in water until day 37 ( Figure 3E ) . Chronologically aging aro7Δ mutants were more resistant to heat-shock but more sensitive to acetic acid than wild type yeast ( Figure 4B and 4E ) . In the W303-1A background , the deletion of ARO7 caused an even more severe growth defect and the mutants were short-lived ( data not shown ) . This may depend on a different response to lack of Aro7 in different genetic backgrounds . Notably , extreme growth defects might reflect the inability of old G0-arrested cells to reenter the cell cycle to form colonies , simulating a short-lived phenotype . Of the remaining putative long-lived mutants whose longevity was tested under starvation/extreme CR , far3Δ , far11Δ , ppg1Δ , and bul1Δ lived shorter than wild type ( the reduction of life span was significant for all the mutants except bul1Δ ) while pan2Δ lived approximately as the wild type ( Figure S4A , S4B ) . Far3 and Far11 are part of a complex that plays a role in promoting G1-arrest in response to pheromone signaling [61] . Notably , Far7 , Far8 , and Far10 are found in the same protein complex and the corresponding deletion mutants were all identified as putative long-lived strains ( Table S6 ) . Furthermore , two of the dubious ORFs whose deletion is associated with longevity extension , YDR199W and YMR052C-A , overlap with FAR9 , coding an additional component of the Far complex , and FAR3 , respectively ( Table S6 ) . It is plausible that lack of these proteins may inhibit the G1-arrest triggered by further stimuli , i . e . nutrient shortage , and cause adaptive regrowth . Mutants displaying the adaptive regrowth phenotype may therefore be mistaken for long-lived due to an enrichment of their representation in a pool caused by cell division . To further characterize the long-lived mutants identified in this study , we measured the budding index of each of them in exponential phase and during chronological aging ( Figure 8 ) . Notably , a more complete G1/G0-arrest , measured as a decrease of budding index , has been observed in chronologically aging long-lived mutants and is believed to contribute to longevity extension via the reduction of replicative stress [40] . Our analysis revealed a lower ratio between budded and unbudded cells in all mutants in comparison with the wild type during the exponential phase . The reduction of budding index was statistically significant for all the mutants except cka2Δ in agreement with the mild growth defects observed in the mutants ( data not shown ) . On day 1 the budding index of both acb1Δ and trm9Δ was significantly higher than that of wild type cells ( p<0 . 01 ) and in the acb1Δ mutant it remained higher until day 7 ( p<0 . 01 ) ( Figure 8 ) . By contrast , the budding index of the aro7Δ mutant was significantly lower than that of the wild type on day 3 and 7 ( p<0 . 01 ) ( Figure 8 ) .
The use of the yeast deletion collection combined with a tag microarray detection method has found a wide range of applications , many of which involve drug screening to define their mechanisms of action [62] , [63] . Here we adapted this methodology to investigate how different genes affect the chronological life span . By performing a competitive survival assay on a pool of approximately 4800 haploid deletion strains , we identified several novel life span determinants . Analogously to Powers et al . , we obtained data supporting the importance of functional mitochondria in long-term survival and identified several autophagy-related genes that are required for normal life span [26] ( Figure 2 and Table S7 ) . The autophagic process is down-regulated by the principal pro-aging pathways and work done in yeast , worms , and flies suggests that it is required for longevity extension [46] , [48] , [64] . Interestingly , we identified a significant number of genes whose deletion is associated with short life span , which are included in the “mitophagy” GO group ( GO:0000422 ) ( Table S7 ) . Since in non-dividing cells autophagic breakdown is the only mechanism to remove damaged organelles , we speculate that this is a key element in long-term survival and longevity extension . Furthermore , autophagy plays an important role in the removal of damaged proteins , which are known to accumulate during aging [65] . Because our studies suggest that adaptive regrowth is common in the BY4741 background and also to test the mechanisms of starvation-dependent CLS extension , the longevity tests performed on the individual BY4741 mutants were performed under starvation conditions , whereas the original screen was carried out on cells incubated in medium throughout the experiment . Notably , the great majority of mutations that cause life span extension in medium does so in water [7] , [44] . Yeast cultures were transferred to water at day 3 , a condition that leads to the activation of an anti-aging response analogous to that promoted by reducing the glucose content of the growth medium and controlled by the same mediators [7] , [23] , [44] . Thus , our tests on the individual BY4741 mutants studied the effect of individual genes on the starvation/extreme CR-induced longevity extension . In this context , the results shown in Figure 2B indicate that the protein transport to the vacuole is required for the extended life span associated with starvation/extreme CR . However , the dramatic shortening of longevity observed in the vps27Δ and vps25Δ mutants and their sensitivity to oxidative stress ( Figure 2B and 2C ) strongly suggest that in chronologically aging yeast protein turnover by autophagy is a crucial function for survival in both regular medium and under starvation/extreme CR . Interestingly , Vps27 and Vps25 are components of the Vps27-Hse1 and ESCRTII complexes , respectively . Both complexes are part of ESCRT system and are involved in the degradation of ubiquitylated transmembrane proteins via the formation of multivesicular bodies ( MVBs ) [66] . Their key role in survival underlines the importance of plasma membrane and Golgi protein breakdown in preserving cellular function over time . Of the 14 putative long-lived BY4741 mutants retested , 9 lived longer in water , with 5 of them reaching a significantly extended life span ( Figure 3 ) . Three of the latter ( cka2Δ , trm9Δ , and ydr417cΔ ) were assayed in different genetic backgrounds and their longevity extension phenotype was confirmed in SDC medium . Overall , half of the mutations retested either under starvation/extreme CR or both in medium and under starvation/extreme CR was confirmed to be implicated in life span regulation , underscoring the effectiveness of our experimental approach but also the importance of using the water paradigm to filter out false positives . Interestingly , none of the long-lived mutations identified here has been identified by the high-throughput analysis performed previously [26] . Similarly , we did not identify any mutants of the TOR/Sch9 or of the other known pro-aging pathway , possibly because mutations in these pathways may affect growth rates and interfere with this growth-based method to determine life span . In our assay , the interaction between deletion mutants in the same environment might increase the noisiness of the data . In fact , the death of the short-lived mutants and consequently , the release of nutrients , might facilitate the regrowth of other mutants , which in turn might lead to the accumulation of metabolites ( e . g . ethanol or acetate ) detrimental for cell survival [7] , [18] . This death-regrowth dynamic is supported by the increase of CFUs detected on day 12 and 15 of our initial screen ( Figure 1A ) . A certain degree of noise indeed exists in our data . For example , the trm9Δ and ydr417cΔ mutants were confirmed as long-lived despite the observation that the reproducibility between the two biological replicates as estimated by EDGE analysis did not reach the threshold of q = 0 . 1 ( Table 1 ) . Some noisiness in the data may also explain why none of the genes implicated in yeast apoptosis and known to reduce chronological life span was identified in our screen . Although we do not have results relative to the pro-apoptotic NDE1 gene ( the corresponding mutant was excluded from our dataset , see Materials and Methods ) , other deletions associated with a down-regulation of yeast apoptosis , e . g . yca1Δ , aif1Δ , and ndi1Δ , were not present in the putative long-lived group generated by K-means clustering . This is not surprising given the modest life span extension caused by the deletion of these genes [67] , [68] . Among the long-lived mutants identified is acb1Δ . Acb1 is a ∼10kD acyl-CoA-esters binding protein highly conserved in eukaryotes [52] . In yeast , down-regulation of Acb1 activates a stress response , which includes several heat-shock proteins and the cytosolic catalase ( Ctt1 ) [69] . Although in the BY4741 acb1Δ mutant we did not observe resistance to H2O2 , the mutant was heat resistant , in agreement with the up-regulation of heat-shock genes [69] . The serine/threonine kinase CK2 is constitutively expressed in all eukaryotes so far investigated and hundreds of CK2 substrates are known . [70] . In yeast , CK2 is required for cell viability , it is primarily located within the nucleus and is implicated in the regulation of chromatin structure and global gene expression [55] . Since in mammalian cells both S6K and Akt/PKB are CK2 substrates [71] , [72] , it will be informative to test whether Sch9 , the closest yeast homologue of these kinases , interacts with CK2 directly or indirectly . Trm9 is one of several tRNA methylases present in yeast and is conserved in several other species [58] . Lack of Trm9-dependent methylation at U34 is thought to cause the incorporation of the wrong amino acids into proteins [58] , which may explain the growth defect observed in the trm9Δ mutants . The production of certain defective proteins might: a ) simulate a reduction of translation , which is known to extend the life span of C . elegans [73] and b ) promote chaperone synthesis , which may have an anti-aging role [74] . The deletion of the dubious YDR417C ORF , which overlaps with RPL12b , causes reduction of growth rate , life span extension under starvation/extreme CR and in regular medium , and heat-resistance . Analogous results were obtained when the deletion of RPL12b was analyzed . We have not tested directly whether YDR417C or RPL12b is responsible for the phenotypes observed . However , since a ) in high-throughput gene expression studies no expression corresponding to YDR417C has been detected [59] , [60] and b ) abolishing/reducing the expression of ribosomal proteins causes RLS and CLS extension in yeast and worms , respectively [13] , [75] , lack of Rpl12b may be the cause of the phenotypes we observed in the ydr417cΔ mutant . The deletion of the aromatic amino acid biosynthetic gene ARO7 leads to longevity extension and heat-shock resistance in BY4741 . This observation points to a general role of amino acid-signaling in life span regulation . In fact , several lines of evidence suggest that the reduction of the amino acid/protein component of the diet might be important for life span extension: 1 ) removing glutamate or asparagine from the medium promotes yeast CLS extension [26] , 2 ) amino acid-restriction prolongs yeast RLS [76] , 3 ) reduction of the protein content in the diet of Drosophila was shown to be the key factor to extend life span [77] , 4 ) in rodents , a simple reduction of either methionine or tryptophan from the diet promotes longevity [78] , [79] . It will be important to understand how different amino acids affect life span and whether conserved pathways function to regulate life span in response to amino acid restriction in different species . It is noteworthy to point out that our screen identified long-lived mutants that are more sensitive to acetic acid than wild type ( trm9Δ , ydr417cΔ , and aro7Δ ) . In this regard , they differ from the long-lived sch9Δ and ras2Δ mutants , which were shown to be resistant to acetic acid [18] ( Figure 4D ) , suggesting that resistance to high concentrations of this acid is not a requirement for longevity extension . Further experiments are needed to understand whether acetic acid plays a pro-aging role in a trm9Δ , ydr417cΔ , or aro7Δ context . Notably , the majority of the novel long-lived mutants did not show an increased percent of G1/G0-arrested cells during chronological aging . On the contrary , one of them , acb1Δ , showed a significantly higher budding rate up until day 7 . Together , these results indicate that , while a tighter G1/G0 arrest may improve chronological survival [18] , its role in yeast aging is not central . In summary , we have identified novel yeast pro-aging genes that point to cell functions previously not linked to life span regulation . Several of these genes are evolutionary conserved suggesting that they may also function to control longevity in other species .
Pools of the BY4741 ( MATa his3Δ1 , leu2Δ0 , met15Δ0 , ura3Δ0 ) haploid deletion collection were obtained as described previously [35] . All the other strains used for this study were generated in either the DBY746 ( MATα leu2-3 , 112 , his3Δ , trp1-289 , ura3-52 , GAL+ ) or the W303-1A ( MATa leu2-3 , 112 trp1-1 , can1-100 , ura3-1 ade2-1 , his3-11 , 15 ) genetic backgrounds by one-step gene replacement as described by Brachmann et al . [80] . The complementation tests were performed on the acb1Δ mutant after transformation with either a centromeric MoBY plasmid carrying the ACB1 ORF under its own promoter or a control vector [81] . Yeast cells were grown in synthetic complete medium ( SDC ) supplemented with a four-fold excess of tryptophan , leucine , uracil and histidine . Yeast chronological life span was measured as previously described [3] . Briefly , overnight SDC cultures were diluted ( 1∶200 ) into fresh SDC medium and incubated until day 3 when no residual cell growth is normally observed . Viability was measured by plating aging cells onto YPD plates and monitoring Colony Forming Units ( CFUs ) starting from day 3 , which was considered to be the initial survival ( 100% ) . For starvation/extreme calorie restriction , cells from 3 day-old SDC cultures were washed three times with sterile distilled water and resuspended in water . Every 4–7 days , cells from the water cultures were washed to remove nutrients released from dead cells . To establish significance between the survival of wild type and deletion mutants , the mean life span calculated by curve fitting ( Boltzmann sigmoidal ) and/or the area under the survival curves were used to perform: 1 ) one-way ANOVA followed by Tukey's multiple comparison test ( Figure 2A , Figure S4A ) and 2 ) paired t-test , two-tailed ( rest of the survival studies ) . The statistical software Prism ( GraphPad Software ) was used for the analysis . CK2 inhibitors were added to the aging cultures at day 2 and 5 and resistance to heat was tested at day 3 . The range of concentrations tested was 10–200 µM for TBBz and 5–15µM for TBBt . At different time points the budding index of approximately 500 cells was determined visually by using a phase contrast microscope . Prior to cell examination , cell clumps were removed by brief sonication . The data were analyzed by one-way ANOVA followed by Tukey's multiple comparison test . At day 3 , 9 , 11 , 15 , and 20 , samples from the aging pools were diluted in fresh medium to an initial density of 6 . 25×105/mL and incubated at 30°C until the cultures reached a density of 107/mL . Approximately 2×107 cells were spun down and cell pellets were stored at −20°C until further processing . Genomic DNA was extracted using the YeaStar Genomic DNA kit ( Zymo Research ) as previously described [35] . TAG hybridization and processing of microarrays were performed as described by Pierce et al . [35] . Barcode probe intensities were extracted and processed as described previously [35] . For each time point , the corresponding array was mean normalized and log2 ratios were calculated with respect to the day 3 time point to obtain mutant-specific aging profiles ( Table S2 ) . Approximately 500 mutants whose tag intensity was similar to the background at each time point ( e . g . mutants growing extremely slowly or mutants whose tags hybridize poorly ) were excluded from the analysis . Next , K-means clustering ( K = 10 ) was performed to identify strains with similar aging profiles . Profiles for each strain were averaged between replicates prior to K-means clustering . The root squared mean errors ( RSME ) between the two replicates were calculated and 413 mutants with high RSME ( 90th percentile ) were also excluded from the analysis ( Table S2 ) . Clusters were classified as short-lived and long-lived by manual inspection . A novel significance method was recently developed for identifying differentially expressed genes in longitudinal time course microarray studies [42] . We adapted this method to our aging dataset , which is analogous to a time course microarray as it involves repeated sampling and measuring from a pool of mutants . This analysis was performed using the EDGE analysis software package [43] to identify strains whose representation in the pool over the course of the experiment changed consistently in both replicate experiments . The time course differential expression analysis option ( ‘within class’ analysis ) was used . Using a q-value cutoff of 0 . 1 , we identified 438 strains that were significantly overrepresented or underrepresented in the pool over the course of the experiment . The q-value estimates the false-discovery rate when calling a gene significant [82] , [83] . The survival curves for each individual mutant reported on Table S3 were generated by multiplying the fold change in the microarray intensities ( Table S2 ) by the CFUs of the pool at each time point ( Table S1 ) . The short-lived mutants identified by K-means clustering were subjected to Gene ontology enrichment analysis . Gene ontology analysis was performed using the Gene Ontology ( GO ) Term Finder ( http://go . princeton . edu ) [84] , which uses a hypergeometric distribution to determine whether GO terms are enriched in a list of genes at a frequency greater than that expected by chance . Heat shock resistance was measured by spotting serial dilutions ( 10-fold dilutions , starting from a cell density of ∼108/mL ) of cells removed from day 1 or 3 SDC cultures onto YPD plates and incubating at either 55°C ( heat-shocked ) or 30°C ( control ) for 60–240 min . After the heat-shock , plates were transferred to 30°C and incubated for 2–3 days . For the oxidative stress resistance assay , cells were diluted to a cell density of 107/mL in K-phosphate buffer , pH6 . 0 , and treated with 100–200 mM of hydrogen peroxide for 30 minutes . Serially diluted ( 10-fold ) control and treated cells were spotted onto YPD plates and incubated at 30°C for 2–3 days . For acetic acid resistance , day 3–5 cultures ( 0 . 5 mL ) were treated with 300–500 mM acetic acid for 180 min . After the treatment , serially diluted cells were spotted onto YDP plates and incubated at 30°C for 2–3 days . The pH of the acetic acid-treated cultures was ∼3 and did not differ depending on the mutant , neither it changed depending on the acetic acid concentration used . All experiments were repeated 2–3 times with similar results . All supplementary data can also be downloaded from our webpage: http://chemogenomics . med . utoronto . ca/supplemental/lifespan/ While this article was being revised , Metecic M et al . published a paper describing a screen for long- and short-lived mutants similar to the one reported here [Matecic M , Smith DL , Pan X , Maqani N , Bekiranov S , et al . ( 2010 ) A microarray-based genetic screen for yeast chronological aging factors . PLoS Genet 6: e1000921 . doi:10 . 1371/journal . pgen . 1000921] . | Model organisms have been instrumental in uncovering genes that function to control life span and to identify the molecular pathways whose role in aging is conserved between the evolutionarily distant unicellular yeast and mice . Because yeast are particularly amenable to genetics and genomics studies , they have been used widely as model system for aging research . Here we have exploited a powerful genomic tool , the yeast deletion collection , to screen a pool of non-essential deletion mutants ( ∼4 , 800 ) to identify novel genes involved in the regulation of yeast chronological life span . Our results show that normal life span depends on functional mitochondria and on the cell's ability to degrade cellular components and proteins by autophagy . Our data indicate that a cell signaling protein , CK2 , and diverse cellular processes such as fatty acid metabolism , amino acid biosynthesis , and tRNA modification modulate yeast chronological aging . The high level of conservation of the novel life span regulatory genes uncovered in this study suggests that their role in longevity regulation might be conserved in higher eukaryotes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
]
| [
"genetics",
"and",
"genomics"
]
| 2010 | Genome-Wide Screen in Saccharomyces cerevisiae Identifies Vacuolar Protein Sorting, Autophagy, Biosynthetic, and tRNA Methylation Genes Involved in Life Span Regulation |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.